We look at UK connectedness with EEMEA in financial and trade terms. We would not particularly buy the risk of strong economic contagion into the EEMEA region on a Brexit vote, but proxy trades may well focus the minds on liquid crosses such as Poland and even Turkey; however, Czech Republic comes out as one of the most affected in economic terms. We expect the maximum growth impact of 0.7pp on the Czech Republic from trade on a Brexit vote at worst, down to only 0.1pp for Russia, which on all fronts seems least affected. We do not expect Brexit to markedly shift the already existent narrative of authoritarianism in Poland and Hungary, though it could be used rhetorically. A “Czech-xit” referendum is possible, but we do not see similar risk elsewhere in the region. We think the accession paths (however distant eventual success might be) will remain unchanged in countries such as Serbia, while in Turkey the domestic political narrative is more important on EU relations.
Markets, investors and local participants in EM are commenting and thinking about one risk more than any other at the moment – what a ‘Brexit’ vote would mean and the chances of it happening. Nomura’s house view assigns a 25% probability for Brexit (75% probability for Bremain) – although we acknowledge that the vote might be close – and so keeping the UK within the EU. On a ‘Brexit’ vote, however, we see a 2% peak to trough fall in GDP and recession (however, our current 2017 GDP growth forecast of the UK at 1.7% for next year represents a larger, roughly 4%, shock). Similarly, looking at the eurozone linkages Nomura believes that growth could drop by around 0.5pp. Even with a baseline of the UK voting to remain within the EU, investors are getting concerned about the chunky tail risks involved in a successful Brexit vote. We think that while marginal, and with other larger explanatory factors, some of the weak Q1 growth prints in CEE, and especially in Poland, could be attributed to Brexit uncertainty.
We need to look first at financial, remittances and trade linkages.
Naturally, sentiment would be the faster shock channel, indeed possibly already in effect, as uncertainty over the UK’s future growth rates and relationship with the EU (and world) would be transmitted through UK trade and financial relationships. More risk-averse lending attitudes from UK banks into EEMEA at a time when credit growth is generally seen as slow could be a key shock adding to roll-over risk.
We find that UK banks have a very small exposure in percentage of GDP terms and the share of total foreign exposure to EEMEA markets. These data are not perfect. For example, lending via other entities in third-party countries is meant to be dealt with in this ‘ultimate risk’ basis data, but may not always. Equally lending from UK banks to overseas corporate entities of EEMEA companies is also imperfectly captured. However, it does give us a rough sense of scale and relative exposure. We find that South Africa has the largest exposure at 21.7% of GDP, thanks to Barclays and Old Mutual linkages from London. However, Barclays’ exit from South Africa means this will come down significantly, similarly with the restructuring of Old Mutual. By our estimates underlying lending exposure is much lower, at around 6%of GDP. So South Africa still stands out as somewhat exposed, but in general has low foreign credit linkages thanks to its exchange controls framework.
Turkey is the next most exposed at 3.3% of GDP. Again very low, but the time series is interesting, with a big ramp-up in exposure as HSBC has built up its business in Turkey, lending to corporates especially in recent years, but then a reversal of exposure as HSBC and other banks have scaled back. Russia has minimal financial reliance on the UK in terms of direct lending, with less than 1% of GDP. CEE has much higher overall exposure to foreign banks, but these are mainly Austrian and Italian banks. The Czech Republic comes out on top in terms of overall foreign exposure and UK exposure, but even then only 2.2% of GDP.
Broadly, we find that countries have limited potential impact from direct financial contagion from the UK, though South Africa should be watched. Equally, CEE should be watched from any knock-on effect from Brexit on European banks that have exposure into CEE. As such, second-round effects would be the driver of any impact. Short-run financial linkage risks would mainly be on sentiment and lending criteria, in our view. Longer run we suspect that because EEMEA countries are not in the Eurozone they will not be unduly affected in terms of banking access to the UK and the ability for cross-border financing to occur with the UK outside the EU.
We next examine a key component of balance of payments exposure – remittances.
Poland receives the largest volume of remittances from the UK with USD1.2bn per year according to World Bank statistics. This is followed by Hungary with USD385mn – though in each case Germany is a much larger flow.
South Africa has a very significant share of remittances from the UK, though volume is very small. Other countries do not show up as that significant.
Remittances from the UK to our EEMEA coverage are also significantly lower than to India and Nigeria. We can envisage a small impact from a weaker GBP after Brexit through remittances, but this overall risk exposure seems very low.
The biggest and most obvious linkages are trade. EEMEA exports to the UK were some USD50.2bn last year (for our coverage), out of total trade of USD1.1trn – i.e., around 4.6% of total trade.
Figure 5 shows that the shares of exports going to the UK from EEMEA are generally low and in most countries have dropped somewhat since 2007, with the exception of Poland and Romania, which have increased slightly.
Turkey and Poland have the largest share at 7.1% going to the UK (white goods and textiles from Turkey, intermediate manufactured and investment goods from Poland). Russia has the lowest followed by South Africa.
Because of the potential for second-round effects though, we should also look at total UK and EU export linkages. Here the Czech Republic is highest, at nearly 90%, followed by Poland and Hungary. South Africa comes out bottom, with a very large fall in this share since 2007. Turkey has seen a similar diversification.
We also need to consider the size of the share of trade in GDP. Here Turkey drops right back as a more closed economy, similar to South Africa. The Czech Republic has the highest share of UK trade at 4.6% of GDP and is the most open economy, with Hungary following with 3.3% of GDP. Poland is a more closed economy, coming third with 2.8% of GDP.
Again total linkages with the rest of the EU are much more significant, at around 70% of GDP for the Czech Republic and 65% of GDP for Hungary, falling back to 28% for Poland in 2015.
Overall the size of shock transmission may well be bigger via second-round linkages than through first-round effects.
Slower UK growth will have a double-pronged impact through direct and indirect channels from lower partner country growth rates. This is the short-term impact. The longer-term impact would depend on trade access, single-market membership, free trade agreements and tariff levels with the UK. We suspect these longer-term risks on EEMEA growth may be limited, however, because of the UK’s tendency not to tariff imports with an already weak post-Brexit currency and the possibility of competitive advantage even being gained by (particularly CEE) countries over the UK based on access rights. The risks to lower or higher long-run potential growth of the UK (and indeed the eurozone) would be the more significant long-run factor affecting long-run EEMEA growth.
Trade related growth impact
Applying the trade linkages to our house views on lower UK and eurozone growth we come up with a baseline for the growth impact on EEMEA. However, we note the huge amount of uncertainty surrounding these estimates, the possibility of upside (i.e., bigger shock) risks from sentiment effects and financial linkages that we have not attempted to account for in this note (viewing them as too small).
We see growth affected most in the Czech Republic through eurozone trade especially – Brexit would trim off some 0.5pp from growth. In Hungary, we similarly find 0.5pp, 0.3pp for Poland and 0.2pp for Romania. In Turkey and South Africa we see the UK trade impact as equally as big as the eurozone impact because of the relative shares of the trade basket. Here we see their growth falling by 0.1pp, respectively. For Russia, we see the same level.
These levels seem small, reflecting the direct trade impact (and corresponding import impact as well). Arguably adding in additional sentiment shocks to domestic demand and private sector investment would add another decent size chunk of downside (maybe a few tenths of a percent).
We think we have here a rough ranking of exposure, which is important. The most open economies come out worst as expected.
It is worth examining FDI as a risk.
Figure 9, however, shows there has been a steady trend down in the total volume of FDI from over USD10bn a year pre-crisis to only around USD2-4bn a year now. The number of deals has similarly dropped.
This suggests a fair volume of FDI under risk after Brexit, though not too big. Equally, we consider upside potential from more UK FDI into EEMEA and especially CEE depending on UK trade access to the free market (or indeed Turkey, which does have free market access).
As such, while probably skewed to the downside, it is unclear whether FDI risks after a Brexit would be that big, in our view.
EU structural funds
One of the biggest market concerns after a Brexit is what would happen to EU structural fund payments. A snapshot for 2014 shows how Poland stands out for the massive volume of EUR14bn/year it received net from the EU. Hungary follows with EUR5.7bn net, Romania with EUR4.6bn and the Czech Republic with EUR3.1bn.
The next funding semester from 2014-20 saw these figures ramped up further.
We believe that if there was a Brexit the EC would still have to guarantee structural funds because that is part of EU law for this semester to 2020. The Commission would have to run a deficit and borrow to fund them without the UK’s contribution into the end of the period (assuming the UK was still paying for two years at least), or other richer member states could be asked to contribute more. We think the result would most likely be a mixture of the two.
The issue would be more what would happen after the end of this semester. However, note these structural fund payments were due to be scaled back anyway for the 2017- 21 semester as countries become more developed and they are eventually weaned off such payments. Hence, the impact of the UK leaving might be somewhat limited. Removing the UK’s share of net contributions from CEE countries net payments would only mean a fall of around 4% in net funding if it was looked at mathematically. As such we do not view this as a particular concern.
Political impact on CEE / Czech-xit Some market participants believe that Brexit would cause isolationism and anti-EU sentiment in Hungary and Poland (and to a degree in Czech Republic) even worse. We disagree with this.
We think the political dynamic in CEE is very much driven by domestic politics and fiscal and other legal conflicts with the EU arising from such domestic policy agendas. As such, we feel there is a sufficient driving force against the EU irrespective of a Brexit vote or not. True, a Brexit vote could become part of the PR and political narrative, but the foundations of the relationship with the EU (and pivoted vs Russia) we think are far more fundamental. However, Hungary may well see the biggest ramp up in such PR on its referendum on rejecting refugee policy from the EU – but again the vote looks set to go the government’s way anyway.
We see anti-EU sentiment in Hungary and Poland from their governments especially being more combative and asserting sovereignty – not as a desire to leave. We do not see the probability of leaving as increasing in either country should there be a successful Brexit vote in the UK. Each country has and continues to benefit too much from the EU in terms of trade, jobs, migration, structural funds and historical convergences paths, which include normalising legislation and the rule of law (the current spat with Poland not withstanding).
The exception is the Czech Republic, where there have been some expectations of a ‘Czech-xit’ referendum, tapping into the long-run euro-scepticism of the country. As one of the most open economies, the country’s ability to pass a referendum to leave is unlikely, in our view, and even getting one in the first place would be fraught under the current government and current president, who is much less a eurosceptic than his predecessor.
We think a Brexit would marginally strengthen President Erdogan’s lukewarm view on Turkey’s relationship with the EU. But here again domestic politics is likely to already have had enough of its own momentum as power shifts towards the president and away from what was previously a more Europhile government. Brexit could be used to extract further conditionality on the refugee and visa issues, however.