Bulgaria is now poised to join the euro. This comes two decades after the launch of the European Monetary Union (EMU) and over 20 years since Sofia introduced a successful currency board regime. Both anniversaries are a convenient starting point to assess Bulgaria's economic performance not only in absolute terms but also relative to other Central and Eastern European countries that joined the EU and the euro in the interim.
The most revealing figure is the income per capita. Over the past 20 years, this has more than doubled in Bulgaria. Living standards have certainly improved but part of this upsurge should have occurred anyway. Poorer countries tend to catch
up and grow faster if they trade freely with the rest of the world and maintain open markets. All New Member States (NMS) that joined the EU in the last two waves since 2004 have closed the gap with richer members. So Bulgaria should be judged not in absolute terms, but also relative to its opportunities, and the performance of its European peers.
#1. WHAT IS THE EURO'S OVERALL IMPACT?
1.1 THE EURO HELPED CEE COUNTRIES GROW
Figure 1 illustrates the general catch-up process of the NMS from Central and Eastern Europe by comparing theirgrowth rates (of real GDP per capita) since 1999 to the initial level of GDP per capita. It is apparent that there are two groups: the old EU-15 and the NMS. Among the latter there is strong convergence and the (initially) poorer countries grew
faster than the others, narrowing their income differentials with the more affluent EU Member States.
However, among the group of old Member States (in the blue oval), there is an "awkward" trend. The lower-income countries (Greece, Portugal) grew slower than the richer ones (e.g. Germany). Moreover, Italy, and initially high-income country, recorded the worst performance (alongside Greece), substantially diverging from other old Member States.
Some EU sceptics cite this as evidence that the euro has hindered convergence. However, this is not the case, at least not for the NMS. The vertical axis in Figure 1 shows that real incomes have increased by over 100 % in most of the NMS. This alone indicates that these countries have experienced a long period of strong growth and development.
Significantly, all countries (green dots) which are (today at least) in the euro area have performed better than suggested by the regression line. Estonia joined when the euro was in crisis, but this has clearly not undermined its growth which has been above its potential. A raw, direct comparison between the growth of Slovenia and that of the Czech Republic, for example, might suggest that Slovenia (an early euro joiner among the NMS) has underperformed compared
to the Czech Republic which does not even intend to join the euro.
However, this methodology would be flawed since it does not acknowledge their different starting positions. Slovenia had already been much better off in 1999, so one could not have expected a very high growth rate anyway. if the analysis factors in the potential given by the starting level of GDP per capita, Slovenia has actually surpassed expectations.
Bulgaria's dot lies somewhat below the line; this suggests that it could have done better, at least relative to the rest of the NMS, including those that have joined the euro. The better than expected performance (based on their low starting positions) of NMS which adopted the euro earlier has another important implication. Opponents of euro area enlargement claimed that premature accession could trigger national boom-bust cycles, so endangering convergence.
The real convergence which the NMS have experienced, i.e. catching up with the older Member States, however, tends to contradict this. Such cycles have occurred over the past two decades and have proven very destructive. However,
these have also been experienced by countries outside the eurozone, Iceland being the extreme example. And it is far from certain that eurozone membership played any role in these boom and bust developments. So history suggests that the euro is linked to better economic performance among the new Member States.
1.2 IS BULGARIA DIFFERENT FROM OTHER MEMBER STATES?
B Bulgaria is a special case and difficult to compare to other countries which have already joined the euro. For example,
much more advanced industrial economy integrated into the German value chain, especially in the automobile sector. The Baltic countries - like Bulgaria - had currency boards. So it might be assumed that their experiences would provide some insights. But this would be to discount their economic structures.
All three states also had lengthy waiting periods, giving them time to prepare. Lithuania was the last country to join, on 1 January 2015, after Estonia (2011) and Latvia (2014). Bulgaria's structure of exports shows a large proportion
of 'traditional' products in sectors which often characterize commodity-producing emerging markets (e.g. Metals, Mineral products, Vegetable produce). These three sectors make up about one-third of all exports. In the case of Romania, which had the same income per capita in 1999, these three sectors account for only about 16% of the total, one half of their share in Bulgaria. Among the industrial sectors, machines represent only 18% of the total, followed by textiles with 10%.
The low percentage of advanced sectors in exports is one key distinctive feature of Bulgaria, even among the new Member States. Among the NMS outside the euro, Bulgaria has the lowest specialization in Machines and Transportation. For Romania, which has the GDP per capita closest to Bulgaria, these two sectors alone account for one-half of its exports.
Hungary's export composition differs even more markedly and is closer to that of developed economies with cars and machines alone accounting for 60% of total exports. This compares to just 6% for metals and mineral products are taken together.
A comprehensive indicator of the nature of the overall export structure is the 'Economic Complexity Index' (ECI) developed by Cesar Hidalgo, from the MIT Media Lab and Ricardo Hausmann from Harvard University. This index shows the extent to which a country is enmeshed in the global division of labour by looking at the number of products exported by the country and where in the world they are used. By calculating what products a country exports and where they go, analysts can provide a measure of the knowledge accumulated in it. Advanced industrial economies usually have an ECI of around 2. By contrast, exporters of raw materials and commodities have an ECI below 1, often around 0.5, because extracting raw materials from the ground does not require much human capital.
Figure 6 shows the ECI for a number of NMS in 2016. It is apparent that Bulgaria has the lowest value, which suggests that the knowledge intensity of the Bulgarian economy is rather low. Moreover, Table 1 shows that there has been little progress over recent decades. The indicator for Bulgaria has increased but only slightly, leaving the level of 2016 considerably below those of other NMS as shown in the figure above.
Most other NMS have advanced more than Bulgaria, except for the Czech Republic and Slovakia, which already had a high level of economic development from the outset. The low knowledge intensity of Bulgarian exports is one of the factors behind the growth underperformance (relative to expectations as seen in the regression line in Figure 1). Conversely, higher knowledge intensity would increase the gains from the euro since the common currency facilitates the integration of local industry into the European value-added chains.
Although adopting the euro will not have any direct impact on the structure of industry and exports, the lower transaction costs documented below should favour a deeper integration of the economy in the European value chains (e.g. small-scale automotive suppliers). Over time, this would open up possibilities for entering high-knowledge sectors.
1.3 HOW DO WE PROPERLY MEASURE THE IMPACT FROM ERM II AND EUROZONE MEMBERSHIP?
A s demonstrated above, it is difficult to analyze the impact of euro area membership by just comparing two countries without factoring in their different starting points and potentials. However, one comparison can clearly be made, namely between the Czech Republic and Slovakia. Their shared history, similar institutions and a floating exchange rate would seem to make them ideal candidates for a direct comparison. So it would be tempting to just measure the performance of the two countries before and after Slovakia joined the euro in 2009.
But this would be misleading since Slovakia's economy was far behind the Czech Republic's at the time. Comparing only the years after the introduction of the euro is to overlook a crucial fact - most of the impact of this final step had already come earlier through ERM II membership and anticipation effects. Therefore, the analysis needs to start earlier.
Hence three authors from the OECD decided to estimate the macroeconomic effect of the euro on Slovakia by comparing its economy with a 'synthetic' country that has similar characteristics. This 'synthetic control method' (SCM) consists of comparing the income per capita of Slovakia to a mix of 2/3 Czech Republic and 1/3 Romania. The authors found that between 2006 (when the decision to join the ERM II was made) and 2011, the Slovak economy grew much faster than its synthetic counterpart.
The overall effect by 2011 was that the euro increased the real GDP per capita in Slovakia by approximately 10%. Slovakia had underperformed before deciding to adopt the euro. But this apparently boosted its economy and overall standing. Several governance indicators improved until about the time the decision was taken (2006), which probably contributed to the ensuing growth spurt.
The OECD study finds: "Two-thirds of the positive gain is observed already by 2008, emphasizing a strong anticipation effect. Nevertheless, the gap in GDP per capita widens between 2008 and 2011 by an additional 3 percentage points.
Further, in line with the standard literature, we find that had Slovakia kept the floating currency regime during the recession in 2009, the economy would have been temporarily better off by roughly 2%."
Similar studies are unavailable for other countries, but this result concurs with our contention that, on average, euro area countries from Central and Eastern Europe have slightly surpassed expectations given their starting positions.
#2. WHAT IS THE EURO'S IMPACT ON ECONOMIC FUNDAMENTALS?
2.1 HAS EUROZONE MEMBERSHIP IMPROVED THE COMPETITIVENESS OF THE NEW MEMBER STATES?
It seems clear that the adoption of the euro facilitates trade. But it is important to distinguish between two effects.
The Figure 7 shows the ratio of intra- to extra-EU trade using the average of imports and exports. It is apparent that all NMS are very much integrated into the EU's economy. Most values are above 2, implying that intra-EU trade is twice as important as trade with third countries. Brexit will impact this ratio only marginally because the UK is not a key trading partner for most new Member States. The figure below also shows that intra-EU trade has become relatively more important in all of the countries considered.
For example, Lithuania is already in the euro (and was in the ERM II for a long time), but despite that, it is difficult to see any difference with Bulgaria - the two countries have a very similar ratio of intra- to extra-EU trade and a very similar increase over the period under review. Hungary and Romania, which have floating exchange rates, both have a much higher ratio than Bulgaria (which is much closer to the euro with its currency board). Moreover, these two countries also show considerable increases, which, in Hungary's case, is much larger than Bulgaria's.
Hence too little data is available to check whether the euro has been the main driver of the increase in trade of the new Member States which recently joined the eurozone. This makes it difficult to make a strong prediction for Bulgaria. There is, however, a lot of research on the impact of the euro on trade for the 'old' Member States. For example, Volker Nitsch and Mauro Pisu have generally concluded that the euro has increased trade within the currency union. Of course, the old Member States had had floating exchange rates, so their experience may tell us little about what Bulgaria should expect given its 20-year pegging to the euro.
As with trade, it is difficult to disentangle the impact of the euro, the ERM II and EU membership on Foreign Direct Investment (FDI) flows into the new Member States. The available studies are mostly based on data from the old Member States.
Acquiring know-how and new technology was a prerequisite for the new Member States to catch up and expand their exports as rapidly as they did (most NMS from Central and Eastern Europe have seen an average growth rate of exports of about 10% per annum for almost 20 years). Analysts could even argue that countries which already have a large stock of FDI, and where a large proportion of their industry is dominated by foreign firms, have reached saturation point. This would leave little room for new FDI inflows. Even though it is difficult to isolate the effects of the euro adoption from other possible factors at play, the European Central Bank recently concluded: "On average, joining the EU increased inward FDI flows from other EU countries by 43.9%, but did not have a significant impact on a country's capacity to attract FDI from non-EU countries.
On average, adopting the euro increased FDI from other euro area members by 73.7%. Thus, the additional effect of belonging to the common currency area can be estimated at around 20 p.p." The reasons for this are clear: the EU reduced the cost of cross-border business among its members, and the euro area stimulated cross-border capital flows as exchange and liquidity risks were eliminated.
As explained, the data for the new Member States is less clear-cut. The figure 8 shows the inflows of greenfield investment (projects) in four countries from CEE. It is apparent that these flows were highly variable, especially for Bulgaria, which had at times the highest value; higher in absolute terms than Hungary, the Czech Republic or Slovakia.
However, this value was reached only in 2006, close to the peak of the boom (and probably some of these projects were not implemented when the bust came).
This figure tells us that, overall, all four countries show a very similar overall trend despite their different currency regimes. Slovakia's case is again instructive. As explained above, Slovakia should not be compared directly to the Czech Republic, but rather to a weighted average of Romania and the Czech Republic (with weights of 1/3 and 2/3, respectively). This is done in Figure 8 showing the actual inflows for Slovakia and its hypothetical counterpart which has not joined the euro (both expressed as a percentage of GDP). It is apparent that after 2007/8 the two lines remain rather close. The largest difference occurred in 2008 before the country joined the euro. Part of these higher FDI inflows before 2007/8 can be ascribed to ERM II membership and the anticipation of this step as well as eventual euro membership.
However, this interpretation is difficult to square with Slovakia having experienced very high FDI inflows in 2003/4, i.e. years before it opted for euro membership. Bulgaria provides a similar experience. It used to attract rather high inflows as a percentage of GDP but since about 2013 the rate of inflows into Bulgaria is close to the average of the CEE countries.
However, given the higher flows of the past, the stock of FDI is now higher as a share of GDP for Bulgaria than for these other countries. In short, it is very difficult to predict with any certainty the impact on FDI flows following the announcement that the country will eventually join the euro. Moreover, in Bulgaria's case, for the first time, membership of the Banking Union will precede euro membership. This should reinforce the confidence effect. But since there is no precedent, it is difficult to say how important this will be for foreign enterprises mulling a new investment in the country.
2.2 IS ERM II MEMBERSHIP A BOOST FOR REFORMS?
E RM II membership brings no material changes to the economy of a country with a stable currency board. This is true for Bulgaria, but it was also true for the three Baltic states. However, it seems that the expected adoption of the euro (and the commitments made in this context) did spur reforms and improvements in governance even in these countries (which joined between 2007 and 2015). The figure below shows the change in the WGI (World Governance Indicators) of the World Bank over the two years leading up to the euro adoption and the posteuro period.
2.3 ARE EURO MEMBERSHIP FEARS OF INFLATION OVERBLOWN?
Some worried that introducing the euro into the countries concerned would lead to higher prices. However,
concerns about high inflation through the eurozone proved groundless. Firstly, it is important to distinguish between a one-off jump in prices and persistently high inflation. The real concern is that traders would exploit the new currency to 'round up' prices, whereas salaries would be converted directly.
Few would contend that the euro would lead to permanently higher inflation. In reality, introducing the euro is usually accompanied by a very small - indeed marginal - uptick in prices, essentially during the first few months. No reputable academic study has found a rise of more than 0.5% - if official consumer price indices are accepted.
Of course, there are a lot of anecdotes abound about hefty hidden price increases. In Germany, it was widely believed that retailers had often converted prices in the old Deutsche Mark onetoone into euro prices. Given that the DM-to euro exchange rate was almost 2:1, this would have implied an effective doubling of prices.
Something similar is to be expected in Bulgaria since the lev was pegged at 1:1 to the DM when the currency board was introduced in 1997. In social networks, such stories will certainly circulate. In reality, these cases were extremely rare and numerous academic and other studies on this subject invariably find a negligible or non-existent impact on inflation. Higher prices through 'rounding up' were thus the exception rather than the rule. It is true, though, that this type of price 'gouging' was more commonplace in some areas widely used consumer services (personal services, restaurants, or household services), so perpetuating the impression of a hefty overall price hike.
Careful studies of the impact of the euro on general prices revealed no evidence of a material uptick in average prices. Most people, however, were convinced that prices had increased.
For example, a European Commission study found that 58% of Latvians thought so. For Estonia, an ECB study finds a moderate uptick in inflation (less than 0.5%) six months before the changeover, with higher price increases in smaller shops. Prices in Lithuania were falling at the time of the changeover (deflation of over 1%), so there was no uptick in prices. But the European Commission finds that the switch to the euro led to prices falling about 0.1% less than they might otherwise have done. Improved price transparency is generally more important than marginal price increases in the
wake of the introduction of the common currency.
Consumers can, of course, already easily compare prices across countries thanks to the internet. But high currency exchange costs limit their ability to buy across borders, even online, if they do not have euros. Recent research has shown that the introduction of the euro leads very quickly to almost complete price convergence, especially for goods offered by large retailers across many countries. Three professors from the MIT and Harvard investigated the prices at the Latvian stores of global clothing retailer Zara around the time of the euro's introduction.
They found that as long as the lat existed, prices at Zara's stores in Latvia frequently varied by between 5-10 % for the same goods in euro area countries (translating lat prices into euro at the fixed exchange rate of the currency board). However, within a month of the introduction of the euro, these differences disappeared. This elimination of price differentials did not lead to higher average prices. Nevertheless, the public focuses on certain price hikes, thus creating the false impression that the euro has increased prices overall. Moreover, there is evidence that the mere change of currency (and units of account) is unsettling for many. The authors of an erudite academic study entitled "Psychological Costs of Currency Transition: Evidence from the Euro Adoption" contend that "there is evidence of substantial psychological costs associated with currency transition, especially for the old, the unemployed, the poorly educated and households with children".
For all these reasons it seems unavoidable that the introduction of the euro leads to complaints about higher prices, even if average prices remain largely unaffected. However, experience has shown that such complaints wear off quickly because prices remain very stable after the one-off rounding up of prices at the time of the switchover.
2.4 WHAT ARE THE EFFECTS ON ASSET VALUATIONS (STOCK EXCHANGES, REAL ESTATE) AND FINANCING POSSIBILITIES (CREDIT RATING, INTEREST SPREADS, ACCESS TO CREDIT)?
It is difficult to establish whether euro entry or ERM II membership have had a significant impact on stock markets since markets tend to anticipate events. It makes little sense to look at the evolution of stock markets after the ERM II entry because it was usually known a long time beforehand. However, given that euro adoption brings more transparency and predictability, allowing more cautious investors to enter, it can be assumed that market liquidity should increase.
House prices, which do not react instantaneously to news, might provide a more reliable guide to the impact of ERM II membership on asset valuations. However, property prices are driven by long cycles, which do not seem much
affected by ERM II or euro area membership. In recent years, price increases in the New Member States have tended to outpace those in the rest of the EU, as Figure 13 shows. So it's difficult to detect any definitive correlation between ERM II or even euro membership on house prices. For example, the Slovak real estate market experienced considerable growth prior to the country's euro adoption, a trend that can be attributed to the anticipation effect.
However, the boom in real estate prices coincided with very strong economic growth at the outset of the 2009 economic crisis. Then the real estate market dipped. The data shows, however, that the euro brings more stability, and reduces price volatility in real estate markets. The trend of rising house prices in Bulgaria might continue for some time, not so much because of the euro, but because higher growth and reforms associated with the introduction of the euro tend to increase confidence. This, in turn, lifts the value of property assets.
Bulgaria's credit rating has improved considerably over past decades, thanks to a conservative fiscal policy which brought the debt/GDP ratio down from over 90% to about 20% today. Despite the positive trend, the country could have done even better. Moody's credit rating for Bulgaria was the last set at Baa2 with a stable outlook. Standard & Poor's credit rating for Bulgaria stands at BBB- with a positive outlook. This level is considerably lower than, for example, the Czech Republic, whose fundamentals in terms of debt and growth are considerably worse. Euro membership should lead to further improvements.
However, these expected improvements, which materialized in other late adopters of the euro, should not be considered an automatic given irrespective of government policy. Experience and research show that the rating of a country's sovereign debt depends only partially on the stock of debt. Figure 15 shows the weight of a number of different factors which empirically concur to determine a rating. These follow a study which examined the ratings of dozens of countries over a long period (Haspolat (2015)). The figure shows that the stock of debt (relative to GDP per capita) is only the third most important factor. It comes after GDP per capita and government effectiveness (measured by the WGI indicators of the World Bank used above). Adhering to the rule of law is also almost as important as having a low debt. Higher growth (and higher investment) also helps to improve the rating. Exchange rate volatility is one factor, but only a minor one.
This means that the currency board and perhaps now ERM II membership might lift the rating by one notch. But further advances will depend on progress in other areas. The continuing improvements of governance in the three Baltic States might therefore explain their constantly upgraded ratings rather than their adoption of the euro. Interest rates remain in general higher in the new Member States, even in those countries which now have the euro, then in most other euro area countries. This should not be surprising since the risks for banks are also higher. For example, as bankruptcy procedures are often slower in many NMS and their outcome more difficult to predict, the recovery of collateral is more uncertain. However, it is more difficult to explain why rates have remained even higher in Bulgaria.
Figure 16 shows the interest rate charged on new business loans in the four countries which have most recently adopted the euro (Estonia, Lithuania, Latvia and Slovakia). It is apparent that for these countries there have been considerable fluctuations, but little trend movement in recent years. For Bulgaria, there has also been little movement until the last quarter of 2018. But then there was a sudden fall in rates in the last two observations in 2018. It is too early to say whether this is the first consequence of the announcement of the coming of the ERM II and accession to the Banking Union.
The banking system of every country has its own specificities. Hence, it might thus be better to compare Bulgaria to the average of the new euro area members.
Figure 17 shows this average and also the value for Italy (whose value is above that of the euro area average). It is clear that rates charged on new business loans remain at above 3% in Bulgaria, far exceeding the average of the new euro area Member States; and much higher than in a country like Italy. This implies considerable room for further reductions in the borrowing costs for enterprises in Bulgaria.
2.5 WHAT HAPPENS WITH TRANSACTION COSTS?
There have been many studies of savings in transaction costs arising from the single currency. They are real enough, but they are also sometimes over-egged. Much of Bulgaria's trade with the euro area (and indeed most EU) countries are already being conducted in euro. Moreover, many firms already have accounts in euro. This means that the formal adoption of the euro will change little for them. However, since they have to pay their employees and most other domestic costs in local currency (lev), they still need to exchange most of their euro earnings into levs. And the reverse also applies: importers have local revenues in levs, which they need to exchange into euro to pay suppliers from the euro area
Exchanging currencies involves costs, which can be measured by the bid-ask spread. This spread expresses the cost of a round-trip transaction (i.e. the loss that arises if one exchanges 100 levs into euro and then exchanges the euro amount received back into levs). For small cash transactions such as for tourism, these costs can be high, sometimes up to 10%, considering the buying and selling rates for cash in airports or currency exchange booths in tourism centres. The table below shows the currency exchange rates offered by the largest bank in Bulgaria on its website.
The last column in this table shows the bid-ask spread, which is calculated as the percentage difference between the rate at which the bank sells and at which it buys foreign exchange for noncash transactions. The column entitled BNB gives the reference rate of the Bulgarian National Bank, which, however, does not represent a rate at which enterprises can make transactions. Larger enterprises might be able to negotiate better rates than those published on the website of this bank. But a large part of the exports of Bulgaria to the rest of the EU is made by SMEs, which probably do not have this market power. Bulgaria's foreign trade turnover (imports plus exports) denominated in euro should benefit from a saving of 0.4% (half of the bid-ask spread as shown in table 2). To calculate the total transactions costs savings, the total trade conducted in the euro needs to be determined.
This analysis has two components. First, Bulgaria's trade with the euro area is worth about 80% of GDP (resulting from the fact that two-thirds of the overall trade is with the euro area and that total trade to- GDP ratio exceeds 120%). But then there is trade with other countries, which might also be conducted in euro already. Eurostat data suggests that about one-half of Bulgaria's exports and about one-third of Bulgaria's imports (to and from third countries, respectively) are denominated in euro. If we use the upper limit given by exports we would find that trade turnover worth about 100% of GDP would be affected by these savings.
As the transaction costs, as calculated above, amount to 0.4% of every euro trade turnover, their elimination should save about 200 million euro per annum. Even firms which trade with non-euro countries will benefit from the adoption of the euro as the spreads between the euro and the US dollar, for example, are usually lower than the lev-dollar spreads. This additional gain will be real because there are indeed very large differences in the bid-ask spreads between euro area countries and Bulgaria.
Figure 18 shows the bid-ask spreads (in percentage terms for non-cash transactions) from the websites of two banks belonging to the same group - one in Austria and one in Bulgaria. The Austrian bank should be representative of a relatively small euro area bank. It is apparent that in all cases the bid-ask spread is several times higher in Bulgaria than in Austria. For example, for the US dollar, the bid-ask spread is 0.9% in Austria, while in Bulgaria it is 3.5%. Austria, where the exchange rates are always quoted against the euro, offers much better terms for foreign exchange transactions than Bulgaria.
The bid-ask spread for the US dollar in a euro area country is about the same size as the spread for the euro in Bulgaria. This is the natural consequence of the euro being the main reference currency for foreign exchange transactions in Bulgaria, whereas in a euro area country such as Austria the US dollar is the main foreign reference currency. However, the market for the bilateral 'cross rate' of the lev against the British pound or the Turkish lira will be rather thin, resulting naturally in a large bid-ask spread. By contrast, the markets for the British pound or the Turkish lira against the euro are very large and liquid, with much lower bid-ask spreads. Once Bulgaria joins the euro, transaction costs for other currencies should also fall considerably for Bulgarian firms and individuals.
Substantial savings in transaction costs can also be expected in Bulgaria's trade with non-euro area countries. Per unit of transaction value, these savings might be even higher compared to lev-euro deals. For example, the bid-ask spread for lev-dollar transactions is 3.5% in Bulgaria, but the spread for euro-dollar is only 0.9% in Austria. The potential fall in costs for US dollar transactions would be 2.6 p.p., and the same for the British pound. In the case of the Turkish lira, the fall would be much larger (from 10% to 1.3%).
Applying the rule of thumb that half of the bid-ask spread should be taken as the implicit cost per oneway transaction, we would estimate that the savings would be 1.3% for US dollar transactions and over 4% for transactions in Turkish lira. To be conservative, it might be better to use the US dollar savings for all non-EU trade. The fall in the spread to the UK
pound would be the same size as that relative to the US dollar. But the savings would be a massive 4% against the Turkish lira. Since a large part of Bulgarian trade with Turkey is presumably conducted in euro or US dollar, the exchange rate savings should apply only to a small volume of transactions.
Moreover, SMEs are less important for the extra- EU trade, which is dominated by larger enterprises. They might be better placed to obtain more favourable terms than those advertised on the website of the bank. To be conservative, we might assume that the average bid-ask spread effectively applied for Bulgaria's trade with non-euro area countries and not denominated in euro is about 1%. These savings should be applied to about 1/6th of Bulgaria's trade turnover, or in other words about 20% of GDP. This would lead to an estimated saving on non-euro trade of about 0.2 % of GDP or 100 million euro.
On top of this, there will be considerable savings in the costs of making transfers. Within the euro area, there is no difference in costs between domestic and cross border transfers. Considerable costs, however, are still incurred in making a transfer from a bank account in Bulgaria to an account in the euro area. The cost of a transfer can vary but Bulgaria's
Institute for Market Economics recently estimated that the average cost might be around 20 euro.
In principle, this is a good base to calculate overall savings by looking at the overall number of transactions.
But this is difficult to determine precisely as all money transfers from different banks would have to be calculated.
The European Central Bank collects the numbers in special transactions statistics and reports that there are about 4 to 5 million cross border credit transfers to and from Bulgaria each year. This would imply that the additional total transactions cost savings from participating in the low-cost euro transactions system (TARGET 2) should be around 100 million euro or about 0.2% of Bulgarian GDP. An assessment of the Latvian central bank arrives at a similar, but somewhat smaller, estimate as a percentage of GDP. The total transactions cost savings, combining all three elements, should thus be worth about 0.8% of GDP or 400 million euro per annum. This figure might be somewhat higher than estimates from
other countries because it uses more granular data on bid-ask spreads and because the direct savings through the lower transactions costs via the TARGET system are often not considered.
#3. HOW MUCH IMPACT WILL BANKING UNION ACCESSION HAVE ON BULGARIA'S BANKING SYSTEM?
Bulgaria's road to the euro is unique in terms of the sequence of steps it needs to undertake. Before entering
the euro. However, in Bulgaria's case, the European authorities thought it wiser to scrutinize the country's banking system more carefully before joining the euro. The 2014 crisis of the Bulgarian banking system triggered by the collapse of Corporate Commercial Bank (Corpbank) created deep concerns about the quality of supervision.
Academic research has long emphasized that even non-euro area countries might profit from joining the Banking Union because membership would provide a stable arrangement for managing financial stability. This applies in particular to Bulgaria and other CEE countries because their banking systems are in any case dominated by banks from the euro area, i.e. banks under the supervision of the ECB. The table below shows the importance of euro area banks for the banking systems of the NMS which are not yet part of the euro).
The second column shows the share of cross border assets, which is highest in the Czech Republic and Croatia, with Bulgaria coming third with 77%. Most of this presence is аffected via subsidiaries, although Bulgaria also had (in 2014) a nonnegligible share of cross-border assets via branches. Not all foreign-owned lenders in Bulgaria have their parent banks based in countries of the Banking Union but banks from those countries had a sufficiently high share of 58% of the overall market in terms of assets, which is similar to Romania and Poland.
Foreign dominance of the local banking system has not always been a blessing. Financial instability in the home countries of foreign banks has tended to spill over when at the height of the first leg of the financial crisis (2008/9), the national supervisors in those countries tightened restrictions on intra-group cross-border transfers. The move limited the ability of multinational banking groups to re-allocate capital and liquid assets to their subsidiaries in CEE countries, many of which were facing even deeper problems. This experience led local supervisors and regulators to lift their own standards, imposing higher capital and liquidity requirements to make sure that the subsidiaries in CEE could survive as 'stand-alone' entities. This naturally leads to a higher cost of capital for banking in the host country.
At the time, the problem was solved through the so-called Vienna Initiative under which the major banks present in the CEE region agreed to keep lines of credit open and provide the necessary capital to keep reasonable lending conditions.
But such ad hoc mechanisms are unreliable longterm. Banking Union membership can resolve this problem of national ring-fencing during both booms and busts, because it allows consolidated supervision, as opposed to sub-entity standalone supervision.
During the boom years, national authorities in CEE countries had a hard time addressing excessive credit creation through national supervisory action, because banks indulged in supervisory arbitrage by relying on funding and capital from their parent banks. Similarly, national authorities had difficulties in preventing massive withdrawals by western European banks; hence the need for the Vienna Initiative. Within the Banking Union, these issues could be more easily addressed.
Finally, membership of the Banking Union will give more regulatory certainty in times of crisis. Investor and banking managers do not need to fear unpredictable national regulatory measures. Most banking systems in the NMS are dominated by foreign banks, as documented above. Normally this helps to stabilize the domestic market in case of national shocks. For example, when their outsized booms turned into a bust, the three Baltic countries benefitted enormously because their banks were fully-owned subsidiaries of resilient Swedish and Danish banks.
The mother banks were strong enough to absorb the large losses of their daughter banks in Estonia, Latvia and Lithuania. These mother companies were also patient enough to see through the problems and keep their subsidiaries afloat. This was possible, of course, because the crisis did not affect Denmark and Sweden unduly and their banks were strongly capitalized.
Other Central European countries had a more difficult experience with foreign banks because those from Austria had a weaker home base than those of Denmark and Sweden. However, even though Austria experienced some difficulties in 2008/9 (because of the massive investment of its banks in CEE), the Austrian state remained strong. It kept the confidence of investors and its rating was downgraded, but only from the very best, AAA, to second best, namely AA+. Austria's rating remained much better than in any of the countries where Austrian banks were operating. Unfortunately, the case of Bulgaria today does not fit this pattern. Some of its foreign banks come from countries which are not stronger.
UniCredit Bulbank, a unit of Italy's UniCredit, is the largest bank in Bulgaria. Its rating is, of course, tightly linked to that of the Italian government. In late 2018 Moody's downgraded most Italian banks following its downgrade of the Italian government's bond rating to Baa3 with a stable outlook from Baa2. Standard & Poor's credit rating for Italy stands at BBB with a negative outlook.
So it is that Bulgaria has at present the same rating level as that of Italy, the home base of the largest foreign bank in the country. Moreover, Bulgaria's rating is improving which implies that its rating might soon overtake Italy's. Naturally, this potential discrepancy creates a concern for the Bulgarian authorities, as weak foreign banks might become a source of instability. In this context, the Banking Union has become particularly important for Bulgaria in at least two aspects.
First, the ECB's pledge to supervise the local subsidiaries of large foreign banks (technically the Single Supervisory Mechanism, part of the ECB) protects against improper capital transfers or other intra-company transactions that an Italian bank might consider in the event of domestic funding problems. Of course, this does not apply only to Italian banks. Any foreign bank might be tempted in times of crisis to shift capital and liquidity from its Bulgarian subsidiary back to the mother company if the parent suffers difficulties.
The ECB will be a much more impartial supervisor than the national ones, which tended to favour banks headquartered in their home country. Moreover, market confidence in the mother companies should also improve as the SSM enforces the adjustment to underlying problems (large stocks of NPLs, for example), something which national supervisors had neglected to do for political reasons.
Second, the Single Resolution Fund provides strong protection against problems at mother banks. The Single Resolution Fund will have in a few years over 50 billion euro at its disposal (and could raise quickly an additional 25 billion euro). This is more than the annual GDP of Bulgaria. Also, the European Stability Mechanism is soon likely to be able to provide a backstop to the SRF in case a very large bank has problems. This is to say that the financial means to rescue
banks or help them recover are large enough even for the big lenders which have subsidiaries in Bulgaria.
#4. ENTRANCE FEE FOR THE ECB?
Membership of the euro and the Banking Union requires some payments to be made, but this is not an 'entrance' fee.
It is important to bear in mind that the contribution of the BNB to the capital of the ECB will be remunerated because the BNB will participate in the profit of the ECB. The total capital of the ECB amount to 7.750 billion euro, but its profit has been around one to 1.2 billion euro annually, which represents a rate of return of over 15 %.
Moreover, the BNB will, of course, participate in the so-called monetary income of the entire euro system.
The experience of other countries confirms this. For example, the figure below shows that the profits of the Central Bank of Lithuania increased after the introduction of the euro by over 30 % over the previous two years.
4.1 ENTRANCE FOR THE SINGLE RESOLUTION FUND (SRF)
I n the first four years of its existence, the SRF has already collected over 20 billion euro (which will probably increase by another 5 billion this year) of ex-ante contributions from banks. These contributions will continue until the SRF reaches its target size of 1 % of covered deposits, or probably around 55 billion euro.
This sum represents a considerable burden for the banking system. However, the SRF can, among other things, be used to guarantee the assets or the liabilities of an institution under resolution, or purchase its assets or make loans to it. It can also make a contribution to the institution under resolution under specific conditions (Bail-in of at least 8% of the total liabilities including the funds of the institution under resolution and the contribution from the SRF is pegged at a maximum of 5% of the total liabilities including the funds of the institution under resolution).
The overall cost of creating the SRF is justified by the need to ensure the stability of the European banking system by implementing the key elements of the European Banking Union. The financial muscle of the SRF is needed when a bank gets into difficulties. This is why the distribution of contributions across banks (the cost of building up the capital of the SRF) should be linked to the risk a bank poses to the stability of the system and the SRF in particular. The current methodology for calculating contributions is complex. But it brings important advantages for Bulgaria.
The overall target size of the SRF has been determined as 1 % of covered deposits. However, the contribution each bank has to make each year is not just proportional to (the same fraction of 1 % of) the covered deposits at that bank. This would not be appropriate since two banks with the same amount of covered deposits might represent a very different risk profile.
This is why the contribution of each bank to the build-up of the SRF is determined in a two-step procedure. In the first step, the annual contribution expected from the entire banking system to the SRF is determined (so as to reach the 1 % target by 2025). In the second step, this overall sum is distributed across individual banks according to their relative (not absolute) risk characteristics. It is this second step which needs to be analyzed. The figure below shows the outline of the methodology used by the Single Resolution Mechanism (which governs the SRF) to calculate the contribution each individual bank has to make to the SRF.
The basic approach followed so far to determine the contributions banks have to pay to the SRF is based on a number of different balance sheet indicators, mostly using ratios, which constitute key elements of key prudential rules. The impact of these risk factors on the contributions (relative to the size of the national banking system) depends on the relative position of the banks concerned within the euro area. Table 5 shows the result of this risk profile calculations as published by the SRB. This data is available only for countries already part of the Banking Union (and thus under the SRB). It is apparent that most of the New Member States have a rather low-risk profile. The average of the three Baltic States and that of Slovakia would be only less than 15 %, less than half of that of the French banks.
The SRB risk profile for Bulgaria is obviously not yet available, but it should be towards the lower end of the range. This would imply that the contribution of the Bulgarian banks might be considerably less than 1 % of covered deposits.
4.2 EUROPEAN STABILITY MECHANISM (ESM)
By joining the euro Bulgaria will also be obliged to join the ESM. This has two implications: Bulgaria has to pay its share of the paid-in capital of the ESM and Bulgaria will have to guarantee a small share of the outstanding loans of the ESM (and the EFSF), including those to Greece. The first element, namely the contribution to the paid-in capital, should not be a major concern.
The rules of the ESM foresee for poorer Member States for 12 years a considerable reduction (by about 50%) of the regular share the country would have in the ESM, which, in turn, is equal to the share in the capital of the ECB.
Article 42 of the ESM Treaty foresees a temporary reduction (for 12 years) for the share of countries with a GNI per capita of less than 75 % of the EU average.
The formula for the reduction in the share is as follows: the reduced share is equal to a quarter of the normal share plus three-quarters of the country's share in the GDP of the euro area. For Bulgaria, an approximate calculation gives a reduction in the capital share to be paid up by about one third. The share of Bulgaria in the capital of the ECB is about 0.88 % (of the euro area countries total). One-quarter of that is equal to 0.22 %. The share of Bulgaria in the GDP of the euro area is about 0.5 %.
Three-quarters of that are equal to 0.37 %. These two elements together give a total share for Bulgaria in the paid-in capital of the ESM of about 0.6 %, or about one third less than the normal share. The total amount is about 500 mln. euro and it needs to be subscribed for a period of five years. Despite the lower paid-in capital, Bulgaria will have normal voting rights in all ESM decisions. *This article is an adapted version of "Bulgaria on the Road to the Euro and the Banking Union. Effects for Business" research paper
|Daniel Gros is Director of the Centre for European Policy Studies (CEPS) since 2000. He holds a Ph.D. in |
economics from the University of Chicago.
In the past Mr Gros worked at the IMF, collaborated with the European Commission as economic adviser to
the Delors Committee that developed plans for the EMU, and taught at several leading European Universities.
He has been member of high-level advisory bodies to the French and Belgian governments and was
provided advice to numerous central banks and governments, including Greece, the UK and the US, at the
highest political level.
Mr Gros is currently an adviser to the European Parliament and a member of the Advisory Scientific Council
(ASC) of the European Systemic Risk Board (ESRB).
He has published extensively on international economic affairs, including on issues related monetary and
fiscal policy, exchange rates, banking as well as climate change. He is the author of several books and editor
of Economie Internationale and International Finance. He contributes a globally syndicated column on
European economic issues to Project Syndicate.