Brexit Damage Limitation: Tariff-Jumping FDI and the Irish Agri-Food Sector

Following the United Kingdom's withdrawal from the European Union at the end of January 2020, an eleven-month transition period came into effect during which existing rules on trade, travel, and business remained in operation. An agreement on future relations is to be negotiated over this period. The complexities involved in international trade negotiations notwithstanding, the UK government has announced that it will not seek an extension of the negotiating period.

In terms of the types of possible Brexits, the spectrum running from ‘hard’ to ‘soft’ refers to the height of the barriers that would be applied to UK-EU trade. The hardest of ‘hard Brexits’ would have arisen had the UK crashed out of the European Union without a withdrawal agreement, in which case trade would have defaulted to World Trade Organisation (WTO) rules, as explained below. The eventual signing of a withdrawal agreement, after much delay, allowed the transition period to come into effect. The term ‘hard Brexit’ was also used generically to refer to the UK choosing to leave both the Single Market and the EU Customs Union. The UK's rejection of free movement of labour rules out membership of the Single Market, while its stated desire to negotiate free trade deals of its own with non-EU countries is inconsistent with membership of a customs union (which is defined by its common external tariff): hence a hard rather than a soft Brexit seems to be the guaranteed outcome at this juncture.

Unless a substantive free trade deal can be negotiated between the UK and the EU then trade between them will default to WTO rules. A key WTO rule is the ‘most favoured nation’ clause: in the absence of a substantive free trade deal, WTO members are required to charge each other the same tariffs they apply to other WTO member states. As trade in agriculture and food is subject to particularly high WTO tariff rates

See Figure 3 of Lawless and Morgenroth, 2019.

Though the policy implications are of broader general applicability, as will become clear later, the main focus of the present paper is on beef and dairy. Table 1 shows Ireland's export destinations for these sectors. The UK takes around half of Irish beef exports. For the narrowly-defined dairy sector – which comprises milk, milk powders, whey, butter and cheese – the UK accounts for around 40 per cent of Irish exports. When the sector is defined more broadly to include infant formula and caseinates the proportion is closer to 20 per cent.

Internationally traded products are classified by HS (Harmonized System) codes for customs purposes. The narrow definition uses 6-digit HS codes. The broader definition includes certain 8-digit product lines representing infant formula and caseinates (or ‘baby and bodybuilder’ products, as they are referred to colloquially) which are included under ‘Miscellaneous Edible Products & Preparations’ in the 6-digit system. The broader definition is employed by the Irish agri-food industry and the Department of Agriculture, Food and the Marine.

The range of possible consequences of Brexit for Ireland extend well beyond the agriculture and agri-business sectors and are surveyed in ESRI (2015) and Copenhagen Economics (2018). Reflecting the high degree of uncertainty as to the future trading regime, these and the numerous studies conducted at the broader European level work through a range of scenarios based on differing combinations of tariff and non-tariff barriers. The present paper adopts, purely as a thought experiment, the most extreme of scenarios, which brings the issues of particular interest to us most clearly to the fore. This ‘doomsday scenario’ posits a complete cessation of UK-EU trade in beef and dairy products. Irish products lose access to the UK market but a vacuum is created in the markets of the EU27 by the prohibition on imports from the UK. How might this vacuum be filled, and specifically what share might be capturable by Ireland?

The answer will depend, on the demand side, on the degree of substitutability between the products of the individual EU27 member states and those of the UK, and, on the supply side, on available capacity and on the match between EU member state resources and the resource requirements of the particular products under consideration.

This assumes no change in the level of imports accepted from outside the EU27.

Revealed comparative advantage (RCA) is a measure of the relative advantage or disadvantage a particular country has in a particular sector, as evidenced by existing trade flows.

The revealed comparative advantage index compares the share of commodity i in country j's exports to this commodity's share in the exports of some aggregate group of competitor countries. An RCA index greater than 1 indicates that country j has a comparative advantage in commodity i relative to the aggregate group of countries.

The answers provided by our RCA analysis are found to offer little comfort to Irish farmers. This is where the second key element of the paper comes in. It has been noted from an earlier era of structural transformation in Ireland that RCA served as a poor predictor of future developments because of the substantial sectoral foreign direct investment (FDI) inflows that occurred as part of the process (Barry and Hannan, 2001). The reason for this is clear: RCA analysis is based on a country's existing factor endowments, and factor endowments change as a result of FDI inflows.

The FDI flows of relevance in the present case are of the ‘tariff jumping’ variety.

Tariff-jumping foreign direct investment (FDI) arises when a foreign firm locates production within the destination market in order to avoid a trade barrier. We use the term in this broad sense to include FDI undertaken to avoid non-tariff barriers (NTBs) also.

The issues we address are of considerable importance. The UK investments of Irish agri-food firms, though they make business sense, provide no insurance to Irish farmers. Irish agriculture will have to find new markets: given the scale of the adjustments required, this may take a considerable number of years. A second consideration is political. The positive effects of the expansion in international financial services will accrue largely to Dublin. The adverse shock will be much more substantial – agri-business industries and the agricultural jobs they support dwarf international financial services employment by a factor of five or six – and will impact primarily on rural regions. Such geographically-divergent outcomes have been a factor in the anti-globalisation backlash seen across much of the developed world over recent years.

The paper is structured as follows. The next section provides a review of the literature on tariff jumping FDI, comprising both the academic literature and more recent newspaper coverage of Brexit-related FDI flows. Sections 3 and 4 present the RCA analyses of the beef and dairy sectors respectively, while Section 5 returns to the issue of tariff jumping FDI to discuss the policy implications of the analysis. The paper closes with some brief concluding comments.

From the foundation of the Irish state until the policy shift towards export orientation in the mid-1950s most inward FDI was ‘tariff jumping’ in nature (Barry et al., 2016; Sweeney, 1973). Such FDI flows are frequently two-way in nature, the significance of which will become clearer later. The erection of a customs frontier on the island of Ireland in 1923, for example, induced a number of British tobacco companies to establish production facilities in Dublin, while Dundalk tobacco firm P. J. Carroll & Co. jumped the trade barrier in the opposite direction, establishing a factory in Liverpool in 1923 to protect its British sales (Irish Times, 19 Aug. 1930).

The domestic market is of little significance for most of the well-known foreign MNCs that Ireland hosts today. For the likes of Google, Apple, Intel and Pfizer, Ireland serves as an ‘export platform’ from which to produce for the international market. In thinking about the implications for Brexit however it is important to recognise that the distinction between ‘tariff jumping’ and ‘export platform’ FDI is not as clear-cut as might be supposed. Though the UK in the 1920s imposed tariffs on imported cars, the fact that preferential tariff rates applied to production in Commonwealth countries caused US car manufacturers to cross the border from Detroit and establish export-platform factories in Southern Ontario from which to ‘tariff jump’ into the protected UK market (Kindleberger, 1990, 141). A similar blend of FDI types was apparent in pre-EEC Ireland when one of the attractions that the Industrial Development Authority (IDA) advertised to foreign investors was the preferential access to UK and Commonwealth markets that an Irish location offered (Barry and O’Mahony, 2017).

This same blend of motivations is apparent in the case of the London-based financial services firms that have been establishing in Ireland since the 2016 UK referendum to allow them to retain access to the European Single Market.

Note that this ‘barrier jumping’ conforms to our broad definition of tariff jumping FDI.

Ireland's largest agri-food businesses already had substantial operations in the UK before Brexit appeared on the radar, and this geographic diversification affords them a degree of protection against the possibility of a hard Brexit. In the case of Kerry Group, for example, more than 90 percent of its UK sales are already manufactured at its more than one dozen UK plants (Irish Times, 11 Aug 2017). In a mirror image of developments in financial services, other Irish dairy and beef firms have been working to ensure that they have adequate facilities within the UK to service that market in the event of trade barriers being erected.

Lakeland and Glanbia are among the largest Irish dairy firms, while many smaller co-operatives use Ornua, the former Irish Dairy Board, to market and sell their produce. Lakeland's purchase of Northern Ireland company Fane Valley's milk business in May 2016 was reported to provide the company with a “massive safety net” against Brexit (Irish Independent, 4 May 2017). Lakeland then went on to acquire another Northern Ireland firm LacPatrick Dairies which also had processing facilities on both sides of the border (Irish Times, 6 Mar 2019). The company's expanded cross-border presence, it was suggested, would “ensure that it avoids the worst of any Brexit fallout” (Irish Times, 7 Dec 2018).

Glanbia, though it has only a low exposure to the UK market, has contract manufacturing arrangements in place in that jurisdiction in case imports from the EU are disrupted (Irish Times, 24 Jan 2019). The company, which is described as “Europe's biggest manufacturer of mozzarella”, has begun construction of a mozzarella-processing plant in Ireland to ensure EU access if exports from its current facilities in Wales and Northern Ireland are disrupted (Irish Times, 7 Dec 2018). Ornua acquired a British cheese plant in 2017 and has indicated that it has contingency plans to use British cheddar to supply the UK market in the event of trade disruption (Irish Times, 25 Jan 2017; Irish Independent, 25 Feb 2019).

Similar developments are apparent in the beef processing sector. Two of the largest Irish beef processors are Dawn Meats and ABP. Dawn Meats agreed a strategic partnership with Northern Ireland firm Dunbia in 2017. The latter's two large Northern Ireland plants along with its seven in Britain were described in newspaper reports as providing the Irish firm with a ‘Brexit buffer’ (Irish Times, 24 May 2017). Its rival, ABP, already employed 6,000 in the UK by 2017 and has been investing further since then. Its UK arm has been described as being firmly positioned “as a supplier of British beef and lamb to British retailers” (Irish Independent, 5 Jun 2018).

Why, in this case, do British agri-food businesses appear to have been slower to insure themselves against the prospect of a loss of EU market access? A recent report by Breinlich et al. (2019) finds that though the Brexit referendum vote has led to a 12 percent increase in new investments by UK firms across the rest of the EU, these investments have been confined to services sectors. The authors suggest by way of explanation that the UK government has been perceived to have prioritised the interests of manufacturers in its negotiations with the EU. Membership of the Single Market, which is of particular importance to services firms, was ruled out from the beginning: the focus was instead on minimising customs frictions. There can be no guarantee that this will continue to be the case, particularly in the case of agri-food trade given the possibility of regulatory divergence between the UK and the EU.

The logic of the processes outlined above – Irish agri-food firms ‘tariff jumping’ into the UK and London-based financial services firms ‘non-tariff barrier jumping’ into Ireland – suggests a way by which Ireland might be able to avoid some of the damage that a hard Brexit would entail. The type of FDI focused upon here has been largely overlooked in the studies conducted for the Irish government by the Economic and Social Research Institute and consultancy firm Copenhagen Economics. In discussing the FDI implications of Brexit, ESRI (2015) concentrates on the possibility of ‘FDI diversion’ whereby inflows that would otherwise have gone to the UK are diverted to Ireland. This is an entirely different focus from that of the present paper. The ESRI study, furthermore, is based on analysis ‘of patterns of the location choice of new FDI projects in Europe over the past ten years’, a methodology that cannot be used to explore the issue of interest to us. Copenhagen Economics (2018) makes only a passing reference to tariff jumping FDI, noting merely that “increasing costs of final goods trade to the EU can [make] it more attractive for the UK to use Ireland as an export hub for EU destinations”.

Consider, purely as a thought experiment, that Brexit brings all UK-EU trade in beef and dairy products to a halt. Table 2 shows the trading position in beef and dairy (broadly defined) as of 2016, where the figures have been scaled down for ease of exposition. Ireland exports more beef to the UK than the UK exports to the rest of the EU (including Ireland). Even were Ireland to capture the entire EU market share vacated by the UK, Irish beef would remain in massive oversupply.

There may be some slight comfort to be had in the fact that Ireland and the UK export to the same third-country markets, that they export the same main product category (though there are quality differences in the beef exported to the various national markets), and that Ireland has a strong revealed comparative advantage in this main export product category.

As Table 3 shows, Ireland and the UK are each other's most important destinations for beef exports, followed in descending order by the Netherlands, France, Italy and Germany. Together these markets account for 85 percent of Irish and 83 percent of total UK beef exports.

It is of interest to note however that neither Ireland nor the UK are the main export suppliers of beef in any of these continental markets. Germany, for example, is the main export supplier to the Netherlands, though its key product categories are different from those of the UK and Ireland.

As measured by international trade code categories, Ireland and the UK export similar beef products. In fact, as seen in Figure 1, the vast bulk of Irish and UK beef exports are in the same product category, i.e. HS code 020130. (The definitions of the various categories are provided in Appendix 1).

Figure 1

The export price data for this category suggest however that the two countries export different cuts of meat to the various national markets. Table 4 for example shows that while Ireland receives better prices than the UK on average across the EU28, it receives lower prices in France. This is presumably a reflection of the ‘carcass balance’ complications referred to earlier.

The flows in premium cuts can also vary over time. Irish exports to the UK in this particular trade category achieved higher prices than UK exports to Ireland in 2012 (not shown), but the position was reversed in each of the following four years.

Table 5 shows that Ireland has a stronger revealed comparative advantage in this export category than its principal competitors, the UK and the Netherlands.

Though the Netherlands has a lower RCA than Ireland in its EU trade in this category, it exports marginally more than Ireland to the EU28.

We now conduct a similar analysis for the dairy sector, where we largely concentrate on the broad (8-digit) definition employed by the Irish agri-food industry and the Department of Agriculture, Food and the Marine. The infant formula and caseinates included in the broader definition of the sector have experienced substantial growth in recent years and – as is apparent from Table 1 – are exported largely beyond the EU. Their inclusion yields a more optimistic assessment as to the post-Brexit prospects for the sector. There is another reason too why, at first glance, the situation in dairy might appear more benign than that facing the beef industry. Table 6 shows that the vacuum created by the exclusion of UK dairy exports from the EU market would exceed the loss in Irish dairy exports to the UK. (This may be seen more clearly in Table 2 above).

It is inconceivable of course that the Irish dairy industry could capture the entirety of this newly available market. The situation is even more challenging however, firstly because – while Ireland and the UK largely export to the same group of EU countries (not shown) – they export very different dairy products (unlike in the case of beef), and secondly (and again unlike in the case of beef) Ireland does not have a comparative advantage in segments that account for the bulk of the UK's dairy exports.

Recall that in the case of beef, as seen in Figure 1, Irish and UK intra-EU exports are strongly concentrated in the same 6-digit product category, and both Ireland and the UK have a comparative advantage in this category. While there is some overlap in the case of dairy, Ireland does not have a comparative advantage in two of the top three UK 6-digit dairy export categories (Table 7), indicating that Ireland would face strong competition in attempting to fill the vacuum in these segments. Capacity matters as well as comparative advantage however, so competition would come particularly from countries with sizeable dairy herds as well as a high RCA.

While several small EU27 states may have a strong revealed comparative advantage in some of these segments, their dairy herds would not be large enough to capture much of the market vacated by the UK. Accordingly, Table 7 focuses only on countries with significant dairy herds.

Eurostat (2020) Number of Dairy Cows – annual data, 2019: https://ec.europa.eu/eurostat/databrowser/view/tag00014/default/table?lang=en

The degree of dissimilarity between the exports of the UK and Ireland is apparent also when we move to the 8-digit product level. Though Ireland has a comparative advantage in a number of the UK's main export sub-sectors at this level, these add up to only a fraction of total UK intra-EU dairy exports (Table 8).

The revealed comparative advantage (RCA) analysis does not offer great hope for Ireland's prospects of filling the dairy vacuum created by a presumptive UK exclusion from the EU market. Adjusting to Brexit will of course require Irish agri-food firms to diversify away from the UK market. Opening up new foreign markets is no easy task however. The issue explored here is whether there are grounds for believing that the adjustment can be eased, or the process supplemented, by seeking to attract particular types of UK-based firms to produce in Ireland for the post-Brexit EU market, i.e. by seeking to stimulate the mix of tariff-jumping and export-platform FDI discussed earlier in the paper.

The more sophisticated segments of the dairy sector differ from beef in being characterised by ‘marketing intangibles’, of which brand names are the most visible form; others include strong customer relationships and established distribution channels (OECD, 2010). Sectors characterised by such proprietary knowledge, which also includes R&D-related ‘trade intangibles’, tend to be dominated by MNC incumbents which are difficult to displace (Markusen, 1998).

Thus historian Mary Daly (1984), in discussing the characteristics of early FDI inflows to Ireland, recognised that these were more likely in ‘large-scale consumer industries dominated by a small number of large firms who (fight) for market shares with heavy advertising expenditures and the creation of strong brand images’.

If customer bases have a strong attachment to incumbent dairy firms then it makes sense for Ireland to leverage its considerable experience in FDI attraction to seek to induce these UK-based firms to establish operations in Ireland. The raw materials they use are precisely those that Ireland would have in excess supply in the doomsday scenario under consideration.

There are numerous historical episodes which can be drawn upon to illustrate the use of Irish dairy raw materials by locally-based ‘intangible capital’-endowed foreign firms. Fox and Breathnach's (2014) history of the mechanised creamery system in the early 20th century reveals a strong initial presence of British firms before their displacement by the burgeoning co-operatives. In most cases, as they note, these firms had been sourcing butter from Ireland in the pre-creamery era and established their own creameries to copper-fasten supply to their well-established distribution channels.

Several decades later, Rowntree and Cadbury, which had been producing chocolate confectionery in Dublin since the 1920s and 1930s respectively, commenced production of the key ingredient – chocolate crumb – at Mallow and Rathmore in the 1940s when milk and sugar remained subject to post-war rationing in Britain. Chocolate crumb subsequently became a major export industry, with Ireland reported in 1962 to be the largest exporter of chocolate products by tonnage in the world (Irish Times, May 25, 1960).

Two further historical parallels bear mention. Ballyclough's emergence as the largest co-operative creamery in the country in the early 1970s was credited to its “diversification into the best-paying product lines” (Irish Farmers Journal, 22 May 1971). This was based on the manufacturing and sales relationships it had developed with nearby foreign MNCs – Rowntree in the case of chocolate crumb, Cow & Gate in the case of baby food, the former Manx Creameries in the case of cheese, and US company Borden in the case of high grade milk powder (Irish Farmers Journal, 22 May 1971). The final example relates to the present-day infant formula sector, which currently accounts for some 35 percent of broadly-defined dairy exports. Three of the four key players, along with Kerry Group, are foreign MNCs. Danone now owns Cow & Gate and operates baby-nutrition sites at Macroom and Wexford. Wyeth Nutrition supplies infant and maternal nutrition products from its factory in Askeaton, Co. Limerick, and has since 2012 been part of Nestlé, while Abbott Nutrition processes milk, whey and lactose at Cootehill, Co Cavan, to produce toddler, infant and follow-on formula for distribution across the globe.

See investor profiles at www.Enterprise-Ireland.com.

These historical episodes demonstrate the feasibility of attracting UK food processing firms to Ireland to service foreign markets. For smaller UK producers, we know from an earlier era that the familiarity of the Irish environment (in terms of language and legal system for example) can represent a significant advantage (Barry, 2019). More generally, while land bridge issues and the seasonality of milk production will militate against an Irish base, a 2009 change in the UK tax system – which abolished dividend taxes on foreign repatriation from many low-tax countries, including Ireland – will work in the opposite direction (Liu, 2018). The challenge, furthermore, plays to the strengths of Ireland's development agencies. The modus operandi of the IDA was described by former managing director Padraic White as first identifying the sectors that represented a good fit for the country and then targeting particular companies and seeking to persuade them to establish operations in Ireland (White, 2000, 272).

Brexit is likely to be extremely damaging to Ireland's beef and dairy sectors. While Irish beef may be well-positioned to capture a strong share of the EU markets to be vacated by UK exports, quantitatively this will be nowhere near sufficient to prevent a sharp fall in the aggregate demand for Irish beef.

The analysis is more complicated in the case of dairy. The United Kingdom exports substantially more dairy products to the EU than Ireland exports to the UK. The products exported by the two states differ however, and Ireland does not display a strong comparative advantage in many of the products exported by the UK. This suggests that it would not be in a strong position to capture the EU market share vacated by UK-based firms. To the extent that Irish dairy raw materials are substitutable for UK raw material inputs, damage from Brexit could be alleviated by seeking to attract UK and UK-based downstream dairy firms to establish export platform operations in Ireland.

The implications of the present analysis are not confined to the dairy sector. Customer loyalty will also play a significant role in other advertising-intensive and R&D-intensive sectors (Appendix 3), and many UK firms, faced with the prospect of diminished EU-market access, may be amenable to the attractions of an Irish location.