The Costs of Non-Unification by mattcarthy5

The Costs of Non-Unification Brexit and the Unification of Ireland

KLC Consulting Vancover, BC. Canada

JULY 2018

Impressum

Bibliographical information: Autor: Kurt HĂźbner, KCL, and Renger van Nieuwkoop, Modelworks Title: The Costs of Non-Unification. Brexit and Irish Unification Place: Vancouver and Thun Year: 2018

Responsibility for the information and views set out in this report lies entirely with the authors.

Kurt HĂźbner www.klconsult.ca KLC Consulting Vancouver BC Canada

Renger van Nieuwkoop Modelworks Goldiwilstrasse 16 F 3600 Thun Switzerland info@modelworks.ch www.modelworks.ch

Contents Comments by Marcus Noland (Peterson Institute for International Economics)..………….………….i Research Team.......................................................................................................................... 1 Brief Summary .......................................................................................................................... 2 Executive Summary................................................................................................................... 5 1

Background................................................................................................................. 14

Scenario “Brexit” ........................................................................................................ 21 2.1 2.1.1 2.1.2 2.1.3 2.1.4 2.1.5 2.1.6

Introduction ............................................................................................................ 21 Scenario “Brexit_TR” .............................................................................................. 23 Scenario “Brexit_IM”: Less Immigration ................................................................ 25 Scenario “Brexit_FD”: Reduction in Foreign Direct Investment ............................. 26 Scenario “Brexit_XR”: Devaluation of the GBP. ..................................................... 27 Scenario “Brexit” .................................................................................................... 28 Scenario “Brexit_SM” ............................................................................................. 29

2.2 2.2.1 2.2.2

Imports and Exports ............................................................................................... 30 Overall Effects on Imports and Exports .................................................................. 30 Regional Imports and Exports ................................................................................ 31

Hard Brexit versus SM-Brexit and Unification ............................................................. 32

Conclusions................................................................................................................. 35

References .............................................................................................................................. 37 5

Appendix: Sectoral Effects .......................................................................................... 46 5.1

Sectoral Effects on Trade........................................................................................ 46

5.2

Effects on Domestic Production ............................................................................. 46

Appendix: The Recalculated Unification Scenario ....................................................... 47 6.1

Assumptions for the Unification Scenarios ............................................................ 47

6.2

Overall effects of unification .................................................................................. 48

6.3

Harmonization of taxes and tariffs ......................................................................... 48

6.4

Optimized government .......................................................................................... 50

6.5

Reduction of NTBs .................................................................................................. 51

6.6

Productivity ............................................................................................................ 52

6.7

Euro Conversion ..................................................................................................... 52

Appendix: Modeling and Data Selection ..................................................................... 55

7.1

The choice of CGE as modeling tool ....................................................................... 55

7.2

The NIROI Model .................................................................................................... 57

7.3

Data and Base Year ................................................................................................. 61

List of Figures Figure 1-1: Comparing the effects of Unification, Brexit, and Brexit_SM: Overall GDP and changes in GDP .............................................................................................................................................................. 12 Figure 1-2: Comparing the effects of Unification, Brexit, and Brexit SM: Overall GDP and changes in GDP per Capita ............................................................................................................................................. 13 Figure 2-1: Changes in GDP for Northern Ireland and the Republic of Ireland due to tariff changes after Hard Brexit........................................................................................................................................... 25 Figure 2-2: Changes in GDP for Northern Ireland and the Republic of Ireland due to immigration changes after Hard Brexit ............................................................................................................................ 26 Figure 2-3: Changes for Northern Ireland due to FDI changes after Hard Brexit......................................... 27 Figure 2-4: Changes in GDP for Northern Ireland and the Republic of Ireland due to devaluation of the pound after Hard Brexit ......................................................................................................................... 28 Figure 2-5: Overall changes for Northern Ireland and the Republic of Ireland after Hard Brexit................ 29 Figure 2-6: Overall changes for Northern Ireland and the Republic of Ireland when Northern Ireland remains in the SM ........................................................................................................................................ 30 Figure 2-7: Effects on exports and imports in Northern Ireland .................................................................. 31 Figure 2-8: Changes in regional exports and imports .................................................................................. 31 Figure 3-1: Impact of the Unification on GDP for Northern Ireland and the Republic of Ireland ................ 33 Figure 3-2: Overall effects on GDP of Brexit and Unification ....................................................................... 34 Figure 5-1: Changes in exports (left) and imports (right) ............................................................................. 46 Figure 5-2: Effects of Brexit on domestic production .................................................................................. 46 Figure 6-1: Impact of the Unification on GDP for Northern Ireland and the Republic of Ireland ................ 48 Figure 6-2: Impact of the harmonization of taxes (including tariffs) for Northern Ireland after the unification: New versus Old Simulation ....................................................................................................... 48 Figure 6-3: Impact of the reduced NI-government size after the unification .............................................. 50 Figure 6-4: Impact of the NTB changes for Northern Ireland after the unification ..................................... 51 Figure 6-5: Impact of the productivity changes for Northern Ireland after the unification ........................ 52

Figure 6-6: Impact of the exchange rate changes for Northern Ireland after the unification ..................... 54 Figure 7-1: Production structure (source: Lofgren & Robinson, 2002, p. 9) ................................................ 58 Figure 7-2: Flows of marketed commodities (source: Lofgren & Robinson, 2002, p. 12)............................ 59

List of Tables Table 1-1: Results from actual studies on the impact of Brexit on GDP ...................................................... 16 Table 2-1: Brexit scenarios and assumptions ............................................................................................... 22 Table 2-2: Tariffs, export taxes, and NTBs after the Hard Brexit ................................................................. 24 Table 6-1: Assumptions of the Unification scenarios................................................................................... 47

Comments on “The Costs of Non-Unification: Brexit and the Unification of Ireland” Marcus Noland Executive Vice President and Director of Studies Peterson Institute for International Economics Washington, DC. USA July 2018

The Costs of Non-Unification: Brexit and the Unification of Ireland presents an overly modest or conservative case for the economic advisability of Irish unification in the face of Brexit. When the full spectrum of considerations are taken into account, the case for unification may be even more compelling than presented in this analysis. Ever since UK government began Brexit negotiations in March 2017, following the June 2016 referendum to exit the European Union, negotiators have been wrestling with “the Irish issue.” The membership of the Republic of Ireland and the United Kingdom in the supranational European Union had facilitated the Good Friday Agreement, which ended decades of political violence in Northern Ireland. Specifically, common EU membership had supported the removal of the physical border separating the two parts of the island, enabling economic integration marked by high levels of free movement of goods and people across the border. Brexit puts these achievements at risk, leaving Northern Ireland at a critical juncture. (A majority of Northern Ireland residents who participated in the June 2016 UK referendum had voted against exiting the European Union.) A “hard Brexit” path, maximizing British sovereignty as envisioned by some referendum supporters, would involve withdrawal from the Single Market and the Customs Union, necessitating the reestablishment of the border dividing the two parts of Ireland and the reintroduction of restrictions on the movement of goods and people, not only between the two parts of the island but also between Northern Ireland and the rest of the European Union. A second path, unification, examined in an earlier report by KLC Consulting, Modeling Irish Unification, would deepen integration with the Republic of Ireland (and maintain participation in the European Union) but now at the additional cost of barriers to trade with the United Kingdom implied by that country’s withdrawal from the European Union. A third path, “soft Brexit” or Brexit-SM is an intermediate solution envisioned by some in the European Union, which involves giving Northern Ireland special status, permitting it to remain in the Single Market and Customs Union upon British withdrawal. EU negotiators indicated in November 2017 that the Good Friday Agreement necessitated an “all island” solution, in which Northern Ireland stayed within the Customs Union. The Costs of Non-Unification: Brexit and the Unification of Ireland critically examines the costs and benefits of these three alternative paths. Business as usual is not an option. In some sense the

study’s results are unsurprising: As a small, very open economy, Northern Ireland is highly susceptible to disruptions in its trade relations. Brexit, in whatever form, will damage the economy. But the potential gains from unification with the Republic of Ireland remain large enough to generate an improvement over the no longer viable status quo, even in the face of the dislocation associated with Brexit. Basic Modeling Challenge Analysts attempting to model the impact of Brexit on Northern Ireland confront two challenges. The first is that as a subnational jurisdiction, much economic data necessary to conduct the analysis may not be collected for Northern Ireland as an independent reporting entity. Working off their prior study, the authors have used a variety of techniques to generate estimates for Northern Ireland when the relevant data are not reported. In some cases, the needed data can be backed out of the UK accounts fairly easily. In other cases, the authors make informed adjustments to the UK data to construct admittedly more speculative estimates of the Northern Ireland figures. These approaches appear reasonable. The current study has been aided by the publication of new data, in particular, new input-output tables for Northern Ireland and the Republic of Ireland. The deeper issue is that the status quo is no longer viable, and hence it is no longer clear what the “baseline” ought to be. The authors have adopted the approach of analyzing their simulation results as deviations from the unification scenario in the earlier report, Modeling Irish Unification (but adjusted for the new data and the recalibration of one parameter). In this way, the Brexit scenarios can be interpreted as representing the opportunity costs of forgoing the unification option. Modeling Approach The authors use a 25-sector, 3-region dynamically recursive computable general equilibrium (CGE) model. It model is calibrated for 2013 and run over the period 2014-2025, with unification occurring in 2018 and Brexit coming into effect in 2021. The authors’ modeling of Brexit on Northern Ireland is sensible: Northern Ireland is treated as a “small country” relative to the Republic of Ireland; both parts of the island are treated as “small countries” relative to the rest of the world. The United Kingdom is not broken out separately in the modeling; one implication is that to the extent United Kingdom is negatively affected by Brexit, which is the consensus of the existing modeling work, that feedback on Northern Ireland and the Republic of Ireland (such as a slowdown in demand for Northern Ireland or Republic of Ireland sales to the United Kingdom excluding Northern Ireland) will not be captured. The implementation is done using a Bayesian approach used in previous CGE work, where the researchers need to construct social accounting matrices in cases where underlying data may be missing or subject to significant measurement error, a common experience in many countries. Other details of the modeling approach and model parameterization are quite conventional and do not raise any red

flags that unusual assumptions or approaches are being employed to generate particular results or outcomes. Brexit itself is modeled as four separate shocks:    

the imposition of trade barriers between the United Kingdom and the European Union, restricted immigration into the United Kingdom, resulting in a decline of the workforce of 0.85 percent, decline in foreign direct investment (FDI) into the United Kingdom, treated as a 1 percent decline in total factor productivity, and a 10 percent devaluation of the British pound.

The “hard Brexit” scenario is the imposition of all four shocks; in the soft “Brexit-SM” scenario, Northern Ireland experiences labor market, productivity, and exchange rate effects but avoids the imposition of new trade barriers with the European Union. Relative to the previous modeling work on Irish unification, the exchange rate shock implies that Northern Ireland would experience a real appreciation of its exchange rate with unification and entry into the Eurozone. The shock could depress output at least for a transitional period and hence make unification appear less attractive than in the previous modeling. Interpretation of the Results Since maintaining the status quo is not a viable option, as an alternative to constructing a baseline, the authors present the simulation results as deviations from unification. The appendix chapter 6 in this report presents a comparison of the unification modeling results obtained using the present and previous models. The updated data and recalibration of a public expenditure parameter (which has the effect of making public expenditure—basically the National Health Service—more efficient with Northern Ireland as part of the United Kingdom than it had been treated under the earlier model) have the effect of making unification marginally less attractive than previously calculated. As shown in Figures 1-1 and 1-2 in the report, Brexit in both hard and soft forms has a negative impact on GDP. Staying within the Single Market and Customs Union mitigates the impact. The positive impact of unification is sufficient to offset the impact of Brexit in either case. GGE models can be seductive in the amount of highly precise results that they generate. It is worthwhile to keep in mind that the models are not reality; they represent a highly rigorous abstraction of reality, embodying many assumptions. When interpreting the results of the models, it is best to think of them as pointing to or reminding us of the relevant channels through which policy may shape outcomes and giving us some guidance or insight into the impact of those policies, rather than fixating on the final decimal point of some simulation outcome. This

iii

caution is particularly apt in the case at hand, since some of the underlying data have been estimated or constructed and may not be precise. That said, the modeling suggests that a hard Brexit would cumulatively reduce Northern Ireland income by approximately €10 billion over 2021-2025, or about €5,000 per capita. The soft Brexit-SM option would generate smaller losses on the order of €3.8 billion or about €2,000 per capita. These are not small shocks. Yet there is reason to believe that the model results may be an underestimate of the actual impact. The modeling in The Costs of Non-Unification: Brexit and the Unification of Ireland is state-of-the-art. But this class of models does not do a very good job of capturing the economics of cross-border supply chains. In Northern Ireland these have grown significantly over the past two decades and are quite extensive in food processing, for example. The actual disruption to the Northern Ireland and Republic of Ireland economies is likely greater in the hard Brexit scenario than the model suggests. A second issue is transfers. The model makes certain assumptions on the magnitude of public transfers in the Brexit and unification scenarios. Northern Ireland agriculture receives significant subsidies under the European Union’s Common Agricultural Policy (CAP), which may not be fully captured by the assumption that transfers are spread uniformly across the UK economy (i.e., Northern Ireland agriculture is a disproportionate beneficiary of EU transfers). Withdrawal of CAP subsidies to Northern Ireland agriculture would probably not be fully offset by new fiscal transfers from London in any Brexit scenario. Hence the fall in Northern Ireland incomes could be greater than the modeling suggests. Finally, as previously observed, the model does not capture the likely adverse impact of Brexit on the British economy and associated spillovers to the economies of Northern Ireland and the Republic of Ireland through this channel. The upshot is that the results presented in this report may represent an overly modest or conservative case for the economic advisability of Irish unification in the face of Brexit.

Research Team Kurt Hübner Dr. Kurt Hübner received his Ph.D. in Economics and Political Science from the Free University Berlin, Germany. He is a professor at the Political Science Department at the University of British Columbia and holds the Jean Monnet Chair for European Integration and Global Political Economy. He has published 12 books and numerous articles in journals. His most recent books are ‘Europe, Canada, and the Comprehensive Economic and Trade Agreement’ (Routledge 2011) and ‘Global Currency Competition and Cooperation’ (publication date: Routledge 201 2019). 9). Hübner’s expertise is in the area of European integration in the context of the global political economy as well as in the political economy of Germany. His main focus is on the Euro, and the role of the Euro in global currency relations as well as th thee economic mode of governance. A further area of expertise is the relation between international competitiveness, innovation, and sustainability, and where he headed several projects in the past. His most recent project in this area deals with ‘National Pa Pathways thways to Low Carbon Emission Economies’ (Routledge 2018). Over the past few years, he also contributed to the analysis and assessment of CETA and TTIP. He has also directed projects for Vancouver Vancouver-based based KLC – a consulting company that focuses on European aand nd North American economic and political relations.

Renger van Nieuwkoop Dr. Renger van Nieuwkoop received his Ph.D. in Sciences from the ETH Zurich, Switzerland. He is a senior researcher and lecturer at the Center of Economic Research of the ETH Zuric Zurich h and founder of Modelworks, a Swiss Swissbased consultancy firm specialized in computable general equilibrium modeling and statistical analysis. He has been teaching CGE modeling all over the world for over 15 years and worked as a consultant for international organizations and institutes (e.g., the National Bank of Poland, the FAO, the Slovenian Ministry of Finance, and the Landcare Institute in New Zealand).

Brief Summary The referendum of June 2016 resulted in the decision to leave the European Union (EU). Since the delivery of the â&#x20AC;&#x2DC;Leave Letter' in March 2017, the process of Brexit has developed, not without some unexpected twists. At this point in time it is not decided which exact course of action the UK-government wants to take and what kind of post-Brexit relations would develop. The main challenge is the position of Northern Ireland (NI) in a post-Brexit constellation: Avoiding a hard border on the island and simultaneously leaving the Single Market (SM), as well as the Customs Union (CU), turns out to be a conundrum that seems difficult to solve. This report is interested in the implications of the Brexit vote for NI. Given the fluid negotiation situation, we look into three scenarios. First, a hard Brexit where the UK leaves the SM as well as the CU. Second, the option where NI would stay in the SM and CU and the other parts of the UK go for a hard Brexit. Third, the scenario where the Republic of Ireland (ROI) and NI opt for unification. All three scenarios are anchored in ongoing debates. The hard Brexit scenario reflects the power struggle within the Tories and assumes that a possible impasse may result in leaving the SM and the CU. The dual regime scenario reflects the suggestion of the EU that NI could stay in the SM and the CU to avoid a hard border on the island. The unification scenario refers to political suggestions to overcome the Brexit conundrum by uniting NI and the ROI. To quantify economic outcomes of each scenario for NI, we make use of an updated version of our previous NIROI-model as well as of the most recently published national account data. The modeling consists of four steps: (i) a calculation of a benchmark path where we assume that the Brexit vote would not have happened and established growth trajectories stay fixed (businessas-usual); (ii) a calculation of a hard Brexit where the UK retreats to WTO-tariffs with the EU; (iii) a calculation of a constellation where NI stays in the SM and the rest of the UK trades to WTOtariffs; and (iv) a calculation where NI leaves the UK and stays with the ROI. As preliminary simulations showed that the scenarios Unification and Brexit work in opposite directions and for easier comparison with the previous study, we decided to keep the start year of the Unification as in the previous studio. The results now show for the years up to the Brexit scenario â&#x20AC;&#x153;pure Unification" effects as well as unification effects for NI and ROI after Brexit happened. The "Brexit"-Scenario assumes that NI remains part of the UK.

The outcome of this exercise is straightforward (see Summary Table). A hard Brexit would reduce the GDP of NI over the period 2021 to 2025 by Euro 10.1 billion. This represents a per capita loss of Euro 5035. In the scenario where NI stays in the SM and the other parts of the UK opt for a hard Brexit, the accumulated loss of GDP would be smaller but still amounts to Euro 3.8 billion. This reduction translates into a per capita loss of Euro 1921. The only winning scenario is the case of unification where between 2018 and 2025 NI would increase its GDP by Euro 17.9 billion, i.e. an increase per capita of Euro 9070 over the period.

Summary Table: Three Scenarios

NI 2017 2018 2019 2020 2021 2022 2023 2024 2025 Total

ROI 2017 2018 2019 2020 2021 2022 2023 2024 2025 Total

Change in GDP in Billion EUR Unification Brexit_SM Brexit 0.0 3.6 3.8 4.0 1.1 1.2 1.3 1.4 1.5 17.9

0.0 0.0 0.0 0.0 -0.8 -0.8 -0.8 -0.8 -0.8 -3.8

Change in per-capita-GDP in EUR Unification Brexit_SM Brexit

0.0 0.0 0.0 0.0 -1.9 -2.0 -2.0 -2.1 -2.1 -10.1

Change in GDP in Billion EUR Unification Brexit_SM Brexit 0.0 0.2 -0.4 -0.2 0.7 0.9 1.2 1.4 1.7 5.6

0.0 0.0 0.0 0.0 -2.5 -2.6 -2.8 -2.9 -3.1 -13.8

0 1,862 1,932 2,005 552 601 653 705 760 9,070

0 0 0 0 -387 -386 -384 -383 -382 -1,921

0 0 0 0 -962 -986 -1,010 -1,035 -1,060 -5,053

Change in per-capita-GDP in EUR Unification Brexit_SM Brexit

0.0 0.0 0.0 0.0 -5.9 -6.2 -6.5 -6.8 -7.0 -32.4

0 48 -69 -29 124 167 212 259 310 1,021

0 0 0 0 -457 -479 -502 -524 -548 -2,510

0 0 0 0 -1,105 -1,141 -1,177 -1,215 -1,253 -5,891

We also calculate all three scenarios for the ROI. The results mirror the outcome for NI. The case of a hard Brexit, as well as the scenario where NI stays in the SM, are both causing high economic costs. The only scenario with net benefits is the case of unification. Unification benefits for the ROI are smaller than for NI but continue to be positive. Those quantitative outcomes of the modelling exercise allow us, in a last step, calculating the costs of non-unification. These are economic opportunity costs that occur when political actors deliberately make decisions that reject first best options and instead go for inferior options. The costs of non-unification are the highest in case of a hard Brexit where they amount to Euro 28 billion over the period. The concept of costs of non-unification refers to the implied economic costs of political choices. Political actors may for a variety of reasons favor a particular option, and the concepts helps to understand the economic implication of choices.

Executive Summary 51.9 % of UK voters voted in favour of leaving the EU in the referendum on June 23rd, 2016. However, the outcome of the referendum varied across areas of the UK as well as between social groups of citizens. Northern Ireland (NI) voted to stay in the EU, with 56% voting to remain within the EU. After triggering Article 50 in March 2017 following a highly unsuccessful bid to create a larger Tory majority Parliament by calling early elections in June 2017, Brexit negotiations between the EU and the UK started in earnest. The negotiations have proved difficult and the original schedule for a successful outcome in March 2019 has become less and less unrealistic. A transition period for 21 months has been agreed between both parties which postpones a final agreement about future relations between the UK and the EU until the end of 2020. For the time being, negotiations are dealing with the terms of separation. Only when those terms are handled to the satisfaction of both sides will negotiations turn to the shape of future UK-EU relations. One of the three requirements for opening talks about future UK-EU relations put forward by the European Council deals with the future of Northern Ireland (NI), with the intention of protecting the Good Friday Agreement and avoiding the erection of a hard border between the Republic of Ireland (RoI) and NI. In November 2017, the EU indicated that an ‘all-island’ approach whereby NI should stay in the Customs Union was required. The joint report released by the British government and the EU on 8 December 2017 states that an ‘agreement in principle has been reached on the package, as opposed to individual elements’. The document set out to deal with the tenuous issues of the divorce bill, the rights of citizens, and the issue of the Border in Ireland. In politically refined language, the section on the Irish Border – encompassing points 42 to 56 of the document – speaks positively and diplomatically about North-South issues on the island of Ireland. The section clearly expressed Britain’s recognition of the unique position of Ireland in the face of Brexit. The British government makes clear their aim to foster North-South relations further into the future and to uphold the Good Friday Agreement and safeguard the peace process through a strong and workable agreement. However, mere recognition by the British government on these issues tells us little or even nothing about the future form of Brexit and the path that Britain will eventually take. Where a viable agreement cannot be made on UK-EU trade relations, the British government promised to maintain full regulatory alignment between the

North and South of Ireland. In the course of the negotiations regarding the transition agreement the UK government put it in writing that as a default mechanism Northern Ireland and the Republic of Ireland would remain in regulatory alignment in the instance that EU and the UK do not reach an agreement: “DESIRING to create a common regulatory area on the island of Ireland in order to safeguard North-South cooperation, the all-island economy, and protect the 1998 Agreement” (TF50, 2018). Such a common regulatory area would imply that NI would stay in the Common Market and the Customs Union. We include this fall-back position into our modelling exercise. From a purely economic perspective, the political decision to leave the Single Market, as well as the Customs Union, are logical steps if the goal is to make the UK a truly sovereign country regarding trade policy, economic laws, norms, and standards as well to avoid any ‘meddling’ on the side of the European Court of Justice into British affairs, at least in a formal, de jure sense. Staying inside the Single Market and/or the Customs Union would prohibit a UK government the kind of sovereignty they are seeking. However, leaving both the Customs Union as well as the Single Market implies a hard Brexit. The price of a hard Brexit is high, in economic as well as in political terms. In economic terms, a hard Brexit implies that the UK becomes a third party when it comes to the EU, and thus will be treated as a non-member economy. In essence, this means economic relations on the basis of WTO terms. In political terms, this means the re-installation of a hard border on the Irish island that divides Northern Ireland and the Republic of Ireland in serious legal ways. Such a decision would probably also jeopardize the benefits of the Good Friday Agreement and thus revive a whole range of past issues between both entities. Early assurances of the British government as well as from the side of the EU to not violate the spirit of the Good Friday Agreement may have been serious statements but are contradicting harsh economic realities for a situation where one entity is a member of the Common Market and the Customs Union, and the other is not. In other words, the Irish question is back on the table. This report is not discussing the political implications of a hard Brexit. Nor does it intend to investigate the range of potential outcomes of Post-Brexit negotiations between the UK and the EU. If we see the outcome of Brexit negotiations as a continuum where one pole is ‘staying in the EU,’ and the other pole is ‘leaving the Customs Union and Single Market’, then we are mainly interested in the latter. However, we also include a scenario whereby NI would stay in

the SM as well as in the CU as a means to avoid a hard border on the Island. We include this option because the ongoing debate about solving the ‘Northern Ireland problem’ assumes that such a route may be a viable way to overcome any political impasse. Our modelling quantifies this option. The emphasis on a hard Brexit scenario in this report is driven by purely economic reasons, as well as by simple pragmatic reasons. The economic rational is straightforward: the customs union and single market are well-defined concepts which come with economic benefits that are shared only by members. Any step to extend the benefits to non-members would undermine the economic rationale for both schemes. Hence, we assume that the EU will act in self-interest when it restricts benefits of membership solely to members. Given that the British government stressed strongly that potential models offered by other countries who have close relations with the EU without being a member of this entity are not on the wish list, we conclude that a WTO-option may become reality. There are options to engage in relations with the EU other than continuing to be a member of the Single Market and the Customs Union, often labeled as the ‘Norway option’ or the ‘Switzerland Model.’ At this point, all those options have been turned down by the UK. Thus, there is no compelling need to model quantitative effects of such routes. The pragmatic reason for focusing on the hard Brexit axis is grounded in the principle of parsimony: We focus our efforts on one extreme and one less extreme case, both of which seem to us probable outcomes of the ongoing negotiations. Rather than speculating about a wide range of outcomes of Brexit negotiations, we decided to look more closely at a scenario of a hard Brexit and a specific NIscenario. During the time of the preparation of the report, some indicators, in particular, the circumstances of the ongoing negotiations, suggest that one of those scenarios may become the actual outcome. In the recent past, we published a study that looked into the economic effects of a Unification scenario, i.e., an economic and political unification of Northern Ireland and the Republic of Ireland. This study was prepared before the Brexit vote of June 2016. Our study showed that under specific assumptions, Unification generates strong positive welfare effects for both entities, more so for the North than for the South. For the new study, we created an updated model with data that was not available at the time of the first report. We did this exercise with the intention not only to quantify in a model setting the effects of Irish unification but also to quantify the costs of non-Unification under the conditions of a hard Brexit. The costs of non7

unification can be interpreted as opportunity costs for a particular policy choice. In other words, our new study is interested in calculating the effects of a hard Brexit on NI as well as the effects of a unification of RoI and NI. We then can compare the effects of a hard Brexit and the effects of unification. The difference of economic effects is interpreted as the costs of non-unification (see graph 1). The advantage of this approach is that it shines a light on opportunity costs that emerge if political actors decide not to go for an economically superior outcome but prefer to stick to a less beneficial option. The costs of non-unification indicate the effects of policy choices. In this respect, our approach is in line with the â&#x20AC;&#x2DC;Cecchini Reportâ&#x20AC;&#x2122; (1988) that at the time calculated the costs of non-completion of the European Single Market and made the case that an inferior policy decision comes with (avoidable) economic costs. The new modeling not only includes new data but also takes into consideration that the Brexit vote changed some of our previous parameters. One of the immediate responses to the outcome of the referendum was a depreciation of the British Pound Sterling against the Euro as well as the US-Dollar. Financial markets priced in their negative sentiments about the outcome of a potentially hard Brexit. Rather than pushing British exports, the depreciation of the British Pound added a further dose of uncertainty that resulted in a relatively weak investment climate that eventually lowered economic growth. Even before Brexit actually happens, we have seen the first negative repercussions. The first negative repercussions of Brexit have already been experienced prior to the completion of Brexit. In our previous modelling exercise, we identified a strong positive currency effect for NI in case of Irish unification. The Brexit referendum changed this variable fundamentally. The loss of trust of financial markets in the economic strength of the UK has implications for any form of Irish Unification: Northern Ireland, unlike in our first modeling exercise, would now move from a now relatively weak currency zone (British Pound) into a relatively strong currency zone (Eurozone). One would expect that the exchange rate effect will, ceteris paribus, dampen potential unification benefits as NI-companies would now have to adjust to an environment with relatively better price competitiveness. One would also expect that the overall negative effects of a hard Brexit are pretty strong for NI. Thus, it was relatively open whether the new modelling with newly provided statistical data would still result in net benefits of Irish unification. The outcomes of our modeling exercises show high costs of non-Unification as well as positive unification effects. This result is implied by two distinct modeling procedures. First, we

recalculate the Unification scenario and show that the whole set of new data confirm our previous result of positive welfare effects. The unification effects remain strongly positive. The positive but smaller effects are due to two factors. One reason for this difference is the adjustment of the assumption on the effect of government harmonization: Our previous assumption of a considerable lower efficiency and quality of NI government expenditure was criticised after the release of the first study, especially with regard to the health system in Northern Ireland. We, therefore, tempered the assumptions in the recalculation of the unification scenario. The second reason for the reduced positive effect is the impact of the Brexit decision which had immediate negative repercussions for the UK, and thus also for the calculation of the economic effects of Irish unification. Given that the referendum decision led to immediate negative economic developments, in particular to a strong depreciation of the British Pound, the overall positive result is a surprising outcome of the modelling. This indicates that it is not only the currency effect that drives Irish economic unification. In a further exercise, we use our model to calculate the effects of a hard Brexit for Northern Ireland. The effects of a hard Brexit are strongly negative. Compared to a hard Brexit, unification is by far the better option. As a matter of fact, it is the only option with positive net effects. If political actors nevertheless prefer a hard Brexit, or if they are willing to accept a hard Brexit for overarching political reasons, then they accept willingly high negative economic costs. Our modeling shows that unification is also the superior path compared to a benchmark scenario (business-as-usual) that assumes that NI would continue to be an integral part of the UK and that the Brexit referendum would not have happened. Finally, we calculate a scenario where NI stays in the SM as well as in the CU, and the rest of the UK leaves the EU in a hard Brexit fashion. In economic terms, such a scenario is superior to a hard Brexit as it avoids the high negative costs, but it is inferior to a unification option as its net effects for NI are smaller than the effects of the Unification scenario.

Pound Exchange Rate, Brexit, and Irish Unification Foreign exchange dealers were as surprised as most political observers when the Brexit referendum on June 16, 2016, resulted in a majority vote to leaving the EU. A first and most lasting consequence was a sharp devaluation of the British Pound against the main currencies. The exchange rate with the Euro devalued by 19 %, compared to the pre-Brexit vote rate. Since then, the exchange rate showed a slight appreciation, despite the fact that negotiations between the EU and the UK have run into challenges. One can argue that the exchange rate has already illustrated the consequences of a potential hard Brexit. The devaluation of the British Pound has consequences for the scenario of Irish unification. Unlike in a pre-Brexit vote environment, NI would now – post-Brexit vote – move from a low exchange rate zone to a relatively high exchange rate zone. This would make – ceteris paribus – NI exports to the UK and the ROW more expensive, and at the same time imports from relatively cheaper. Without compensating policy responses, one would expect that unification effects would be affected. We are testing the sensitivity of a currency appreciation for NI in a scenario of unification. To do this, we still assume an appreciation effect for NI of 10 %. The modelling shows that in such a case economic unification effects would become larger than in our genuine calculation. Exchange rates are forward-looking asset prices and as such less driven by economic fundamentals than by expectations on the side of critical foreign exchange market actors. The formation then is guided by more or less theoretically founded narratives that help actors to navigate in a world of high uncertainty. Narratives, as Robert Shiller (2017) reminded us, are not per se ‘right' or ‘wrong', but still can be key drivers for the formation of expectations that decide about asset prices today. Irish unification and Brexit are both scenarios without any precedent, and thus are characterised by uncertainty. The more clearly these scenarios are communicated, and the more coherent the narratives are presented, the stronger their impact on asset prices. The exchange rate of the British Pound and the Euro are not solely driven by Brexit. This is particularly relevant to the Euro exchange rate which – also due to its status as second most important international money – is driven by other narratives, of which the internal stability of the Eurozone and global political economy processes like trade relations are most important. It is thus a possibility that the Euro exchange rate may depreciate over time, and this will have implications for the British Pound and in consequence for a scenario of Irish unification. A slightly lower depreciation of the British Pound would result in stronger positive unification effects, compared to our modelling procedure as shown in the next figure. Here we assume an appreciation rate of 10% for NI in case of unification with the Republic. We compare such a benign scenario with the more conservative scenario in the report.

Unification with 10% appreciation

Main Unification scenario

We see a clear increase in GDP compared to the main Unification scenario (right figure). The 10% appreciation appreciation--scenario scenario increases the GDP gains from 382 to 390 Billion EUR (an increase by 2%). This clearly shows that the development of the exchange rate iiss a key factor for the Unification scenario: any further appreciation of the British Pound would increase the gains of unification.

Our approach allows us to compare four trajectories: (i) A business as usual path where whereby by the UK mov moves es along without lea leaving ving the EU; (ii) A unification path where whereby by NI now is unified with the South South; (iii) A Brexit path where whereby by Northern Ireland experiences the effects of a hard Brexit as a sub sub--entity entity of the UK UK;; and (iv) A scenario in which NI remains in the Single Market and the CU (Brexit_SM).

Figure 1-1 1:: Comparing the effects of U Unification nification nification, Brexit Brexit, and Brexit_SM Brexit_SM:: Overall GDP and changes in GDP

Figure 1--1 shows the outcomes of the exercise exercise.. A hard Brexit would result in a GDP loss of about 10 billion Euro for our calculation period (calculated as a difference to the benchmark). Compare this to the effects of Unification wh where ere the gains amount to about 18 billion Euro over the period. The costs for opting for a hard Brexit compared to the unification scenario is a staggering Euro 28 bn over the period period.. Those figures stress the opportunity costs associated a hard Brexit. In n the case in which NI stays in the SM and the CU, the effects on GDP would be roughly neutral, compared to the benchmark. However, such an option is not more than a second best, as tthe he costs of non non-unification unification in comparison to staying in the SM and the CU are still about Euro 14 bn bn. Ultimately, u uni nification fication is by far the superior scenario and has in itself strong positive GDP GDPeffects.

Figure 1-2 2:: Comparing the effects of Unification, Brexit Brexit, and Brexit SM: Overall GDP and changes in GDP per Capita

In a per capita capita--perspective, perspective, a hard Brexit would result in a relative lloss oss per capita of about Euro 5,0 5,000 00 over the period of six years, compared to the benchmark. The losses start to pile up after the first three year yearss of Brexit. In the case in which NI would stay in the SM and the CU, the result will be more or less the same in the hypothetical situation as in the benchmark scenario that assumes that NI would continue to be an integral part of the UK and there is no Brexit. Unification is by far the better option option,, with overall per capita gains, compared to the benchmark, of Euro 10,000. Per capita costs of non non-unifi unification cation â&#x20AC;&#x201C; relative to a hard Brexit â&#x20AC;&#x201C; wo would uld amount to about Euro 14 14,000 ,000 over the mode modeling ing period.

Background The economic effects of a hard Brexit have been calculated by a multitude of studies. As much as they differ in methodologies and underlying analytical models, a convergence of results can be observed: in economic terms, Brexit will be costly for the UK. The negative GDP-effects will increase the more closely the outcome of the negotiations will resemble a hard Brexit. Economic effects do not wait for the end of negotiations, however, and the level of uncertainty is already creating negative effects. Born et al’s (2018 study shows that the Brexit vote has already resulted in significant economic costs of about 300 million BP per week (Born et al., 2018). This figure is derived from a specific modeling procedure that uses an algorithm to create a synthetic UK-benchmark’ economy and then compares actual UK growth with the growth of this ‘double’economy. The Brexit vote generated quite some uncertainty on the side of economic actors and that this change in sentiment contributed to the relatively weak development of the UK economy in recent periods. At a time where forecasts of economic growth for large economies as well as for EU members indicate improved growth prospects, forecasts for the UK are reversed downwards (IMF 2018). Typically, scenario studies either make use of a gravity model or a computable general equilibrium (CGE) framework that allows for calculating feedback loops on GDP. Rarer are sectoral studies that would create numerical outcomes of Brexit effects for particular sectors. An exemption is (Gasiorek et al., 2018) who make use of a partial equilibrium model that allows the modelling of the impact of Brexit on prices, exports, imports, and overall output for 122 manufacturing sectors of the UK economy. They examine five post-Brexit scenarios, none of which produce net positive effects for manufacturing. The majority of studies examine the UK as a whole, and state severe losses in the scenario whereby the UK would leave the Single Market (SM) as well as the Customs Union (CU). A modelling exercise undertaken by Copenhagen Economics (2018) applies a CGE-model to simulate four scenarios overall, all of which show negative effects for the Republic of Ireland (RoI). Most devastating is a WTO-scenario that would result in a 7% loss of GDP until 2030. Modelling exercises demonstrate that Northern Ireland (NI) will be significantly hit by a hard Brexit. A study commissioned by the UK-government shows that the GDP losses for NI would be 12% over a period of fifteen years. Taking account of the inferior economic position of NI within

the UK in terms of GDP per capita, such an outcome would be no surprise, for a variety of reasons. First, a hard Brexit would create a land border between an EU- member state and a non-EU member state on the island of Ireland. Given the high level of economic transactions between both entities and given the unequal economic conditions between North and South, one can expect relatively higher costs for the weaker entity. In a hard Brexit scenario, the currently invisible North-South border between the Republic of Ireland and NI will become the EU’s external border with the UK, and this will require forms of a hard border, inclusive of extensive controls which will result in a strong increase of immediate transaction costs. Second, NI is strongly intertwined with RoI regarding goods and services as well as regarding crossborder engagements of workers. Trading costs will increase for both sides. In particular, one can expect additional transaction costs for economic actors alongside the border who will feel the new hard border very directly. According to the Centre for Cross Border Studies (CCBS), there are between 23,000 to 30,000 cross-border workers between NI and the RoI. For the year 2015, Eurostat manifests 7,600 cross-border commuters from NI to RoI and 12,100 conversely. In contrast to the UK, whose cross-border commuters as a share of overall employment add up to 0.25 %, the 1% number for NI is comparatively high, which in turn indicates the severity of a restriction on free movement of people for NI. Third, NI’s trade mainly takes place within the EU. In 2014, these exports accounted for £3.63 billion, whereas trade with non-EU countries totaled £2.35 billion. NI’s relatively greater dependence on the EU as an export market suggests more negative consequences on NI than on GB. When it comes to trade with the EU, the RoI constitutes the most critical trade partner. In 2015, NI’s exports towards the RoI represented 61 % of all its exports to the EU. NI’s export share towards the RoI constituted 34 % of its total exports. In the case whereby GB left both the Common Market as well as the Customs Union, those figures suggest that NI will experience negative repercussions. In advance of the referendum, a study was undertaken in order to examine the potential impact of a Leave-decision upon the economy. The eventual outcome of the referendum then triggered more efforts to calculate possible effects of various forms of Brexit. Table 1-1 shows some of the most prominent studies put forward in the last two years. Northern Ireland Assembly, Oxford Economics, and the National Economic Research Institute (NERI). The ESRI report focuses predominately upon the economic consequences of Brexit on the RoI. The HM Treasury report utilises the well-established ‘gravity modeling approach by calculating effects on the UK. Those

studies differ in many respects, not least in the choice of methodological approaches. All of the studies demonstrate a reduction in GDP. Table 1-1: Results from actual studies on the impact of Brexit on GDP

Study

Scenario

GDP % Change Relative to Base

NIESR

WTO (short-term)

-2.3

HM Treasury

(short-term)

-3.6

OECD

(short-term)

-3.3

IMF

Downside (short-term)

-0.9

European Commission

Mild (short-term)

-1.0

NIESR

Severe (short-term) EEA

-2.7 -1.8

HM Treasury

FTA EEA

-2.1 -3.8

OECD

FTA WTO/FTA (Optimistic)

-6.2 -2.7

LSE/CEP

WTO/FTA EEA/FTA (Central)

-5.1 -7.9

PwC

FTA

-3.0

Oxford Economics

WTO Liberal Customs Union

-5.5 -0.1

Bertelsmann Foundation

FTA EEA

-2.8 -0.6

WTO

-3.0

Source: Bergin et al., 2016

The HM Treasury calculations indicate that by 2030 the GDP of the UK could shrink by 3.8 % under a European Economic Area (EEA) agreement; by 6.2 % under a less favorable bilateral agreement; and by 7.5 % under a WTO scenario. In other words, whatever option is selected, Brexit would create high economic costs for the UK. Impact analysis is difficult to undertake on a sub-national basis, thus most Brexit studies have been limited with respect to assessing regional effects. A study undertaken by Budd (2015) for

the Northern Ireland Assembly attempts to extrapolate overall UK data to NI in order to estimate the impact of Brexit on NI. When assuming a growth rate differential between NI and the UK of one third and an unemployment rate that is twice as much for NI than for the UK, then a decline in UK’s GDP of 2 % would be equivalent to a 3 % lower GDP in NI. The study conducted by Oxford Economics, commissioned by the Department of Enterprise, Trade, and Investment (DETI), distinguishes nine scenarios and focusses on a comparison of NI with the UK as a whole from a long-term perspective. In order to quantify the economic implications of the nine scenarios, a separate model for the UK’s NUTS 1 regions has been developed. This model simulates the impact of changes in input factors labor and capital due to Brexit. The overall result of the study shows that in seven out of nine scenarios there are much higher economic losses (expressed in Gross Value Added – GVA) for NI in comparison to the UK. GVA losses range from a minimum of 0.1 % to a maximum of 5.6 %, depending on the scenario (Oxford Economics, 2016, p. 6.). A similar calculation is presented by Flynn (2016) whose work focused upon the prominent role of EU markets for NI and concludes that a loss of free market access for NI would generate over-proportionate costs for NI, compared to the negative costs for the UK. Most assessments of the economic consequences for the UK of Brexit under alternative scenarios indicate that the economic losses are likely to outweigh savings from cancelled annual payments to the EU budget (Ottaviano et al., 2014) and thus contradict a major aspect of the argument used by representatives of the ‘Leave camp’ during the referendum campaign. Ottaviano et al. (2014) estimate that real GDP losses in the UK would be between 1.1 and 3.1 percent, depending on the underlying Brexit scenario. This will be larger if one assumes that Brexit will have a negative impact on productivity with economic losses of up to a maximum of 9.5 percent. A simulation by Felbinger from the Ifo Institute in Munich (Bertelsmann Study) concludes that a UK Brexit in 2018 would result in a lower GDP per capita in 2030 by a minimum of 0.6 and a maximum of 3.0 percent, compared to GDP per capita if the UK were to stay in the EU. The same study finds that among EU countries, Ireland will be hardest hit by Brexit, with an anticipated GDP per capita loss of 0.8 to 2.7 percent. It should be stressed that those calculations focus upon trade and do not take into consideration the effects of other variables’ effects on R&D in case of a hard Brexit where the UK will get cut off from free research cooperation with EU entities. Using a standard quantitative general equilibrium trade model, Dhingra et. al. (2017) calculate that a hard Brexit would lead to a welfare loss of 2.7% for the average UK household. The losses would be smaller if the UK follows the Norway-model but still 17

be in the range of 1.3%. The gravity model applied by Oberhofer and Pfaffermayr (2018) suggests that a hard Brexit would result in a reduction of real income between 1.4% and 5.7%, depending on the kind of trade agreement with third parties the UK could close. Simulations on the base of the HERMES Model suggest that the effect of a 1 % reduction in UK GDP is the reduction of Ireland’s GDP and GNP by 0.3 % in the medium term; employment would fall by 0.2 percent, and unemployment would rise by 0.2 percent. Another methodological approach explores the effect of general trade agreements in order to estimate the effects of Brexit. Morgenroth (2015) uses numerical estimates from a study by Hufbauer and Schott (2009) to assess the impact on aggregate trade flows between Ireland and the UK. The work of the latter authors shows based on an extensive dataset, that bilateral trade between the UK and EU countries would be reduced by 21.6% in the event of a soft Brexit. The Central Bank of Ireland (2016) published an estimate of the impact of Brexit upon the Irish economy, based on a Bayesian Vector Autoregression approach that makes the case that a hard Brexit would result in a reduction of GDP of more than 3 % over a period of ten years compared to a no-Brexit constellation. The COre Structural MOdel of the Irish Economy (COSMO, see Bergin et al., 2016) explores the impact of Brexit on other parties besides the UK and thus takes into consideration the negative effects on ‘the rest of the world’ and considers the implications of the effects of Brexit on Ireland. The results of this modeling are straightforward: Under an EEA scenario, the output level of the RoI would decrease by 2.3 %; in case of an FTA, output would shrink by of 2.7 %, and in a WTO scenario, output would shrink by 3.8 %. The output reductions are mainly ascribed to the reduction in trade-weighted demand for Irish exports. Although these studies differ in methodological terms as well as in underlying assumptions, it is fair to state that they do converge when it comes to estimations of the outcomes of the various versions of Brexit which have been explored. In the medium-term, they all suggest a loss of welfare for the UK and for NI. The impacts of a hard Brexit are the most severe. We are adding to this literature in a particular way: Rather than modeling only Brexit effects for NI, we also model a scenario of economic unification of NI and the RoI. This allows us to calculate the costs of non-unification. The costs of non-unification, then, is the difference between unification effects and GDP-growth along a path that assumes the Brexit vote would not have happened (benchmark) and the costs of a hard Brexit relative to the benchmark. The costs of non-

unification, thus, are the costs if political choices occur. If the British Government (i) opt for a scenario that comes with high negative costs and (ii) thus exclude a scenario that comes with positive net benefits. In other words, our contribution gives numerical values to such policy choices. Not only does our modeling exercise show effects of specific economic pathways but we can also calculate the numerical values to relative economic effects of chosen pathways. If the UK exclude unification and go for a hard Brexit, then the implied costs are very high. The calculation of the costs of non-unification under conditions of Brexit follows the approach taken by the Checchini Report (1988) which aimed to calculate the the costs of the noncompletion of the Single Market in regards to the EU. This report presented a modeling framework which illustrated that creating a unified single market for the EU economies would generate significant economic gains compared to a scenario in which member states would keep up their existing barriers. The costs of non-integration were seen as unreasonable, given the superior alternative. The same can hold for the concept of costs of non-unification. When the effects of a hard Brexit are negative, and the effects of unification are positive, then not opting for the superior scenario implies the acceptance of high opportunity costs. Our modeling shows that real costs of a hard Brexit for NI are significant. Deselecting a Unification scenario and sticking to a hard Brexit implies that Brexit not only comes with â&#x20AC;&#x2DC;hardâ&#x20AC;&#x2DC; costs but also does not make use of positive economic possibilities. From an economic perspective, this would be a non-Pareto optimum: the economic outcome for average citizens in NI could be significantly improved by unification. If NI would get the right to stay in the SM and the CU whereas the rest of the UK leaves the EU, most of the negative effects of a hard Brexit for NI would be avoided. Still, the positive effects of unification continue to be large and actually are higher than the effects of staying in the DSM and CU. Hence, staying in the SM and the CU is only a second-best option for NI. Our calculations are conservative, not least because the argument can be made that the actual economic development in the UK and NI has moved those economies already on an inferior growth path. Most studies modelling Brexit effects illustrate the impacts that are likely to occur after the UK leaves the EU. At present, the UK remains part of the EU. Nevertheless, more recent economic forecasts indicate that the initial negative effects of the leave vote are already apparent within the economy. As a consequence, the costs of non-unification are probably higher than our model shows. The more detailed negotiations between the UK and the EU are 19

becoming, the more obvious it becomes that our assumptions regarding the size of the transaction costs are on the lower scale. At this point, a hard Brexit with a hard border on the Island would be likely to create significant extra costs in the form of policing, and additional administrative burdens, including things such as increases in the number of police and border staff. Any hardening of the border in Ireland will add extra time, and create problems for crossborder supply chains. The fiscal effects of Brexit may be much higher than we assume, with the consequence that transfers to NI may be reduced in absolute terms. Depending on the strength of inflation effects and also potential capital flight, the Bank of England may have to increase its key lending rate, thus discouraging investments and consequently further reducing growth dynamics. The option to keep NI in the SM and they CU is definitely superior to a hard Brexit, but still inferior to Unification. Such an option would probably create severe political problems: if NI stays and the rest of the UK leaves, the result would be a two-tier regime of regulatory practices within one entity. NI would be a part of the EU in both regulatory and political terms, while the rest of the UK would establish its own distinct regulatory regime. The question would then arise as to why the implied costs of such a regime should be accepted and why this duality should not be overcome by going one step further towards unification?

Scenario “Brexit”

2.1

Introduction In the following chapters, we will have a look at the effects of Brexit on Northern Ireland and the Republic of Ireland. The Brexit-Scenario is based on several assumptions regarding the possible effects of Brexit on the economy of Northern Ireland and the Republic of Ireland. One of the main questions is how the UK will leave the EU. There are several options for the UK which range from completely leaving the EU (“Hard Brexit”) to a close association agreement within the single market and the customs union (see, for example, Dunt, 2016 and Kauders, 2016). In this report, we assume the most extreme case, “Hard Brexit”, as many scientific studies foresee the largest negative economic effects for the UK and thereby for Northern Ireland (for a critical evaluation of several studies see Whyman & Petrecu, 2017). If the event of a Hard Brexit in 2021, together with the failure of UK to close any trade or other agreements with the EU and the relevant countries in the rest of the world, this could, according to several studies1, have the following four main effects: 1. The UK is forced to apply the tariffs and quotas according to the WTO trade agreement to all non-UK countries and an increase in non-tariff-barriers (NTBs); 2. it would see a drop in immigration from the EU; 3. the exchange rate of the British pound will devaluate; and 4. foreign direct investment will decrease. These assumptions are quantified and translated into changes in model parameters. We make a distinction in four decomposition scenarios and one scenario that combines the four main effects. The assumptions regarding the model parameters are shown in Table 2-1.

The most relevant studies are discussed in Chapter 1.

Table 2-1: Brexit scenarios and assumptions

Scenario

Description

Assumptions

Brexit_TR

Adopting WTO

NI: Adoption of tariffs, etc. of GB for EU, ROW; ROI: adoption of tariffs of GB; an increase of NTB costs

Brexit_IM

Immigration effects

Reduction of the workforce by 0.85%

Brexit_XR

Devaluation

Fall GBP (devaluation of 10%) for both NI and ROI

Brexit_Prod

Fall in FDI

Fall in productivity by 1%/a

Brexit

all above

Note that the NIROI model only has an explicit treatment of Northern Ireland and the Republic of Ireland. Only the flows of goods, services, labor, and capital, are considered for Great Britain, the EU, and the rest of the world (ROW). We, therefore, assume that the estimated effects for the UK affect Northern Ireland in the same way (e.g., the same percentage of change in labor migration, same tariffs, same devaluation, etc.). This is also the reason why we have refrained from adding an additional decomposition scenario, in which the positive overall effect of fiscal transfers, which probably will account for four billion GBP net (no annual payments to the EU minus the “divorce” payments). This total income gain for the UK would only have a minor impact on Northern Ireland. Note that in the model we can only show the effects of Brexit on the economies of NI and RoI. This means that the Brexit effects on the UK are not reflected in the results. For the UK, Brexit could lead to a lower GDP, which would also have a negative impact on both NI and RoI. Estimates for these effects range from -1.3 to 7.9% (see Whyman and Petrecu, 2017, Table 1.1, p.7). Therefore, the effects shown could be considered to be optimistic. The results are compared to a “Benchmark scenario”. The benchmark or baseline scenario is a constant growth rate recursive dynamic forecast of the economies in NI and the ROI. It runs for 12 years from 2014 to 2025. Annual growth rates for exogenous transfers, government consumption, and investment in NI and the ROI are set to a long-run trend of 1.5% and 3%, respectively. It should be re-emphasized that neither the benchmark scenario nor the counterfactual scenarios are economic forecasts. Rather they are trending scenarios used comparatively to measure counterfactual effects. This means the same growth rates used in the benchmark are used in the counterfactuals.

Benchmark and counterfactual labour supply grow at compound annual growth rates derived from population data for the two regions from 1966 to 2011. The ROI annual population growth rate is 1.05%, while in NI it is 0.45%. Capital in both regions accumulates at a rate derived from an internal-rate of return, a depreciation rate and initial capital stock and investment levels.

2.1.1 Scenario “Brexit_TR” In the decomposition scenario “Brexit_TR”, we have implemented the WTO tariffs based on the work done by Ottaviano, Pessoa, and Sampson ( 2014). Table 2-2 shows the tariffs and NTBs according to the sectoral aggregation of the NIROI model.

Table 2-2: Tariffs, export taxes, and NTBs after the Hard Brexit Sectors AGR

Agriculture, forestry and fishing

Tariff

Export Tax

5.9%

5.6%

NTB 5.8%

MIN

Mining and quarrying

FOO

Food products, beverages, tobacco

7.3%

5.0%

23.9% 5.8%

TEX

Textiles, wearing apparel, leather

9.6%

9.7%

16.0%

WPP

Wood and paper

2.4%

3.6%

16.0%

OTM

Coke, refined petroleum products, furniture, other manufacturing, repairs

2.7%

2.8%

16.0%

CHE

Chemicals and chemical products

1.8%

1.9%

16.0%

BAS

Pharmaceutic industry

1.8%

1.9%

16.0%

PLA

Rubber, plastic and other non-metallic mineral products

4.6%

4.2%

16.0%

MET

Metal products

2.1%

1.9%

16.0%

COM

Computer, electronic and optical products

2.1%

1.9%

16.0%

ELE

Electrical equipment, machinery

2.0%

1.8%

6.5%

TRE

Vehicles and other transport equipment

8.1%

7.2%

22.1%

WAT

Electricity, gas, steam and air conditioning supply; Water supply; sewerage, waste management and remediation activities

WRT

Wholesale and retail trade; Repair of motor vehicles and motorcycles; Transportation and storage

11.7%

ACC

Accommodation and food service activities

11.7%

INF

Information and communication

11.7%

FIN

Financial and insurance activities

11.3%

REA

Real estate activities

11.3%

PRO

Professional, scientific and technical activities

11.3%

ADM

Administrative and support service activities

14.9%

PUB

Public administration and defense; compulsory social security

14.9%

4.6%

EDU

Education

14.9%

HES

Human health and social work activities

14.9%

ART

Arts, entertainment, and recreation; Other service activities

14.9%

OTH

Activities of households as employers; undifferentiated goods- and services-producing activities of households for own use

14.9%

Imports and exports are getting more expensive due to the increase in tariffs, export taxes, and the NTBs. This will reduce NI exports as well as NI imports. The reduction of exports will have a negative effect on GDP. The reduction of imports will lead to a higher reliance on domestic production. However, this will cause an increase in domestic prices and therefore in domestic production. The overall effect on GDP will, therefore, be negative, as can be seen in Figure 2-1. The total effect for the years 2021 to 2025 is a decrease of around 1 billion EUR for Northern Ireland (or a yearly 2% of GDP) and for the Republic of Ireland almost 20 billion EUR (or a yearly 1.6% of GDP). Note, however, as explained before, that we refrain from modelling the negative

Brexit GDP effects on GB itself. These negative effects would have an additional negative effect on the exports of NI. The estimate of -2% 2% yearly is in line w with ith the study by Ottaviano, Pessoa, & Sampson (2014 2014 2014)) which projects negative welfare effects due to an increase in tariffs and NTBs for Great Britain at around 1%. Figure 2-1 1:: Changes in GDP for Northern Ireland and the Republic of Ireland due to tariff changes after Hard Brexit

2.1.2 Scenario “Brexit_IM”: Less Immigration With respect to the changes in the labor force, scenario “Brexit_IM “Brexit_IM”, ”, we assume a reduction in the labour force of 0.85% based on the study done by the OECD (2016). According to neo neoclassic classical theory, reductions in immigration lead to a reduction in eco economic nomic growth in the short term. Please n note, ote, however, that this is based on the model assumption of homogenous labor and is dependent up upon on the relative skill skill--level level of the migrants and the existing labor force. If the level of skill of the migrants is higher, the reduction in immigration will be negative. Another important issue relates to the possible reverse displacement of lo local cal workers: locals will now be hired instead of immigrants with positive effects such the reduction in spend upon social benefits for unemployed workers due to a lower unemployment rate. Studies have shown (for example, Dustmann et al. (2005)), that immig immigration ration has almost no impact on the unemployment rate except a small negative impact on the unemployment rate of intermediate intermediately skilled labor labor. Taking this into account and in the absence of actual data regarding the skill level of sectoral labor and of the immigrants, we assumed homogenous labor labor.. We further assumed that 25

unemployment is fixed fixed.. The negative impact shown in Figure 2 2-2 is less than 0.1% and the therefore refore almost negligible negligible. Figure 2-2 2:: Changes in GDP for Northern Ireland and the Republic of Ireland due to immigration changes after Hard Brexit

2.1.3 Scenario “Brexit_FD”: Reduction in Foreign Direct Investment There is only limited d direct irect evidence showing the effect on foreign direct investment (FDI) of entering or leaving the EU. A review of studies undertaken can be accessed in Whyman & Petrecu (2017). Most studies show a negative impact on FDI ranging from -18 18 to – 45% in the case of a Hard Brexit. According to Dhingra, Ottaviano, Sampson, & Van Reenen (2017), FDI will decrease by 22% in the UK. Haskel, Pereira, & Slaughter (2007) found that a change of 10% in FDI would lead to a reduction of productivity of 0.5%. We, therefore, assume a reduction of productivity for Northern Ireland of 1% over a period of 8 years years.. The overall impact on GDP of less FDI, as shown in Figure 2-3 3,, is negative and would add up in the years 2021 2021-2025 2025 to less than 500 million Euro.

Figure 2-3 3:: Changes for Northern Ireland due to FDI changes after Hard Brexit

2.1.4 Scenario â&#x20AC;&#x153;Brexit_XRâ&#x20AC;?: Devaluation of the GBP. We assume a 10% devaluation of the British pound. The exchange rate for exports from Northern Ireland to all other regions drops, which will decrease the price for exports. This will have a positive impact on production. Import prices increase due to the devaluation, which will also benefit the domestic production in Northern Ireland but will have a negative impact on the overall price level. However, as already stated, exports an and d imports to the UK are more than 50% of total trade, the positive impact will only come from the non non-UK UK imports and export. In the Republic of Ireland, the effects are negative as the devaluation of the GBP increases the price of the exports and reduces the price of imports.

Figure 2-4 4:: Changes in GDP for Northern Ireland and the Republic of Ireland due to devaluation of the pound after Hard Brexit

2.1.5 Scenario “Brexit” Scenario “Brexit” is the combination of the decomposition scenarios whereby the changes in tariffs and NTBs, the change in FDI, the reduction of immigration, and the devaluation of the pound are included included.. Figure 2-6 6 shows the overall effect on GDP. As we have seen, the main effects of Brexit are caused by the change in tariffs and NTB. For RoI, the negative effect on GDP is mainly caused by the devaluation of the GBP resulting in exports becoming more expensive and imports reducing in price price.. These results are in line with other studies undertaken on Brexit as the majority predict a negative impact on GDP for the UK (see Figure 1-1 1). ). The overall negative effect amounts to a loss of over 6 billion EU in the period 2021 2021--2025 2025 (or a yearly loss of over 2% of GDP).

Figure 2-5 5:: Overall changes for Northern Ireland and the Republic of Ireland after Hard Brexit

2.1.6 Scenario “Brexit_SM” Although the results of the negotiations of the UK with the EU are not yet concluded concluded,, the draft Brexit treaty by the EU Commission states that leaving Northern Ireland in the Customs Union and securing a close alignment with the Single Market could resolve the problem of the inner border in Ireland. Following this logic, we ran a third si simulation mulation in which Northern Ireland is still part of the SM and the CU. For this scenario, “Brexit_SM “Brexit_SM”, we assumed not only the devaluation of the pound but also that the changes in tariffs and NTBs are now also applied to trade between Northern Ireland and Great Britain. As Figure 2-6 6 shows, this scenario would have a negative effect on Northern Ireland. However, if Northern Ireland remains in the Single Market, this could lead to an increase in FDI instead of a reduction as Northern Ireland and the Republic of Ireland are now of particular interest as an entry to Great Britain. It is impossible to quantify the effect of increased FDI in this case. We saw, however, that the overall effect on GDP was small, even for a rather high reduction in FDI, such as in the case of Hard Brexit.

Figure 2-6 6: Overall changes for Northern Ireland and the Republic of Ireland when Northern Ireland remains in the SM

2.2

Imports and Exports

2.2.1 Overall Effects on Imports and Exports In the previous chapter, we saw that for NI the trade, as well as the devaluation effects, are the main drivers of GDP GDP--effects. effects. This also holds for the effects on exports and imports. Figure 2--7 shows the effects on the exports for the scenarios. The adoption of the WTO trade and higher NTBs leads to a reduction of exports (Brexit_TR). The devaluation of the British pounds leads to a clear increase of exports (Brexit_XR). The effects of these scenarios almost cancel each other out, and the overall effects on exports are almost zero. Imports ((Figure Figure 2--7 on the right) are reduced in all scenarios. Concretely, the price of imports increases and leads to a reduction of imports of 3% in the scenario “Brexit_TR” and 0.5% in the scenario “Brexit_X “Brexit_XR”. ”. In the other scenarios, there is a small negative effect on the imports (less than 1%). The overall effect is a reduction of imports between 3 and 5%.

Figure 2-7 7:: Effects on exports and imports in Northern Ireland

2.2.2 Regional Imports and Exports Figure 2-8 8:: Changes in regional exports and imports

For Northern Ireland, exports to the EU are clearly reduced (see Figure 2-8 8). ). As a reaction, exports to RoI and the ROW increase slightly. In the case of the imports, the devaluation and adoption of WTO WTO-tariffs tariffs both lead to a clear reduction in imports, especially from the EU.

Hard Brexit versus SM-Brexit and Unification In our previous study ( KLC Consulting, 2015), we simulated the economic unification of NI and RoI. In our combined scenarios, NI, as a part of an economically unified Ireland, automatically becomes a member of the Eurozone. From a political-legal perspective, this situation would be nearly identical to the German unification when former Eastern Germany first moved to the Deutschmark regime and then â&#x20AC;&#x201C; even before political unification was eventually ratified â&#x20AC;&#x201C; to the newly established Eurozone. Obviously, such a policy change would not add monetary sovereignty to an economically unified Ireland, as membership in the European Monetary Union (EMU) is by definition a yielding of sovereignty in money affairs to the ECB. However, the longstanding differential between the GBP and the Euro, and the international monetary trends that saw the Euro devaluing relative the Pound, would result in a rapid devaluation in currency. At the prevailing exchange rates of the time, this would have devalued the currency for the NI, causing a shift in international terms of trade that would favor NI relative to the UK. This devaluation would initially lead to a per-capita gross domestic increase of 5%. Brexit changed this picture. Rather than giving an upward push to the British Pound, investors took a pessimistic stance that resulted in a depreciation of the British Pound and led to a dampening of the growth dynamic of the UK. Post-Brexit-referendum, NI would now move from a relatively weak currency zone (BP) to a relatively strong currency zone (Eurozone). We further assumed that in the long run, unification would involve the adoption of the Irish tax system, greater openness in NI to FDI, and diminished trade barriers between NI, RoI, and other countries. A period of economic catch-up would likely shift the NI economy from low-valueadded to high-value-added industries. These changes would lead to an increase of per capita GDP in the long run of 4 to 7.5 %. For the new study, we continue to assume an FDI-surge for NI. The simulations for the new study were done with the same NIROI-model, with the difference that we used the latest input-output tables for NI and RoI. The original study on unification was done with a preliminary input-output table (IOT) lacking important data that had to be gathered by the research team. An official IOT was released during the work on this study. With the new data, we recalculated the unification scenario to be able to compare the unification with the Brexit scenario.

Comparing the new results of the Unification scenario with the results of the previous st study udy (see Figure 3--1),, we found that the overall effect remains positive but is slightly reduced reduced. One reason for this is the adjustment of the assumption on the effect of government harmonization: The assumption of a considerable lower efficiency and quality of the NI government expenditure was criticised after the release of the previous study, especially with regard to the health system in Northern Ireland. We We,, therefore, tempered the assumptions in the recalculation of the unification scenario. The second factor is the overall weakening of economic growth iin n the UK due to the Brexit vote vote. A complete comparison of the model results can be found in the Appendix. The positi positive ve effects are 2.75% compared to 5% in the earlier study study.. This can be explained by the depreciation of the British pound which will have a negative effect on the economy of NI. Figure 3-1 1:: Impact of the Unification on GDP for Northern Ireland and the Republic of Ireland

If we compare the effects of Unification with those of Hard Brexit, we see that the Unification scenario is in every year better than the Brexit scenario. The difference, starting at 2021 is around 3 billion billion. As already discussed, the devaluation in 2021 in the case of Brexit would also have a negative impact on Northern Ireland if it were to unify with the Republic of Ireland. If Northern Ireland remains in the Single Market, there iiss almost no effect on GDP.

Figure 3-2 2: Overall effects on GDP of Brexit and Unification

Conclusions Our modeling exercise aimed to calculate the costs of non-unification of the RoI and NI by estimating the economic effects of a hard Brexit and calculating the economic costs and benefits of a unification scenario. Both calculations were complrted utilising our NIROI model that allows for the derivation of quantitative effects on industry as well as at the macroeconomic level. Our model is well tested, and the fact that our new calculation of the unification scenario is in the range of our previous modeling, despite the fact that the data input has been critically changed, speaks for the quality of the model. When it comes to empirical data, in particular for NI, each empirical modeling is confronted with shortcomings, either because data are not getting published or because they are not getting collected in the first place. Our way out of the situation is well explained in the Appendix. It needs to be stressed that we did not include into our modeling exercise effects that stem from direct and indirect reductions of financial and intellectual resources, like research cooperation within the EU-space, due to leaving the EU. In this regard, our results are on the conservative side. The modeling exercise focuses on the economic implications and effects of a hard Brexit. Nobody knows at this point how the negotiations between the UK and the EU will unfold and what the outcome will be. On face value, the speech of PM Theresa May from January 2017 where she confirmed that the UK will leave the Common Market as well as the Customs Union, is the guide for our decision to focus on a hard Brexit. We then decided to also calculate a scenario where N stays in the SM and the CU, and the rest of the UK follows through with a hard Brexit. Up to the present, the realities of Brexit negotiations suggest that one of both options may become a reality. Unlike other studies, we then make use of our NIROI-model and calculate the economic effects of theunification of NI and RoI. The outcomes of our quantitative NIROI model are straightforward: (1) Unification is the superior path; (2) a hard Brexit comes with high and potentially severe costs for NI; and (3) the costs, thus, of non-unification are extraordinarily high. Political actors have to make choices. Our modeling provides insights of the benefits and costs of particular choices. The results are straightforward: Policy choices other than unification come with high avoidance costs. Economic reasoning can differ from political reasoning. Brexit is creating enormous political problems, for all sides, but in particular for the UK and NI as well as the RoI. A hard Brexit implies 35

a hard border between the North and the South of the island and thus significant welfare losses as well as a threat for the Good Friday Agreement that was instrumental for the vast amount of transactions alongside the currently invisible border. Hard Brexit has the potential to come with a cascade of negative economic impacts that may undermine the stability of the UK and the North and South of Ireland. Our study does not explore these potential developments, not least because it would require quite some speculation about future developments. In contrast, we keep the study on a sound methodological level by considering only the economics of Brexit. Over the period 2018 to 2025, the per capita loss in NI due to a hard Brexit adds up to an amount of Euro 5000. This amount is not negligible, not at least because it is the result of a lower growth path of NI. The per capita costs of non-unification due to an inferior policy choice are even higher, with a staggering amount of Euro 15000 over the modeling period. The option that NI stays in the SM and the CU is definitely superior to a hard Brexit but by far not as beneficial as the unification scenario. It is a second best in economic terms. In regulatory terms, such an option may create a difficult situation as it would require that NI would have not only to follow EU rules, laws, and regulatory practices but it would also have to adhere to UK laws that may differ from EU practices after Brexit. Running two regulatory systems simultaneously may come with extra economic costs which we did not include in the modeling exercise.

References Agriculture and Horticulture Development Board (AHDB). (2017, May 11). UK, GB and NI Farmgate Prices. Retrieved from AHDB: https://dairy.ahdb.org.uk/market-information/milk-pricescontracts/farmgate-prices/uk,-gb-and-ni-farmgate-prices/ Ahern, B. (2016, December 7). Britain And Ireland: Keeping Common Trade And Travel Areas. Retrieved from Social Europe: https://www.socialeurope.eu/2016/12/britain-and-irelandkeeping-common-trade-and-travel-areas/ Barrett et al. (2015). Scoping the Possible Economic Implications of Brexit on Ireland. Dublin: The Economic and Social Research Institute, No. 48. BBC News. (2013, October 10). UK immigration acts through the ages. Retrieved from UK Politics: http://www.bbc.com/news/uk-politics-24463873 Bennett and Kirkup. (2017, March 10). How much money does Britain currently pay the EU? Retrieved from telegraph UK: http://www.telegraph.co.uk/news/0/how-much-do-wespend-on-the-eu-and-what-else-could-it-pay-for/ Bergin et al. (2016). Modelling the Medium to Long Term Potential Macroeconomic Impact of Brexit on Ireland. Dublin: Economic and Social Research Institute, Working Paper No. 548. Bergin et al. (2017). COSMO: A new COre Structural MOdel for Ireland. Dublin: Economic and Social Research Institute, Working Paper No. 553. Biondi and Goncalves Raposo. (2016, December 22). The impact of Brexit on Northern Ireland: a first look. Retrieved from Bruegel: http://bruegel.org/2016/12/the-impact-of-brexit-on-northernireland-a-first-look/ Blonigen and Piger. (2011). Determinants of foreign direct investment. Cambridge: NBER Working Paper Series, Working Paper 16704.

Böcking, D. (2016, Mai 30). Iren müssen Trump und Brexit fürchten. Retrieved from Spiegel Online: http://www.spiegel.de/wirtschaft/unternehmen/brexit-und-donald-trump-iren-muessenneue-steuerparadiese-fuerchten-a-1149731.html Boland, V. (2017, March 31). Brexit brings Irish reunification back into the spotlight. Retrieved from Financial Times: https://www.ft.com/content/7a48e040-0d67-11e7-b030-768954394623 Bord Bia. (2014). Factsheet on the Irish Agriculture and Food & Drink Sector. Retrieved from Bord Bia, Irish Food Board: http://www.bordbia.ie/industry/buyers/industryinfo/agri/pages/default.aspx Benjamin Born & Gernot J. Müller & Moritz Schularick & Petr Sedlacek, 2017.The Economic Consequences of the Brexit Vote, https://ideas.repec.org/s/cfm/wpaper.html">Discussion Papers Bradley, J. (2015). Brexit, Northern Ireland and the island Economy: An Update. Dublin: The Institute of International and European Affairs. Brennan, D. P. (2016). Brexit - Energy and Carbon Implications. Public Policy Advisors network. Budd, L. (2015). The Consequences for the Northern Ireland Economy from a United Kingdom exit from the European Union. Northern Ireland Assembly, 2/15. Cechini Report (1988). The European Challenge, 1988. Gower Centre for Democracy and Peace Building, The UK in a Changing Europe, Queen´s University Belfast. (2015). EU Debate NI: To Remain or Leave? Northern Ireland and the EU Referendum. Deutsche Bundesbank. (n.d.). Begriffe und Definitionen, Finanz- und Währungsstabilität. Retrieved from https://www.bundesbank.de/Navigation/DE/Aufgaben/Finanz_und_Waehrungssystem/Stab ilitaet/Begriffe/begriffe_und_definitionen.html Devenport, M. (2016, August 10). Foster and McGuinness in Brexit talks call. Retrieved from BBC News Northern Ireland: http://www.bbc.com/news/uk-northern-ireland-37039683

Dhingra and Sampson. (2017). Brexit and the UK Economy. The London School of Economics and Political Science (LSE), Paper EA040. Dhingra, S., Ottaviano, G., Sampson, T., & Van Reenen, J. (2016). The impact of Brexit on foreign investment in the UK. Tech. rep., Center of Economic Analysis (CEP). Doherty et al. (2017). Northern Ireland and Brexit: the European Economic Area option. Brussels: European Policy Centre Discussion Paper. Donaghy, P. (2016, October 2). Hard Brexit, trade, and the border. Retrieved from sluggerotoole: http://sluggerotoole.com/2016/10/02/hard-brexit-trade-and-the-border/ Doyle, D. (2017, April 6). Trouble is brewing for Guinness after Brexit. Retrieved from Bloomberg: https://www.bloomberg.com/news/articles/2017-04-07/guinness-exposes-debate-over-ahard-border-with-ireland-after-brexit Duncan, D. (2017, March 30). Farmers in Northern Ireland know Brexit will be bad for them - they just donÂ´t know how bad. Retrieved from Public Radio International (PRI): https://www.pri.org/stories/2017-03-30/farmers-northern-ireland-know-brexit-will-be-badthem-they-just-dont-know-how-bad Dunt, I. (2016). BREXIT - What the hell happens now? Canbury Press. ECB. (2003). What is behind discrepancies between growth in GDP and gross value added? Monthly Bulletin, ECB. Eurolink-EU. (2016). EU Funding in Northern Ireland. Retrieved from http://www.eurolink-eu.net/eufunding-2/ European Commission. (2016). European co-funded programmes in Northern Ireland 2007-2013 and 2014-2020. European Commission. European Parliament. (n.d.). EU funding; Management of funds. Retrieved from European Union: https://europa.eu/european-union/about-eu/funding-grants_en

European Parliament. (n.d.). Northern Ireland PEACE programme. Retrieved from European Parliament: http://www.europarl.europa.eu/atyourservice/en/displayFtu.html?ftuId=FTU_5.1.9.html Finance Northern Ireland Government. (n.d.). European Structural and Investment Fund Programmes in Northern Ireland. Retrieved from Department of Finance: https://www.financeni.gov.uk/articles/european-structural-and-investment-fund-programmes-northernireland#toc-2 Finance Northern Ireland Government. (n.d.). European Structural and Investment Fund Programmes in Northern Ireland. Retrieved from Department of Finance: https://www.financeni.gov.uk/topics Finance Northern Ireland Government. (n.d.). European Territorial Cooperation (ETC) Programmes. Retrieved from Department of Finance: https://www.finance-ni.gov.uk/articles/europeanstructural-and-investment-fund-programmes-northern-ireland#toc-2 Gallagher, J. (2016, October 25). Brexit has reopened the Irish question. Retrieved from financial times: https://www.ft.com/content/79b9432e-9a0b-11e6-8f9b-70e3cabccfae Gasoriek, M., Ilona Serwicka, Alasdair Smith , 2018. Which manufacturing sectoes are most vulnerable to Brexit? Briefing Paper 16, UK TRADE POLICY OBSERVATORY Giles and Cadman. (2016, January 3). Economists´ forecasts: Brexit would damage growth. Retrieved from financial times: https://www.ft.com/content/1a86ab36-afbe-11e5-b9551a1d298b6250 Great Britain: Parliament: House of Commons: Northern Ireland Affairs Committee. (2011). Air passenger duty: implications for Northern Ireland, second report of session 2010-12, report. London: The Stationery Office Limited. Haefliger, M. M. (2016, Dezember 23). Vom Brexit überfahren, Irland die Beziehungen zu Grossbritannien. Retrieved from Neue Zürcher Zeitung: https://www.nzz.ch/wirtschaft/irland-und-der-brexit-gewinner-und-verlierer-ld.136421

Haskel, J. E., Pereira, S. C., & Slaughter, M. J. (2007). Does Inward Foreign Direct Investment Boost the Productivity of Domestic Firms? The Review of Economics and Statistics, 89, 482-496. HM Government. (2016). HM Treasury analysis: the long-term economic impact of EU membership and the alternatives. London: Her Majesty´s . HM Treasury. (2016). HM Treasury analysis: the immediate economic impact of leaving the EU. London: HM Treasury. House of Commons, Northern Ireland Affairs Committee. (2016). Northern Ireland and the EU referendum, First Report of Session 2016-17. House of Commons. House of Commons, Northern Ireland Affairs Committee. (2016). Northern Ireland and the EU referendum, First Report of Session 2016-17. House of Commons. House of Commons. 2018 . EU Exit Analysis. Cross Whitehall Briefing, January 2018 House of Lords - European Union Committee. (2016). Brexit: UK-Irish relations. London: House of Lords, HL Paper 76. House of Lords. (2016). Brexit: UK-Irish Relations. Selected Committee on the European Union. Hüttl and Martí Romero. (2016, 5 3). Northern Ireland and EU funds. Retrieved from Bruegel: http://bruegel.org/2016/05/northern-ireland-and-eu-funds/ IMF (2018). World Economic Outlook, April 2018. Cyclical Upswing, Structural Change Irish Congress of Trade Unions. (2016). The case against Brexit. Dublin: Trade Union Labour. Kauders, D. (2016). Understanding Brexit options. Sparkling Books. KLC Consulting. (2015). Modeling Irish Unification. KRB. Inc. (Knights of the Red Branch. Inc.)

KPMG. (2016). Brexit: A Northern Ireland Perspective. KPMG.

Kula, A. (2016, February 27). Where do EU grants to Northern Ireland go? Retrieved from News Letter UK: http://www.newsletter.co.uk/news/business/where-do-eu-grants-to-northern-irelandgo-1-7233873 Lofgren, H., & Robinson, S. (2002). A Standard Computational General Equilibrium (CGE) Model in GAMS. Tech. rep., International Food and Policy Research Institute (IFPRI). Flynn, M. . (2016). The Economic Implications of BREXIT for Northern Ireland. Dublin: Nevin Economic Research Institute (NERI) Working Paper Series, No. 35. Makki, S. S. (2006). Impact of Foreign Direct Investment and Trade on Economic Growth. The World Bank. Mars et al. . (2017). The Common Travel Area: Prospects After Brexit. Durham: Durham University. Mars et al. (2016). Brexit, Northern Ireland and Ireland. Durham: Durham University. Morgenroth, E. (2016). Economic Consequences of Brexit for Northern Ireland. Dublin: Economic and Social Research Institute (ESRI). Nahuis, R. (2004). One size fits all? Accession to the internal market; an industry-level assessment of EU enlargement. The Hague: Journal of Policy Modeling, 26 (2004), 571-586. Newman, H. (2017, March 27). To make the most of Brexit, Britain must leave the EUÂ´s customs union. Retrieved from telegraph UK: http://www.telegraph.co.uk/news/2017/03/27/makebrexit-britain-should-leave-eus-customs-union/ Northern Ireland Affairs Committee. (2016). Northern Ireland and the EU referendum. House of Commons Paper HC 48. Northern Ireland Centre for Economic Policy (NICEP). (2014). Outlook Autumn 2014. Coleraine: Ulster Business School. Northern Ireland Food and Drink Association. (2016). Brexit: Challenges & Opportunities for Northern Ireland Food & Drink. Deloitte.

Oberhofer,H. and Pfaffermayr, M. 2018. Estimating the Trade and Welfare Effects of Brexit: A Panel Data Structural Gravity Moidel, CESifo Working Paper Series No. 6828 CESifo Working Paper Series No. 6828 O´Carroll and McDonald. (2016, October 18). Brexit to blast huge hole in Northern Ireland´s fragile finances. Retrieved from the Guardian: https://www.theguardian.com/world/2016/oct/18/brexit-to-blast-huge-hole-in-northernirelands-fragile-finances OECD. (2016). The economic consequences of Brexit: A taxing decision. Tech. rep. Ottaviano, G., Pessoa, J. P., & Sampson, T. (2014). The Costs and Benefits of Leaving the EU. Tech. rep. Oxford Economics. (2016). The Economic Implications of a UK Exit from the EU for Northern Ireland. Oxford: Oxford Economics. Phinnemore and McGowan. (2016). After the referendum: Establishing the Best Outcome for Northern Ireland. Belfast: Centre for Democracy and Peace Building. PricewaterhouseCoopers. (2016). Brexit: The implications for Irish business. PwC. PricewaterhouseCoopers LLP. (2016). Brexit: The implications for Irish business. PwC. PricewaterhouseCoopers LLP. (2016). UK Economic Outlook. PwC. Queen´s University Belfast. (2017). Event Report: Brexit and Citizen's Rights in Northern Ireland. Belfast. RalSe. (2016). EU Referendum Update. Northern Ireland Assembly, Research, and Information Service Research Paper, Paper 46/16. Riddle Hall, Queen´s University Belfast. (2017). Event Report: Brexit and Citizen´s Rights in Northern Ireland. European Citizen Action Service.

Sapir, A. (2016, October 21). Beyond hard, soft and no Brexit. Retrieved from Bruegel: http://bruegel.org/2016/10/beyond-hard-soft-and-no-brexit/ Shiller, R.J. (20127). Narrative Economic s. American Economic Review Vol. 107, No. 4, pp. 967-1004 Sinn FĂŠin. (2016). The case for the North to achieve designated special status within the EU. Dublin and Belfast: Sinn FĂŠin. Stennett, A. (2016). The EU referendum and potential implications for Northern Ireland. Belfast: Northern Ireland Assembly, Research, and Information Service Research Paper, Paper 21/16. Taylor, C. (2016, October 16). The harder the Brexit, the worse for Ireland. Retrieved from the Guardian: https://www.theguardian.com/world/2016/oct/16/harder-brexit-worse-forireland-sterling-fall TF50. (2018). Draft Agreement on the withdrawal of the United Kingdom of Great Britain and Northern Ireland from the European Union and the European Atomic Energy Community highlighting the progress made (coloured version) in the negotiation round with the UK of 16-19 March 2018. https://ec.europa.eu/commission/sites/betapolitical/files/draft_agreement_coloured.pdf The Centre for Cross Border Studies. (2016). Brexit and UK-Irish Relations. The Centre for Cross Border Studies. The Centre for Cross Border Studies. (2016). Submission to the Joint Committee on Jobs, Enterprise and Innovation: Economic Implications of the UK Referendum Decision. The Centre for Cross Border Studies. (2016). The UK Referendum on Membership of the EU: CrossBorder Cooperation, Peace-Building and Regional Development. Co-Operation Ireland, EU Referendum Briefing Papers, Briefing Paper 3. The Danish Parliament. (n.d.). Which countries are net contributors to the EU budget? Retrieved from The Danish Parliament, EU INformation Centre: http://english.eu.dk/en/faq/faq/net_contribution The Joint Committee. (2016). Brexit and the future of Ireland. The Joint Committee.

The Joint Committee. (2016). Brexit and the future of Ireland. Tonge. (2017). The Impact and Consequences of Brexit for Northern Irelandnd. European Parliament. UK think tank Demos. (2017). Wales and the manufacturing industry most at risk from a "hard Brexit". UK think tank Demos. UK Trade and Investment. (2015). UKTI Inward Investment Report 2014 to 2015. London: UKTI. Whyman, P. B., & Petrecu, A. I. (2017). The Economics of Brexit. Palgrave Palgrave Macmillan.

Appendix: Sectoral Effects

5.1

Sectoral Effects on Trade Figure 5-1 1:: Changes in exports (left) and imports (right)

5.2

Effects on Domestic Production Figure 5-2 2:: Effects of Brexit on domestic production

Appendix: The Recalculated Unification Scenario

6.1

Assumptions for the Unification Scenarios One of the goals of this report was to see if the new database would significantly change the results from the previous report. We, therefore, implemented the same scenarios and recalculated the results. The description of the several decomposition scenarios is taken from the previous report. Table 6-1 shows all scenarios as well as the assumption. Note, that we did not make a decomposition scenario for each tax separately. Table 6-1: Assumptions of the Unification scenarios

Acronym

Description

Assumptions

Unif_Harm

Tax harmonization

harmonization of 1) activity tax rates, 2) commodity taxes, 3) import taxes and 4) institutional taxes

Unif_Govt

Reduction Government

0.5% annual reduction in NI government expenditures; increase in efficiency 0.25%/ an ROI

Unif_NTB

Reduction NTB costs

reductions of NTB costs at the rate of 5%/a

Unif_Prod

Adjustment productivity

Productivity slowly reaches the level of ROI

Unif_XR

Devaluation

the exchange rate between NI and ROI equal to 1, other exchange rates according to ROI; after 2021 devaluation of GBP by 10%

Unification

Overall unification

All above

6.2

Overall effects of unification Figure 6-1 1:: Impact of the Unification on GDP for Northern Ireland and the Republic of Ireland

6.3

Harmonization of taxes and tariffs Figure 6-2 2:: Impact of the harmonization of taxes (including tariffs) for Northern Ireland after the unification unification:: New versus Old Simulation

a. GDP effects with new data (harmonization of activity, product taxes and import tariffs

b. Previous results: Activity tax harmonization

c. Previous results: Product tax harmonization

d. Previous results: Import tariffs harmonization

Activity taxes: Using the old data, applying the activity taxes from RoI in NI caused a small increase in output because the RoI activity tax structure was slightly more favorable. This resulted in a nominal GDP effect of 1.2 billion Euro. The new IOT tables show a (very) small decrease in output as the overall activity tax structure is slightly better in NI. This results in a total negative effect of around the same size. The simulated change from the NI commercial tax regime to the RoIâ&#x20AC;&#x2122;s commercial tax regime causes a 12.3% drop in aggregated commercial tax rates. The overall improvement to GDP in NI was estimated at 1.06% in the first year of unification, but the rate of annual gain de declines clines slightly to 0.97% by 2025. Though the numbers appear small, the effect of around 449.3 million Euro per year accumulates to 3.7 billion Euro by the end of 2025. Harmonization of NI import tax rates with RoI rates in the previous report: The accumul accumulated ated increase in NIâ&#x20AC;&#x2122;s GDP reached 993.2 million Euro in 2025. For the new IOT of NI, we used the tariffs from the old tables.

6.4

Optimized government Figure 6-3 3:: Impact of the reduced NI NI-government government size after the unification

The assumption of a considerable lower efficiency and quality of the NI government expenditure was crit critic icized ized after the release of the study, especially with regard to the health system in Northern Ireland. We We,, therefore, tempered the assumptions in the recalculation of the unification scenario. Old version version:: The harmonization of government functions lowers NI government expenditure. The multiplier effect effect,, which leverages fiscal expenditure into output, works in reverse and NI GDP is reduced by 296.4 million Euro in 2018 and 2.6 billion Euro in 2025. Across the first eight years of the policy, NI GDP is reduced by 11.2 billion Euro.

6.5

Reduction of NTBs Figure 6-4 4:: Impact of the NTB changes for Northern Ireland after the unification

Old version version:: The reduction to import transaction costs is modeled as a 5% annual reduction in distribution costs. While this reduces prices, it also, unfortunately, reduces distributor revenue revenue. Thus the overall effects on trade and output are negligible. In NI the net effect on trade is only 352.3 thousand Euro in 2018, this is associated with an improvement to GDP of 12.5 million Euro that grows to 102.8 million Euro of GDP by 2025. The accumulated effect on NIâ&#x20AC;&#x2122;s GDP reaches 457.3 million Euro by 2025. Assumptions Assumptions: NTBs from NI taken from ROI.

6.6

Productivity Figure 6-5 5:: Impact of the productivity changes for Northern Ireland after the unification

Old version version:: Gradual improvement in productivity because the convergence in returns to productivity inputs lifts GDP in NI by 120.5 million Euro in 2018, relative to the benchmark. The new post post-unification unification policies attract FDI and prompt movements along the production possibilities frontier, improving NI’s GDP every year. In 2025 the improvements have grown to 1.2 billio billion n Euro over the benchmark. GDP gains in NI accumulate to 4.9 billion Euro by 2025.

6.7

Euro Conversion At this point point,, NI and RoI belong to different currency zones and are thus subject to different monetary policy regimes. In our combined scenarios, NI, as a ppart art of an economically unified Ireland, automatically becomes a member of the Eurozone. From a political political--legal legal perspective, this situation would be nearly identical to the German unification when former Eastern Germany first moved to the Deutschmark regime and then – even before political unification was eventually ratified – to the newly established Eurozone. Obviously Obviously,, such a policy change would not add monetary sovereignty to an economically unified Ireland, as membership in the European Monetary Union ((EMU) EMU) is by definition a yielding of sovereignty in monetary affairs to the ECB. However, the long long--standing standing differential between the GBP and the Euro, and the current international monetary trends that see the Euro devaluing relative the Pound despite the pperiod eriod of low interest rates in the UK that will eventually come to an end, promises NI, under a unified

Ireland, a rapid devaluation in currency. Of any region in the UK, NI demonstrates the economic fall-out from an over-valued currency. In economic theory, when a currency cannot be devalued, either the labor market must adjust, or fiscal transfer must be adequately supplied and adequately used. It can be argued that NIâ&#x20AC;&#x2122;s labor market is no longer flexible enough to adjust, perhaps because of long-periods of under-employment that creates hysteresis effects. It can also be argued that the quantity of fiscal transfer is not optimal, and more importantly, that it is inadequately spent. From this view, unification promises a monetary policy fix, even if there is no change in sovereignty over local monetary policy. In addition to currency devaluation, there are other benefits to membership in a large currency union rather than a small one. For example, as the experience of the RoI during the financial crisis from 2008 demonstrates, the EMU has a history of providing liquidity in times of crisis. Also, a change in currency both increases and decreases transaction costs. Cultural ties and evolved integration in supply chain management suggest increases in these costs would be less likely to cause trade diversion between NI and GB. This limits downside risk and means, to the extent that the new economic unit integrates with other members of the Eurozone and those economies whose currencies are tied in some form to the Euro, the advantages of having the Euro as a common currency would be larger. Technically, exchange rates in the model are fixed and remain at 2009 rates. We assume no changeover costs incurred by Irish banking facilities, as these costs, though relevant in the changeover period, would be small one-time costs in the long run. Note, that in 2021 we model the devaluation of the GBP due to Brexit).

Figure 6-6 6:: Impact of the exchange rate changes for Northern Ireland after the unification

Old version version:: The Euro change change-over over effect on GDP is pushed by the trade creation/trade diversion story. We look at the over over-all all changes in country country--pair pair trade flows on page 55, and by bysector changes to trade for 2018 are shown in the appendix. The isolated effect of only the currency conversion boosts exports in NI by roughly 18.7% annually and imports by roughly 8.4% annually, or 9.4 billion Euro of NI trade creation, which accumul accumulates ates to 79.8 billion Euro by 2025. Currency change change--over over from the GBP to the Euro increases NI GDP by an average of approximately 2 billion Euro annually, which accumulates to a total of 15.8 billion Euro by 2025.

New version version:: The new results show a drop of the positive effects because of Great Britain leaving the UK in 2021.

Appendix: Modeling and Data Selection

7.1

The choice of CGE as modeling tool The NIROI model used in this study is a computable general equilibrium (CGE) model for Northern Ireland and the Republic of Ireland and was developed for the previous report. As already discussed in that report, CGE models, have several advantages but also pose some limitations. For the interpretation of the results, it is important to have these advantages and limitations in mind and therefore, we recap this discussion. CGE models are best suited to quantify the effects of a variety of policy changes on output, trade flows, changes in current accounts balances, changes in various aspects of government budgeting, and changes in factor supply and demand, subject to the quality of data inputted, scenario design and the selection of assumptions. Over the last twenty years, both economic theory and modelling techniques have gone through repeated cycles of rigorous testing, refinement and rigorous testing again, and thus it is no surprise CGE models have become a standard part of the economistâ&#x20AC;&#x2122;s toolbox, used to inform practitioners of all kinds about potential implications of particular policy actions. Our Irish unification model shares all the general advantages and restrictions found in contemporary state-of-the-art modeling procedures. In addition to this virtue, CGE offers surprising flexibility. For example, the static solution process was changed into a recursive dynamic process that strengthens its dynamic scope. This latter development, in conjunction with calibration procedures that capture relationships in the data, lends itself quite readily to situations where time-series data are unavailable. Further to the point, because the model is used to contrast some distinct scenarios with a benchmark, model shortcomings are present on both sides of the analysis excellently isolating effects on indicators from policy changes, ceteris paribus. Another advantage of CGE models are external economies of scale for developers. The rapid and continued rise in CGE use has left behind a large body of accumulated publicly available modeling resources that drastically lowers the cost of developing models for specific regions or policies. Though we have selected CGE as the optimal economy-wide modeling framework for this project, as with all economic models, CGE models also have their limitations. First, CGE models uses optimization mathematics to model the aggregate effects of agent level decisions subject 55

to macroeconomic constraints; but, in its static or recursive-dynamic form, the dynamic scope of that decision is limited because agent decision making is restricted to myopic expectations and excludes rational and adaptive expectations. This means though producers seek to minimize costs or maximize profits and consumers to minimize expenses or maximize utility, they do so without inter-temporal selections between consumption and production in current or future periods. Second, the functional forms chosen in applied CGE work are often of the constant returns to scale variety (constant elasticity of substitution functions).2 Thus productivity in the model occurs when returns to scale are constant, which does not account exactly for rapidly developing infant industries or those in decline. Third, while CGE is apt at simulating changes in prices and quantities of products, it does so in real terms so that there are no mechanisms for modeling changes in nominal variables that prompt real effects, like changes in money supply that can lead to real economic changes. This means there is no modeling of quantitative easing or its absence. Moreover, though there is room in the theoretical framework for changes in consumer preferences, they are not frequently applied because preference changes are not easily substantiated empirically. This means representative consumers with increasing income will not change the share of that income spent on a particular good. It also means changes in bysector output, because of price changes, are not met with commensurate changes in by-sector investment. Finally, CGE does not account in any way for non-economic political or social forces. Therefore, CGE results should be interpreted strictly in the economic sense. Yet, CGE modeling has become a workhorse for empirical studies. In light of this large body of limitations, the question naturally arises, why use CGE at all? The short answers: CGE works. More specifically, CGE works is best suited to quantifying the effects of a variety of policy changes on output, trade flows, changes in current accounts balances, changes in various aspects of government budgeting, and changes in factor supply and demand, subject to the quality of data inputted, scenario design and the selection of assumptions. Despite these disadvantages and in accord with the literature, a CGE model is a highly valuable and reliable tool for interpreting the economic effects of a variety of policy scenarios. The described limitations of the approach should, however, be kept in mind when studying the results of this report. 2

Rutherford and Peroni showed how to incorporate more realistic functional forms. This approach, probably due to its complex implication and high demand of data, is almost never used.

7.2

The NIROI Model The NIROI computable general equilibrium model used in this study is the same as the one used in our previous report. The subsequent text follows, therefore, the model description in that report closely. The NIROI model is based on the famous IFPRI model.3 A key feature of the standard model is its flexibility, which permits the analysis of country-specific aspects of economic structure and functioning. The basic model contains different rules for the treatment of relations between exchange rates and the current account, the treatment of the government deficit, the savings of the households, and the labor market. Depending on the scenario and the actual policy in a country, these rules can be adjusted accordingly. For example, the government deficit can be balanced by adjusting taxes, transfers, or government savings. To build the NIROI model, the IFPRI model was extended to a multi-regional case, comprising NI and the RoI. Exports, imports, and other current account components were differentiated according to their origin and destination from and to both countries, as well as from and to: -

Great Britain (GB);

the rest of the Eurozone, other than the RoI (REUZ);

the rest of the EU, other than the Eurozone, GB, and RoI (REU);

and the rest of the world (ROW). For this report, we combined REUZ and REU to one region (EU). This was done because a complete disaggregation for all transactions for the original regions was not available. The second important improvement over the basic IFPRI model is the temporal resolution in the NIROI model. Instead of a static framework, we use a recursive dynamic framework in which the model is solved forward for consecutive years from 2015 to 2025. This means that some of the exogenous variables are changed over time using transition equations. Capital stock, for example, is updated endogenously given previous investment and depreciation. The updated values are used for solving the next yearâ&#x20AC;&#x2122;s equilibrium. Other updated variables are the population, factor productivity, export and import prices, and transfers. A further important feature of the model is the transfers between the governments and the households. These 3

For a detailed description see Lofgren & Robinson, 2002.

transfers are not fixed but depend on the deficits and the population. This allows a more realistic modeling of actual policies in both countries. To our knowledge, the NIROI model is the first state state-of of-the the--art, art, data data-intensive intensive intensive, recursive dynamic model that has been applied to a unification of Northern and Southern Ireland scenario. No other model has been customized specifically to simulate the uniqu uniquee situation in NI where the local government deficit is not funded with debt issuances by the usual market mechanism, but by transfers within the British public finance system. Additionally, NIROI includes the linkage of such transfers in a way that turns it into an endogenous variable (rather than to an exogenous as in IFPRI). Also, in the NIROI model, the small small-country country assumption is applied between both both-island island regions and off off-island island regions while the large country assumption is applied between between-island island reg regions. ions. In other words, both regions are treated as small countries in regards to other entities; RoI is then treated as a large country in regards to NI. This customization accommodates more accurately for geographical proximity between island regions and rresulting esulting price sensitivities. NIROI follows the production and consumptions layouts given in the IFPRI model albeit with a few changes. At the top of the chain, producers maximize profits with a constant elasticity of substitution technology to arrive aatt activity output, disaggregated originally into 54 sectors. The new data for NI and RoI does not allow for a disaggregation of 54 sectors, and the number of sectors was reduced to 25 25. The production structure is given in Figure 77-2. Figure 7-1 1:: Production structure (source: Lofgren & Robinson, 2002, p. 99)

The elasticity substituting value for intermediate consumption is constant and near to 1, reflecting the relatively constant proportions of value value--added added and intermediate consumption typically found in empirical production analysis. The mix of value value--added added al also so follows a CES structure, enabling the substitution of labor for capital, though this substitution is also relatively inelastic. Just as in the IFPR model, and in the SAMs developed for model use, off off-diagonal diagonal or secondary output follows constant distributions over time according to fixed yields. Figure 7-2 2:: Flows of marketed commodities (source: Lofgren & Robinson, 2002, p. 12 12). ).

Marketed activity output, in prices and quantities ((PXAC PXAC and QXAC, in Figure 77--3), 3), are combined and distributed into exports and domestic sales (QE/PE and QD/PDS QD/PDS-PDD). PDD). Unlike in the supply of imports, exports are supplied without distributions costs, which are born by impor importers ters and third third--party party transportation providers. Domestic sales are imperfectly substituted with imports into a composite commodity (QQ/PQ) that is distributed to the domestic market. On the demand side, while the share of composite goods consumption is cons constant tant across the set of absorbing entities, it varies across industrial sectors according to price price-levels. levels. The supply of imported commodities is selected based on regional distributions subject to regional elasticities and regional distribution costs. While consumers optimize utility according to a Cobb Cobb--Douglas Douglas function, the government consumption path is exogenously determined. The government consumption product mix is not price dependent but fixed fixed-coefficient coefficient dependent. Exogenous final values of investment are linked neo neo-classically classically in a one to one proportion with endogenous savings levels, while by by-sector sector investment again follows a fixed fixed--coefficient coefficient distribution distribution.

Transfers in both data and code (not shown in the figure) are modeled slightly differently than in the IFPRI model. There are relatively large transfers from households to governments, especially in NI, consisting of remittances from home-based small business units, which are relatively large in the data. Government transfers to households and enterprises are standard, but there are no transfers between households and the ROW sectors. In the code, these transfers represent foreign domestic expenditures and domestic expenditures abroad, consequently bypassing institutional income but not private consumption. NIROIâ&#x20AC;&#x2122;s factor transfers paid abroad from island regions are endogenously tied to domestic factor demand and wages in a linear formulation. Similarly, factor transfers received from abroad are tied to factor supply, which in the case of labour grows with population. However, in the case of capital, factor supply accumulates recursively according to an internal rate of return. Domestic consumption abroad and foreign domestic consumption follow an exogenously determined growth path, along with external enterprise transfers and, in the RoI case, external government transfers. NI transfers from within the British Public Finance System are set equal to the short fall between government revenue and government expenditure, where government expenditure includes a portion of capital expenditure that in the model code is included in the savings/investment balance. This is a departure from the usual CGE practice of allowing the government deficit to crowd out private investment by negatively impacting the savings/investment balance. We apply it only to the NI case, because the NI deficit is completely funded with fiscal transfers from the UK, and our unification scenarios do not include the possibility of NI funding their own deficit. Therefore, despite the NI government deficit, NI government capital formulation continues at a modest but consistent rate, in-line with the historical results reported in the NFBR literature. There are several tax categories in the NIROI model that are solved for in the base year and remain exogenous throughout model simulations. Income taxes are deducted from institutional incomes before expenditure calculations and directed to the government revenue function. Consumption taxes and other levies constitute commodity taxes, which are inclusive in composite commodity prices but combine with other tax revenues and transfers to arrive at government income. Net activity taxes, which are added to employee compensation costs and gross profits to arrive at gross value added, and import taxes (tariffs), which combine with

import distributions costs to form part of import prices, are also components of government revenue. The NIROI savings and investment balance formulation is consistent with the IFPRI model in that it works in conjunction with the functions for the government balance and the current account balance to ensure the slack variable is equal to zero. However, a separate equation for enterprise savings has been introduced, and the endogeneity of the government savings variable is partitioned into exogeneity for the NI region (as discussed above) and endogeneity for the RoI region. The entire expenditure on net investment is funded from the savings/investment balance. Model closure ensures proper identification of model variables. In NIROIâ&#x20AC;&#x2122;s external balance the endogenous variable is foreign savings while the exchange rate is exogenously fixed at base year valuations until changes in the scenario are introduced. In the government balance, another component of model closure, government expenditure is exogenously determined, tax rates are fixed, and government savings, as mentioned above, are partitioned by region. In the savings/investment closure, both government and investment expenditure is fixed, while government and investment shares of absorption are left to adjust to ensure the model solves. In NIROIâ&#x20AC;&#x2122;s factor market closure, an economy-wide wage adjusts to ensure factor market demand and factor market supply are equated, while a sector-specific wage-distortion term remains constant. As mentioned above, the quantity of factors supplied is exogenously determined. NIROIâ&#x20AC;&#x2122;s labor supply follows a growth path commensurate with population growth rates taken from historically reported population data points. Its capital supply accumulates periodically by an interest rate derived at the end of each period. This means the capital supply allocated in each period is the product of a constant depreciation rate and the accumulation of capital in the prior period. The quantities of capital demanded are endogenously derived.

7.3

Data and Base Year CGE models are characterized by a calibration process that enables comparative scenario simulation without extensive time-series data. The model is parameterized to a base-year data set, and this parameter structure is maintained and altered subject to scenario design. If there 61

are options regarding data availability, often the case at the national level where SUT data are published more frequently, a base year coinciding with a neutral phase of the business cycle should be selected. For the previous report, data compilation began in the third quarter of 2013, while the 2010 SUT for the RoI was not published until the first quarter of 2014. Our base year selections were then limited to 2009 and 2005, from which we selected the more recent year 2009. For this new study, we could use the supply and use tables for the year 2013 issued by the Central Statistics Office (RoI) and NISRA (NI).4 One problem with the new SUT of both countries is that the off-diagonal elements are missing. These elements indicate that a certain sector, for example, the chemical sector, not only produces chemical products but also products that are the main products of other sectors. For small countries, it is often easy to infer the firms that produce these off-diagonal products. For this reason, the information is often not published. To construct a complete SUT for both countries, we used a least-squares estimation approach with high weights for the off-diagonal elements.

The official tables can be downloaded for NI and RoI can be downloaded here: https://www.gov.uk/government/statistics/northern-ireland-supply-use-tables-2013-experimental-results, respectively here: http://www.cso.ie/px/pxeirestat/database/eirestat/.