European agriculture after Brexit: Does anyone benefit from the divorce?

UK’s exit (Brexit) from the EU would entail disturbances in agri-food markets. This study analyzed three different Brexit scenarios with increasing barriers to trade (EEA+, FTA, WTO), employing the EU focused global agricultural sector model CAPRI. In the UK, food prices will increase, making consumers in the UK the biggest losers. However, provided trade costs are kept low, termination of the contribution to the EU Common Agricultural Policy (CAP) and the gains to producers from higher food prices could offset the losses to consumers. In the EU, declining food prices would benefit consumers but reduce farmers’ incomes.


Introduction
The UK invoked Article 50 of the Treaty on European Union after the national referendum held in June 2016 and is due to withdraw from its EU membership (Brexit) in March 2019. Depending on the outcome of current negotiations on a Withdrawal Agreement and future trade relationship, the UK might lose the economic benefits of access to both the EU customs union and the single market. Thus, in a worst case scenario, trade in goods between the UK and the EU will be subject to tariffs and other border measures. Both parties would be exposed to potentially significant impacts on agricultural markets.
Trade policy changes are key factors in determining the economic consequences of Brexit for agricultural markets in Europe. Various outcomes are possible depending on the negotiation results.
One option is that the UK pursues a freest possible economic relationship with EU27 as well as regulatory alignment (Soft Brexit) so that it maintains frictionless trade as far as possible after Brexit (HM Government, 2017). However, UK's red lines in Brexit negotiations, e.g. no free movement of labor, independent trade policy, no EU budget contribution, and independence from the European Court of Justice, mean that frictionless-trade cannot be replicated (Gasiorek et al., 2016) (See Table 1).
One of the options for a soft Brexit would be for the UK to agree to stay in the European Economic Area (EEA), which would minimize trade friction after Brexit as well as eliminate tariffs in agriculture (so-called EEA+, where the '+' indicates that agricultural tariffs are also removed unlike in the EEA itself). An alternative soft Brexit option is a free trade agreement (FTA) between the UK and EU27 but without regulatory alignment. It means that the UK leaves the single market, but makes a FTA for goods. This option inevitably entails higher trade facilitation costs than in the EEA+ case because Non-Tariff Barriers (NTBs) in trade increase. Both in EEA+ and FTA, Rules of Origin (RoO) will increase trade costs as NTBs after Brexit, but the impacts of food standards such as Sanitary and Phytosanitary measures (SPS) and Technical Barriers to Trade (TBT) could differ depending on the scope of an FTA agreement between the UK and the EU. Additionally, NTBs in FTA will be affected by transport delays due to border inspections as well as the need to show compliance with the regulatory standards of the other party.
If neither of those trade deals between the parties is achieved, the UK and the EU would face the Most-Favored-Nation (MFN) tariff rates of the other party, as for any WTO member without a preferential agreement (Hard Brexit). Consequently, those increases in tariffs and NTBs would lead to trade destruction between the UK and the EU and some trade diversion to third countries that would then have relatively more favorable access to both markets. Furthermore, Brexit will impact on government budgets in both the UK and the remaining EU members.
The UK is the second largest net contributor to the EU budget (€9.8bn in 2016, House of Commons (2018b)). The UK treasury will benefit financially from Brexit, but the EU will suffer from revenue loss and it will indirectly affect agricultural budget spending in CAP. In this study, we assess the economic impacts of Brexit, focusing on agricultural markets of both the UK and EU27, as well as changes in UK's CAP budget contribution. We employ the CAPRI (Common Agricultural Policy Regionalized Impact Modelling System) for the analysis in a comparative static manner (Britz et al., 2014). CAPRI lets us compute Brexit impacts on agricultural supply, prices, and welfare. With the scenario analysis, we aim to identify potentially vulnerable sectors and regions, as well as potential winners of Brexit, taking into account interactions with world markets.
6  (Egger et al., 2015) 1) This study used results from gravity trade model in Hufbauer et al. (2009) and Barrett et al. (2015) 2. CAPRI for Brexit assessments CAPRI is a partial equilibrium (PE) agricultural sector model with a focus on EU agricultural markets and regional agricultural supply. It also considers EU agricultural policy and border policies with the main global trading regions (Britz et al., 2014). The supply module is iteratively connected to a global market model for agri-food products, depicting bilateral trade flows between the EU-27, UK and third countries, with a detailed set of trade policies at the border (ad valorem and specific tariffs, TRQs, etc.). The bilateral trade model is based on the Armington assumption (Armington, 1969). It depicts trade preferences among domestic and imported products from different origins, considering price differences.
The CAPRI simulates bilateral trade flows between selected regions and countries globally, which in particular includes the regions "EU-West", "EU-East" 3 and the UK. Within the two EU regions, there is supply on a sub-regional level (NUTS2 4 ), demand and processing on a national level, but no explicit modeling of trade between sub-regions or countries. Agri-food trade impacts will, therefore, affect the aggregate EU trade regions (EU-West and EU-East), but then price impacts transmit proportionally to all regions and farmers in the member countries. For global trade of agricultural goods, we report trade values at an aggregated regional level like North America ( 3. UK agri-food trade with the EU UK agriculture is highly integrated into the EU single market. Overall, 30% of the food consumed in the UK stems from the other 27 member states of the EU (Defra, 2018). In 2016, UK imported food, drinks, and feed worth £42bn and exported £20bn. EU27 accounts for 65% of the UK's imports and 60% of its exports of agri-food. The UK mainly imports meat (processed and unprocessed), dairy, fruits, and beverages from EU-27 and exports meat and dairy products to EU27. Concerning the main EU trading partners, the Netherlands, Ireland, Germany, and France account for about 60% of the UK's food imports from the EU27. Thus Brexit could imply serious disturbances for agri-food traders (and indirectly producers) not only in the UK but also in a number of EU member countries, with the biggest potential impacts in the meat, dairy and fruits and vegetable sectors. However, the CAPRI database covers only 55% of exports and 77% of imports in the UK's agri-food trade in terms of gross revenues compared to the UK's official statistics (Defra, 2018). Mainly, beverages and some processed food products are not covered in CAPRI compared to the UK's statistics.

Brexit scenarios
In order to systematically analyze the impacts of the multitude of changes that are likely to follow from Brexit, we develop scenarios around two main aspects: the outcome of Brexit negotiations and the post-Brexit UK's agricultural policy (See Table 5). For Brexit negotiations, we consider three negotiation scenarios: EEA+, FTA, and WTO. In all scenarios, we assume that the UK roll over all of the EU's preferential trade agreement after Brexit. TRQs are divided between the UK and EU27 based on domestic consumption levels of TRQ products.  The EEA+ scenario represents a case in which the UK remains in the single market and concludes a tariff-free FTA with the EU27, but gains sovereignty on trade policy. In addition to the current EEA, a tariff-free trade agreement is made for agricultural products. In this scenario, we assume that NTBs related to sanitary, phytosanitay and technical standards do not increase trade costs and only borderrelated NTB costs increase by 5%. The overall increase stems from paperwork of rules of origin (RoO) and additional costs from border controls due to inspection and delay (Abreu, 2013). In this scenario, we assume that the UK would continue to make a transfer into the EU budget equivalent to net CAP transfers less third country tariff revenues. 8 For the UK, tariff revenues in agri-food trade are not transferred to the EU.
The FTA scenario depicts a case in which the UK leaves the EU single market and only makes an FTA in goods including agricultural products with EU27. Agricultural products are traded with zero tariffs, but NTBs increase more than in the EEA+ scenario (7.9% for primary products and 12.7% for processed products). The NTB costs are derived from Egger et al. (2015), which estimate the potential trade cost saving of a deep FTA between the EU and United States. We assume that greater similarity in regulations between the UK and EU27 before Brexit and mutual recognition arrangements would reduce trade restrictiveness by 50% compared with the values presented by Egger et al. (2015). The UK's net CAP contribution is abolished and it would become a financial gain for the UK and loss for the EU.
In the WTO scenario, no trade deal is agreed in the negotiation and most favored nations (MFN) tariffs are charged on agricultural products according to WTO rules between the UK and EU27. We further assume that the UK will apply the same tariffs towards non-EU countries after Brexit as it does today as an EU Member State. NTB costs increase the largest (12.6% for primary products and 24.2% for processed products) among the considered scenarios. For NTB costs, we use the results of NTB reduction estimated by Egger et al. (2015) for the case of the EU single market, and again assume that only 50% of NTB costs materialize as trade barrier in the FTA scenario. Thus, we assume that NTB costs in WTO become larger that in FTA by assuming that a less favorable trade environment is formed between the UK and the EU, and that the UK's food standards and regulations will diverge from the EU to some extent.

Results
This section presents the main results of the Brexit negotiation outcome scenarios on the components of the market balances, prices and welfare.

Impacts on trade
Brexit has a much larger impact on relative trade patterns in the UK compared to the EU27. Imports and exports decrease in both the UK (Figure 3) and EU27 (Figure 4) due to trade frictions in all scenarios. The UK's exports decline more in relative terms (% change on basis of tons) than its imports and meat exports are most affected in all scenarios. Even in the EEA+ scenario, UK's exports decrease by 10-25% in all product groups due to the additional 5% trade facilitation costs. In the high impact scenario WTO, cereals, meat and dairy product exports decline more than 60%. Imports to the UK decline in all product categories except oilseeds. However, imports of oilseeds increase slightly in all scenarios due to zero tariff rates and since demand for feedstuffs increases. In the WTO scenario, dairy imports decline the most (about 50%) among all products.
For the EU27, exports and imports have moderate impacts compared to the UK. Exports of "Fruit&Veg (Fruits and Vegetables)", meat and dairy products are most affected. In EEA+ and FTA, relative changes of exports and imports are less than 3%. In WTO, exports of "Fruit&Veg" and dairy decrease by about 8 and 11%, respectively. Overall reductions in cereals, meat, and dairy imports to the EU27, can be mainly attributed to a decrease in imports from the UK.

Impacts on prices, production and consumption
Due to a decrease in imports to the UK, producer prices in the UK increase in most products ( Figure 5).
In EEA+ and FTA, producer price changes are rather small (less than 5%) for all products. Impacts on cereals and dairy products are smaller than for other product groups. The strongest impacts are found in the WTO scenario, where producer prices of meat and dairy products increase by 12% and 7.5%, respectively. Only in oilseed markets, producer prices decrease in all scenarios because imports at lower prices increase from the rest of the world (ROW). In addition, production response follows the sign of changes in producer prices. In the UK, production of "Fruit&Veg", meat, and dairy products in the UK increases the most in all scenarios. In the WTO scenario, meat and dairy production increases by 12%, 6%, respectively, due to the higher producer prices. Changes in cereal production in the UK partially depend on feed demand changes in the scenarios. With smaller impacts on trade in EEA+ and FTA scenarios than in the WTO scenario, cereals production decreases slightly as net imports of cereals increase. However, in the WTO scenario, the increase is caused by larger animal production which in turn induces a rise in feedstuff demand.
For the EU27, producer prices decrease for most products, because exports to the UK decline. Meat producer prices decline by 1.8% in the WTO scenario. Producer prices of oilseeds, however, increase slightly because oilseed imports from the UK decrease. Compared to the UK, the relative changes in agricultural production in EU27 are rather small and stay below +/-1%. Production decreases for all products except for oilseeds for which an increase can be observed due to decreased imports from the UK.
Regarding the consumer side, in the UK, consumer prices 9 increase in all product groups except oilseeds. In the WTO scenario, meat and dairy products increase substantially by about 9% and 12%, respectively. Consumption of cereals, meat and dairy products decreases in all scenarios and of "Fruit&Veg" decreases in EEA+ and FTA scenarios. In EU7, consumers in EU27 benefit from a decrease in prices, except cereals and oilseeds. 9 Consumer prices in this study refer to the retail prices paid by ultimate food consumers. They are calculated in CAPRI, using average prices of consumed goods domestically (quantity weighted average over domestic and imported product prices) plus a fixed margin (e.g. transport, processing, and other marketing costs). They should not be confused with ex-farm prices paid by first purchasers of farm products.

Impacts on welfare
We compute the welfare impacts as changes in consumer surplus, producer income and changes in government revenue. As explained in section 4, we include a change in UK's net contribution to the EU CAP budget from two sources: tariff revenue from agricultural trade and share of GDP. The results are summarized in Table 6. The level of trade barriers assumed in the Brexit scenarios explain the extent of impacts on consumers and producers. The scenario EEA+ shows the smallest, and the WTO the largest impacts, whereas the FTA scenario results stay in between. In all scenarios, consumers in the UK (-12 to -125 €/capita) and producers in EU27 (income losses, -0.2 to -2.5%) suffer due to higher food prices and lower producer prices, respectively, than the baseline. However, if we also consider changes in taxpayer welfare including the UK's net contribution to the EU CAP budget, the final welfare level become different. In FTA and WTO scenarios, the UK gains €2.2 and 2.8bn of government revenues respectively by leaving the EU membership, whereas government revenue losses (€-1.9bn) occur for the EU27. Here, we assumed that the UK's transfer payment associated with an EEA+ membership is the same as before Brexit. Only the UK gains from tariff revenues in agricultural trade. In EEA+, the UK and the EU27 experience small welfare losses due to increased trade costs (5% assumed in this scenario). In WTO, even with UK's gain in government revenue, the UK welfare loss (€-3.1bn) is larger than the welfare loss for the EU27 (€-2.5bn). However, in FTA, the net welfare of the UK become positive (€668bn) because the country's gain from phasing out CAP contributions (€2.2bn) exceed the market loss (€1.6bn).

Discussion and conclusions
Brexit is expected to cause disturbances in international agri-food trade in various ways. The direction and the extent of impacts will be determined by the outcome of the Brexit negotiations. This study analyzes Brexit scenario impacts not only in the UK, but also in the EU27. The partial equilibrium global agricultural sector model CAPRI is employed to evaluate impacts on trade, prices, production and welfare.
Our results mainly confirm previous studies (Table 1). Brexit causes increases in agricultural prices (mainly meat, dairy products) in the UK agricultural markets. The results are largely driven by our assumptions on NTB costs, which are in the middle of those used in previous studies (Table 2).
Additional border control, delay and certificates requirement will inevitably increase trade costs between the UK and the EU27. The extent of NTB costs will depend on how much the UK harmonizes food standards and regulations with the EU27 in the long term. In addition, our study contributes to the literature by clarifying ambiguous effects on the cereal markets in previous studies (Berkum et al., 2016;Davis et al., 2017). Our results show that the livestock sector and feed demand affect the response in the cereal markets. A relatively high increase of livestock production due to reduced imports (Scenario WTO) would require higher feed demand and lead to higher cereal production than the reference scenario. But small impacts on the livestock sector demand less feedstuff and cereal production decreases. as an option to reduce increases in agricultural prices, but producers would have to increase productivity to cope with decreasing market prices.
For the EU, Brexit would incur relatively small impacts due to its large economic size. Net market impacts (consumer surplus + producer revenue) are expected to be small, but the loss of UK's net contribution to CAP leads to a reduction in the net welfare except in the scenario EEA+. Moreover, the withdrawal of the UK as a net contributor to the overall EU budget would put pressure on the CAP budget (as seen in the Commission proposals for the next MFF period). Reductions in CAP spending may further aggravate farm income. This may lead to increased food prices and could offset the benefits of Brexit to consumers in the long run, but this is not considered in this study.
Furthermore, other aspects not considered in the model may question the robustness of the direction and strength of our model results. The expected negative Brexit impact on the exchange rate might directly affect the prices of imported agri-food products (including intermediates) and also the prices of primary inputs for agriculture (e.g. mineral fertilizers). In addition, labor market disturbances (e.g. restricted mobility of seasonal workers from Eastern Europe to the UK) could impact labor-intensive agricultural sectors, such as horticulture in the UK. Negotiation of TRQs for third countries preferential access to the UK would be of importance. Moreover, CAPRI does not cover food markets such as beverage, alcohol and some processed products such as jam, frozen pizza, pet food. Thus, absolute levels of consumer surplus changes could be larger than estimated in CAPRI and our model may thus underestimate welfare impacts in the UK. Moreover, CAPRI uses the Armington approach, which has 10 Source: The UK's independent factchecking charity (https://fullfact.org/europe/norway-eu-payments/) the well-known disadvantage of not capturing emerging trade flows (zero trade problem) and understating trade creation from small trade shares (small share problem) (Kuiper and Tongeren, 2006).
Finally, our study implies that increased market inefficiency arising from trade barriers due to Brexit could lead to welfare losses for the UK and the EU27. In particular, the consumer surplus losses in the UK could be substantial (-125 euro/capita) in the scenario WTO. However, net welfare gain could be achieved in the UK due to financial gains in producer income and EU CAP contribution in the scenario FTA. For the EU27, net market impacts are small, but loss of CAP contribution from the UK can lead to net welfare loss. For the UK, no deal scenario (WTO) would impose strong markets impacts for consumers. It remains to be seen how UK's post-Brexit agricultural and trade policy can mitigate those market impacts. For the EU27, producers would likely face income losses due to lower food prices and a shrinking EU CAP budget.