Zdenek Kudrna

Financial crisis: testing the relationship between foreign banks and the new EU members

This paper discusses the impact of the recent financial crisis on the relationship between the domestic authorities of new member states and the foreign banks that dominate their financial sectors. Foreign banks played an adverse role during the unsustainable pre-crisis credit booms that rendered certain EU10 economies very vulnerable to crisis. However, they also supported the most affected EU10 economies through the acute phase of the crisis, when new capital and liquidity was essential for the success of domestic and international stabilization programs. It thus seems that the crisis test had proven the viability of the EU10 model of financial integration. However, there are also signs of heightened political risks that tend to be disregarded by economic analysis. This paper provides early evidence from Hungary and Latvia demonstrating that these risks are not entirely covered by EU membership. Domestic tax and regulatory policy changes may have a substantial impact on foreign subsidiaries in EU10 countries and thus strain mutual relationships more than expected. Therefore, it is yet to be seen whether the foreign-dominated banking model is resistant to the politics of hard times.

Introduction

The EU10 countries had developed a unique model of financial integration by selling almost all of their large banks to strategic foreign owners.[1] This strategy stabilized these banks after a turbulent period of economic transition and provided them with easy access to foreign financing. Foreign ownership of banks, however, has proved to be a mixed blessing in certain countries, where excessive credit growth translated into real estate bubbles and a large proportion of total loans denominated in foreign currencies. These two factors made certain EU10 economies and their subsidiaries very dependent on sustained flows of foreign financing provided by foreign parent banks. In turn, sustaining external financing proved a major challenge during the 2007 −2009 financial crisis, which required international policy coordination to prevent a systemic crisis in the most affected EU10 countries.

The financial crisis presented the first serious test of the interdependent relationship between the foreign financial groups and host EU10 countries since the relationship was formed during the previous decade. This paper reviews this experience and highlights the future challenges stemming from increased political risks that are often disregarded by economic analysis. It also provides early evidence from Hungary and Latvia, demonstrating that the heightened political risks that may result from such a scenario are not merely theoretical.
The review of the analytical literature reveals that, although the crisis was triggered by external factors, the macro-prudential domestic vulnerabilities such as pre-crisis credit growth and foreign currency-denominated lending determined the varied impact of the crisis on each EU10 economy. A meltdown – a combined banking and currency crisis – was successfully avoided in all EU10 economies through a combination of domestic and international measures, which were supported by sustained financing from parent banks. On this basis, international financial institutions concluded that the crisis was managed successfully and the relationship between domestic authorities and foreign banks had proved its viability throughout the crisis. They expect no change in the dynamics of this relationship, only minor adaptations of the banks’ business models that will reflect crisis consequences and regulatory reforms.

However, one of the unexpected consequences of the recent crisis may be increased political risk that foreign bankers will face in the EU10 economies most affected by the crisis. The economic analysis implicitly presumes that the political risks are fully covered by EU membership. However, the EU legal framework leaves policy space for the EU10 governments to enact tax or regulatory policies that need not be coordinated on the EU level. The post-crisis period of slow growth, high unemployment and fiscal austerity creates political incentives for shifting more of the crisis related burdens onto foreign banks, who were implicated in the unsustainable pre-crisis booms and generated considerable profits in the process. This paper provides early evidence from Hungary and Latvia demonstrating that increased political risks are not a mere theoretical possibility, but may have a material impact on the subsidiaries of foreign banks in EU10 countries. At the same time, it is too early to judge whether these examples are mere outliers or whether they set a new trend of less consensual relationships between foreign banks and host authorities in the new member states.

EU10 model of financial integration

The EU10 economies followed the same model of economic development at least since the mid-1990s period of their economic and political transition. The institutional core of the model was political integration into the European Union and gradual convergence with the EU laws and operational norms during the enlargement process. On the economic level, the key driver of transition reforms was trade integration with ‘old Europe’ as all EU10 countries are small open economies with their growth heavily dependent on demand for exports. The third pillar of the model was financial integration in the form of privatization of nearly all major banks to foreign strategic investors, primarily from the old EU member states. The degree of financial integration distinguished the EU10 model from that of other emerging economies around the globe, which retained greater domestic control over their banking sectors (EBRD 2009:62). The run up to the financial crisis and the subsequent crisis experience provided the first serious test of this model under adverse economic circumstances.

The EU-focused economic model helped the post-communist countries to fulfill the Copenhagen criteria necessary for EU membership, which was a widely shared political goal. The EU supported bank privatizations to foreign strategic investors as well, so the benefits of financial integration through strategic investors were rarely disputed before the crisis.[2] As a result, financial groups from the old EU member states came to dominate the banking sectors of the EU10 economies (see Unicredit 2009b:15 for overview).

The statistical evidence also generally supports the view that the financial integration made a positive contribution to growth of EU10 economies. The EBRD ran a series of tests of this proposition and found that although it is not a case in general for all emerging market economies, in case of the EU10 economies there is an observable positive effect of financial integration on economic growth (EBRD 2009: 68). Moreover, the corollary benefits of bank privatization to foreign strategic owners was the restructuring of formerly state-owned banks, many of which required repeated and fiscally costly bailouts during the transition period (Berglof and Bolton 2002).

Figure 1: Foreign ownership of EU10 banks

Source: EBRD database

The financial integration has been pursued with varied intensity, although the high proportion of foreign control is shared across the EU10 economies (Figure 1). In some countries, foreign banks contributed to the overheating of local economies by funding excessive credit growth and issuing much of these credits in foreign currencies (Figure 2). These two factors essentially split the EU10 countries into two groups. The Baltic countries, Romania and Bulgaria and Hungary[3] were more vulnerable to the sudden stop of capital inflows and, therefore, they were more affected by the financial crisis. Moreover, the high proportion of loans denominated in foreign currencies seriously constrained their macroeconomics responses to the crisis. The Czech Republic, Poland, Slovakia and Slovenia entered the crisis with less overeaten economy (Figure 2) and either entered the euro zone during the crisis or avoided excessive lending in foreign currencies. As a result, they were less vulnerable to the crisis and thus less affected.

Figure 2: Private sector credit growth and foreign currency denominated loans

Source: Real private sector credit growth IMF (2010:44), FX credit EBRD (2010b:7)
Notes: Slovak FX loans for 2008 from Tuma (2009). Slovenia joined euro before the crisis so the FX loan category was not directly relevant.

In retrospect, it is easy to establish that credit growth was excessive in some countries and cannot even be justified by the inevitable deepening of the financial sector during the first sustained period of growth after the transition (Bakker and Gulde 2010, EBRD 2009:68). The credit growth in the group of the less affected EU10 economies was slower, moreover much of the external capital inflows took the form of foreign direct investment into export-oriented sectors. This contrasts with the more affected economies where higher proportion of capital inflows fueled domestic consumption and real estate bubbles (Figure 3).

Figure 3: Non-FDI capital inflows and real estate prices, 2003 - 2008

Source: Author’s calculations based on IMF (2010).
Notes: Non-FDI cumulative capital inflows refer to residual capital inflows after subtraction of FDI. Real estate price increases are for 2003 to 2008, except for Baltic countries where the boom ended in 2007 and where figures cover the period of 2002-2007. Data on real estate prices for Romania not available.

The foreign banks are responsible for the failure of their risk management systems to contain the excessive credit boom. However, risk management systems were overwhelmed not only in the EU10 countries, but across the globe. The banking regulatory standards were known to be pro-cyclical,  particularly in situations where credit growth led to an increase in collateral values that in turn allowed banks to lend more against the same collateral. However, this issue was neglected during the global negotiations of the Basel II bank capital standards (Stiglitz 2010:107), thus the excessive credit boom in some EU10 countries is as much a failure of parent banks as that of the EU regulatory framework that incorporates global capital rules.[4]

The foreign banks as well as EU10 authorities were aware of the risks associated with fast credit growth. The Nordic banks started reining in the credit boom in the Baltic countries already in the summer 2007, before the onset of the financial crisis. They hoped to orchestrate a ‘soft-landing’, but the crisis made this impossible. Similarly, the foreign banks have joined the effort to contain credit boom in Bulgaria and implemented a series of cooling measures, including credit growth ceilings at individual banks (see IMF 2010c:58-59 for overview). However, these measures were only partially successful as credit flows shifted from banks to non-bank institutions such as leasing companies and to direct cross-border lending, often from parent banks. This experience had demonstrated that the domestic regulator tools are insufficient for containing a credit boom, when the free flow of capital within the single market provides local firms with access to loans from other EU countries. In any case, the cooling of measures implemented by foreign banks proved to be too little too late to contain pre-crisis vulnerabilities.

The large presence of foreign banks in the EU10 explains only the supply of foreign currency loans, but not the local demand. Most of the borrowers were attracted by lower interest rates on foreign denominated loans and were aware of the currency risk involved in taking such a loan.[5] However, in countries where the central banks were unable to keep inflation low and interest rates reasonably stable, borrowers considered the currency risk more acceptable than the risk of high and highly volatile domestic interest rates (EBRD 2009:70-71). This was further encouraged by the pre-crisis stability of EU10 exchange rates. Some currencies were explicitly pegged to the euro and several countries were expected to join the euro area soon.[6] In this context, a euro-denominated loan was a reasonable alternative for borrowers, although it collectively translated into a massive vulnerability on the macroeconomic level. As such, the currency risks should have been addressed by the central banks and regulatory authorities rather than individual banks. Thus the large proportion of loans denominated in foreign currencies again points to the broader deficiency of the pre-crisis EU regulatory framework that ignored potential ‘macro-prudential’ problems.

Testing the relationship in an acute crisis

The relationship between cross-border banks was thus tested more profoundly in the EU10 countries more affected by the crisis. Their authorities and parent banks operate in an interdependent relationship, which is mediated by the EU regulatory and crisis management frameworks. In normal times, the interests of domestic stakeholders and foreign banks are aligned as both benefit from an efficient, profitable and stable banking sector that finances economic development.  During crises, however, short term interests may clash as banks seek to minimize their potential losses and domestic stakeholders seek to ensure the flow of credit and raise tax revenue, for example. Such conflicts may put domestic governments and foreign banks into situations resembling a prisoner’s dilemma game, in which the best outcome results from close cooperation, but under certain circumstances it may be more advantageous for one of the parties to defect from cooperation, especially if other parties stay committed. [7] During the pre-crisis period, the relationships between parent banks and the EU10 countries were mutually beneficial. The EU10 economies had access to external funding, better banking know-how and improved services, without incurring the costs associated with a gradual build up of these capacities by domestically-owned banks. The parent banks, in turn, generated considerable profits from their investments. Nonetheless, during the acute phase of the crisis there were moments of high uncertainty when the commitment to mutually beneficial cooperation was tested.

The EU10 countries were sheltered from the initial impact of the financial crisis that proliferated through the rapidly falling prices of certain complex financial assets. Transnational banking aim at maximizing the economies of scale from their cross-border operations, which leads internal specialization within the group. The cross-border banking groups tend to concentrate investment activities either in one of the EU financial centers or at the home-country headquarters. The EU10 subsidiaries specialize in traditional retail and corporate banking businesses that require only limited direct involvement in complex international finance. Moreover, the traditional banking business in EU10 was highly profitable before the crisis and local subsidiaries had to import capital from parent banks to satisfy rapidly demand for credit. They had hardly any resources for investments into complex financial products. As a result, only a few of the EU10 banks experienced direct losses during the early phase of the crisis.

Moreover, the EU10 economies continued to grow until mid 2008, and it was not immediately obvious that the crisis would spill over from advanced financial markets into the emerging economies. At the time, the ‘decoupling hypothesis’ – which predicted that  advanced countries would experience recession but global growth would be sustained by emerging economies – seemed to apply to the EU10 as well. The flow of external financing continued and risk premiums on EU10 bonds did not indicate any imminent problems (IMF 2010: 46). Only Baltic countries were preparing for a gradual slowdown and Hungary was trying to consolidate its external debts, but these policies were not invoked in direct response to the crisis .

The financial crisis hit the EU10 economies with full impact only in the autumn of 2008. Following the bankruptcy of Lehman Brothers, the global financial markets froze, as did the financial flows to emerging markets. The most acute period of crisis lasted from September 2008 to March 2009; as financial markets were unable to distinguish between the more and less vulnerable EU10 countries, all were affected indiscriminately (Tuma 2009). Most withstood the immediate pressures on their own, but three EU10 governments had to turn to the IMF and EU for emergency stand-by loans. The stabilization policies delivered a degree of macroeconomic stability, which could, however, be undermined by discontinuity in private external financing that came primarily from parent banks.

Parent banks faced severe difficulties in their home markets during the crisis, and economic circumstances may have forced them to reconsider their commitments to financing their EU10 subsidiaries. However, their support was necessary to prevent a systemic crisis in the EU10 countries most affected by the crisis. Ensuring sustained financing during the most acute period of crisis, when parent banks faced simultaneous pressures at many markets, has been the most stringent test of the relationship between foreign banks and host authorities.

The severity of the crisis is observable in basic macroeconomic variables (Figure 3). The real GDP growth of all EU10 countries - except Poland - was negative in 2009, with Baltic economies shrinking at a rate rarely seen in peace time. The capital flows that fueled growth before the crisis fell dramatically, and exchange rates of the non-euro EU10 countries oscillated widely (see IMF 2010:51). All countries with fixed exchange rates managed to prevent outright currency crises, although this required the combined support of the IMF and EU in the case of Hungary, Latvia and Romania.[8]

Figure 4: Real GDP growth in EU10, 2009

Source: Unicredit (2009b:4)

There was no escape from the recession spilling over to banking sectors. The growth rate of credit to the private sector fell from its pre-crisis peak rates that in four cases exceeded 50 percent per year, to negative rates at the end of 2009 (World Bank 2010:14).[9] As many of the indebted households and firms could not cope with the macroeconomic shock and rising unemployment, they were not able to repay loans in time and thus the ratio of non-performing loans (NPL) started to rise. Initially, NPL ratios in all EU10 countries but Romania were below 5 percent at the beginning of 2008, but by the beginning of 2010 they had risen in all countries, but especially in Romania, Lithuania and Latvia (Figure 5). These ratios will continue to climb as losses proliferate from the real economy to banks’ balance sheets.

Figure 5: Peak-to-through change in credit growth and non-performing assets, 2009

Source: Credit growth rates from World Bank (2010:14) and NPL ratios from IMF (2010c:55)

Figure 5 demonstrates the relationship between the fall in the growth rate of credit to the private sector that fueled the pre-crisis boom and the rise in non-performing assets that followed the acute period of crisis. It shows that countries with the fastest expansion of credit before the crisis (and thus the largest difference between the pre-crisis peak rate of credit growth and post-crisis through rate) are also those that have experienced the steepest increases in non-performing assets. This is an indication that the pre-crisis boom was unsustainable and driven by inflated asset values. The figure also highlights that there are two groups of countries; the banking sectors of those in the lower-left part of the chart were less affected by the crisis, whereas those in the upper-right part were more affected.

Although credit growth turned from high positive to low negative numbers, the overall amount of credit to the EU10 private sector shrank much less than GDP (WB 2010:14). This shows that parent banks stayed committed to the EU10 economies throughout the crisis and despite difficulties in their home markets. This was no small achievement given the circumstances. Unlike their EU10 subsidiaries, parent financial groups were not spared the initial impact of the crisis and had to deal with drastic losses in their financial asset portfolios. These losses were exacerbated in September 2008, by the failure of Lehman Brothers, which increased uncertainty about the liquidity and solvency of financial market participants to such levels that private financial markets were brought to a standstill. Parent groups were in the midst of deleveraging, increasing their liquidity and orchestrating capital increases when the bad economic news from the EU10 countries began to accumulate.

As overheated growth turned into deep recession and the non-performing assets started to rise rapidly, some of the EU10 subsidiaries urgently needed new capital. They also needed access to liquidity as some of them faced considerable deposit withdrawals, especially in Baltic countries where banking sector deposits fell by over 10 percent between October 2008 and March 2009 (IMF 2010:50). Moreover, there was a risk of sharp devaluation in EU10 countries with high levels of foreign lending, where the losses could easily reach extraordinary proportions. Rating agencies woke up to the consequences of exposure of Italian, Swedish, Greek and Austrian banks to EU10 economies and questioned the impact on parent groups’ credit ratings (see Fitch 2009, Stokes 2009). The parent banks also faced pressure from stock markets, where their shares lost more value than shares of banks with less cross-border exposure (Unicredit 2009a).

The relationship between parent banks and their EU10 hosts was most brittle and tense between January and March of 2009. The financial media had speculated about potential withdrawal of western financial groups from new member states and Eastern Europe (see Aron 2009, Evans-Pritchard 2009, Stokes 2009, for example). These articles adopted the logic of a prisoner’s dilemma among foreign banks: the groups that left first were expected to face lower losses than those who remained committed for a longer period of time. The early leavers may have cut their losses by selling their EU10 assets while there was still a market for them, thus receiving a higher return than the fire sale price likely experienced by those who stayed put for too long. Some of these concerns were exaggerated, as parent banks were strategic, not mere portfolio investors (see also Unicredit 2009b: 4). They invested billions of Euros into acquiring market shares and building retail banking networks, thus they were unlikely to leave too quickly (Haeberle 2009). Moreover, such reports failed to distinguish between the gravity of the situation among different EU10 countries (Tuma 2009). Nonetheless, they revealed that the high uncertainty had created a crisis of confidence that could lead to overreactions undermining the financial stability of EU10 subsidiaries, their foreign parents and ultimately of the EU economies.[10]

A policy response to market concerns was needed. It could come either on a case-by-case basis or more systematically as a coordinated EU policy (Gros 2009). Austria lead an effort to design a systematic response and tested its political feasibility at the informal ECOFIN Council in March 2009. The idea did not receive sufficient support, as the problems of some EU10 banks were not of equal concern to all EU members (Haeberle 2009).[11] The Council  refused to consider a systematic EU solution and declared that the affected EU countries would be supported on a case-by-case basis.[12] Although the Austrian initiative was unsuccessful, its preparatory meetings led to the less formal ‘Vienna Initiative’ that was important for sustaining policy coordination and ensuring the commitment of parent banks to financing the most affected EU10 countries.
The Initiative was launched by the IMF, the European Bank for Reconstruction and Development (EBRD), and the Commission, all of which were concerned about the commitment of foreign banks to sustaining sufficient levels of external financing necessary to stabilize the most affected EU10 economies. The Initiative (also formally known as the European Bank Coordination Initiative) provided a forum for coordination and information sharing among international financial organizations (IMF, EBRD, EIB, the World Bank), the Commission (and the European Central Bank as an observer), home and host authorities of cross-border banking groups as well as the largest financial groups involved in the EU10. It was based on the mutual ‘quid pro quo’ when international and national authorities provided banks with detailed information on the course of macroeconomic stabilization and international reserves that were crucial for their risk assessment (IMF 2010:63). In exchange, banks committed to sustaining certain levels of external financing for their EU10 subsidiaries that supported the overall objectives of macroeconomic stabilization. To this end, specific agreements on liquidity support and recapitalization of local subsidiaries were signed for the countries most affected by the crisis, including Hungary, Romania and Latvia (EBRD 2010a:1).

Coordination was also important with regard to the legal feasibility. In the case that the parent bank received state aid from a home-country government, the home-country authorities had to ensure that the attached conditions did not include any protectionist provision that could prevent liquidity and capital support of banks’ foreign subsidiaries. Such restrictions could seriously constrain the flow of external financing to the most affected countries, because almost all important Austrian, Belgian, French and Greek banks active in the EU10 region received some form of financial support from their governments (Unicredit 2009b: 13). Similarly important was the participation of the Commission, which helped to ensure that the cooperation of competitors is consistent with the EU competition policy and that adopted measures comply with state aid rules.[13]
The Vienna Initiative also supported coordination with EU10 policy responses to the crisis. The host-countries were responsible for macroeconomic stabilization, liquidity support for local currencies irrespective of bank ownership and, if necessary, support for deposit insurance schemes. They fulfilled their obligations when such measures were called for. Half of the EU10 governments introduced specific support measures for their banks, although in the case of Poland and Slovakia they were not used at all (see IP/10/623). Across the EU27, about 32% of the total financial support made available for banks was utilized. The EU10 take-up rates in Latvia, Hungary and Slovenia were below this average (Table 1). This reflects the fact that bank subsidiaries in the EU10 were not directly affected by the sub-prime crisis that triggered the initial round of state aid. The relatively lower utilization of available state aids also demonstrates that macroeconomic stabilization was successful in containing the acute phase of the crisis.

Table 1: Bank support schemes in EU10 countries (EU approved aid in € bn)

Source: Commission (2010). The take up rates are from IP/10/623 and the Hungarian take up rate is estimated on the basis of information from C(2010) 91 final.

The macroeconomic stabilization policies that were supported by the IMF and EU in three cases, combined with domestic measures and support from the Vienna Initiative, prevented exaggerated market concerns from triggering a systemic banking crisis in any of the EU10 economies (see also Unicredit 2009b :5, EBRD 2010a: 2, IMF 2010: 61). These measures helped reduce capital outflows from the CEE region, which was relatively milder than in Latin America or emerging Asia (EBRD 2010a:2). Bank subsidiaries in the less affected countries such as the Czech Republic, Slovakia and Poland did not require any assistance from host governments or international financial organizations. To the contrary, their continued profitability throughout the crisis helped their parents to provide capital and liquidity support to subsidiaries in the more affected economies.[14] In the end, all parent banks honored their financing commitments and no local subsidiary went bankrupt.[15] As the immediate crisis of confidence subsided, parent banks continue to support those EU10 subsidiaries that need external financing. In short, the acute phase of the crisis was managed successfully by an ad hoc cooperation among EU10 authorities, parent banks and their home-country authorities, IFIs and the Commission.

Back to business as usual?

The EU10 markets remain attractive to foreign banks despite the crisis. Many of the foreign groups generated much of their pre-crisis profits in the EU10 countries, and although some of these profits were derived from unsustainable lending booms, much was a result of gradual economic convergence and financial deepening that are likely to continue.[16] Before the crisis, lending growth was generated primarily by loans to households that experienced the first sustained consumption and investment boom during the post-transition period. As a result, consumer loans have caught up with averages in the euro area, but penetration in other market segments is much lower; for example the mortgage market is at 8 percent of GDP compared to 38 percent in the EMU (Unicredit 2009b: 11). Similar potential exists in corporate lending as local firms shift from retained earnings and foreign direct investment to higher degrees of bank financing that is characteristic of the old EU economies. However, realization of these prospects is conditional on renewed demand for EU10 exports and macroeconomic stability.

Although the crisis did not alter the long term prospects of the EU10 markets, some adaptation of the bank business model will be required. The local subsidiaries of foreign banks will have to strengthen their capacity to resolve the non-performing assets accumulated during the crisis. This is an expensive process both in terms of provisions and write-offs as well as operational costs, thus making the banking in the EU10 markets more expensive than before the crisis. Moreover, there will be further adaptations stemming from the policy lessons of the crisis. Most notably, local banks should reduce the foreign currency lending and curb excessive credit growth during the upswing of economic cycle.

Reducing foreign denominated loans will require local banks to rely more on local deposits. However, if all banks compete for deposits that are likely to be stagnant during the immediate post-crisis period, it may trigger an intensive competition for deposits. Moreover, lending in local currencies is likely to shifts currency risks from households to banks, making it more complicated and expensive for banks. Therefore, banks may be tempted to sustain lending in foreign currencies, if there is demand for such loans. To prevent the recurrence of large-scale unhedged lending in foreign currencies, the non-euro EU10 countries may need to introduce some regulatory limits on such loans.

The excessive credit boom in some EU10 countries was but one indication of a broader problem of pro-cyclical bank risk management rules. Curbing this problem is on the G20 as well as the EU reform agenda, thus parent banks will be forced to adapt to global reforms of the capital adequacy rules. Moreover, on the EU level the credit booms will be watched by the European Systemic Risk Council (ESRC) that was introduced as part of the financial reform package approved in September 2010 (see Kudrna 2010). The ESRC has the power to require the home and host country authorities to act to prevent the re-emergence of unchecked credit booms.

Despite the adaptation pressure of the post-crisis period, it is not likely that any of the parent banking groups withdraws even from the most affected EU10 countries. Firstly, the slowdown will be observable in nearly all economies globally, so there will be no obvious alternative locations for investments. Secondly, parent financial groups are long-term strategic investors with considerable sunk investments, who maintained their commitment during the acute phase of the crisis, thus they are unlikely to depart during the post-crisis slowdown. Therefore, the financial analysts in general expect that the relationship between the foreign banks and their EU10 hosts will continue and there will be no changes to the pre-crisis arrangements, spare for the minor adaptations to the crisis experience and regulatory reforms.

Resurgence of political risks?

The aftermath of the crisis also creates political risks that are generally disregarded by the economic analysis. The excesses of the pre-crisis boom need to be paid for, which is a considerable task especially in the most affected EU10 economies. There will be a period of slow growth, high unemployment, fiscal austerity and increasing taxes as governments try to regain macroeconomic stability under less benign global economic circumstances. On the microeconomic level, unpaid loans have to be resolved, which often means bankruptcy for businesses and consumers, forced sale of their property and evictions from houses with unpaid mortgages. Inevitably, these measures pit the immediate interests of many local constituents against those of foreign-owned banks.

During the pre-crisis booms, the profits of local banks accrued for their foreign owners attracted limited political attention. The EU10 economic model relying on financial integration enjoyed widespread political support and, compared to the first decade of transition, banks were successfully restructured by their new owners, provided much better services, expanded their clientele and no longer required repeated taxpayer-financed bailouts (see Kudrna, Nollen and Pazdernik 2005). The crisis challenged the generally positive perceptions, the constructive role played by the foreign banks notwithstanding. As governments – especially those on IMF programs – cut public spending and increased taxes, banking profits attract more public scrutiny.[17] As elsewhere in Europe, the EU10 countries are considering additional taxes on banks and Hungary has already introduced them.

Hungary was one of the most vulnerable economies among the EU10 because of its high debt levels and dependence on foreign funding. Stabilization measures were introduced already in 2006, when the government began reducing public spending and introducing new taxes. Banks became liable for a 4% increase in a tax on profits as well as a special 5% solidarity tax on interest receipts from loans that enjoy direct or indirect subsidies from the state budget, such as home mortgages (Delloite 2009). As the global crisis deepened, Hungarian economic difficulties and the IMF/EU stabilization package required further rebalancing of the public budget. The new government elected in April 2010 returned to the idea of a special bank tax.

In July 2010, the parliament imposed a bank solidarity tax of 0.5% on assets of each large bank that is expected to yield some € 700 million in annual fiscal revenue (EurActiv, July 6, 2010). This tax comes on top of all other taxes, including the 2006 tax increases. [18] Bank solidarity taxes were discussed and introduced in other countries, but never at such a high rate. Sweden had introduced one at 0.036% of assets, Austria approved  a 0.07% tax and discussions in the UK or US indicate a future solidarity tax in the range of 0.07 to 0.15 percent (Bloomberg, Jul 22, 2010). This makes the Hungarian rate 3 to 12 times higher than similar taxes elsewhere. The tax will be implemented for three years and its proceeds will go to support the general government budget, not a bank resolution fund as suggested by the Commission (see COM(2010) 254 final). Moreover, the tax does not apply equally to all banks, but only to those whose assets exceed a certain threshold. The eight largest banks, which are coincidently owned by foreign investors, are expected to contribute 90% of the total tax revenue (RGE 2010).[19]

The bank solidarity tax in this form has been opposed by parent banks, IMF representative, EU Commissioner as well as the Hungarian central bank. The European Banking Federation called for a modification of the discriminative tax proposal and pointed out that the combined tax burden levy would push some banks into losses (Bloomberg, June 25, 2010). These losses may undermine banks’ capital base and limit their ability to finance economic recovery without additional capital from parent banks. However, additional capital from parent banks or other investors is unlikely to come as it would increase the tax liability. At the meeting of the Vienna Initiative in June 2010, parent banks warned that they may abandon certain activities to reduce their assets in Hungary (Austrian Independent, July 7, 2010). The head of the IMF’s mission to Hungary argued for more durable and non-distortive measures that would have less negative effects on the economy and solve Hungary’s fiscal problems (Bloomberg, July 22, 2010). The Hungarian minister of finance also acknowledged that the EU Monetary Commissioner had “serious doubts”, and the EU has expressed fears that other countries, including Romania and Slovakia, may follow suit (Bloomberg, July 2, 2010, EurArchiv, July 6, 2010). The monetary council of the Hungarian National Bank added to the chorus of criticisms, stating that the “amount the Government aims to collect from domestic financial intermediaries by means of a special bank tax may impair the ability of the banking sector to attract capital and its capacity to lend, which in turn may result in significant output loss in the short and long term. The drain on bank earnings through the planned levy may undermine the ability of the domestic banking sector to collect funds and, ultimately, the stability of the Hungarian economy” (MNB, July 5, 2010).
The government maintained that the extraordinary financial sector tax is the only alternative to protracted austerity measures and although it may sacrifice some economic growth, the long term austerity package would have an even deeper negative impact (Budapest Business Journal, July 19, 2010). The government, which enjoys a two-thirds majority in parliament, has since introduced additional taxes on other sectors receiving substantial foreign-investment such as energy industries, telecommunication, retail and distribution. It also postponed talks with the IMF and the EU about the continuation of their financial support for macroeconomic stabilization in Hungary (IMF 2010).

The targeted tax increases stretch the amicable relationship between foreign investors, international organizations and Hungary into a more conflictual territory than at any time during the last two decades. It is too early to tell what the impact of the new taxes will be and whether the government predictions or those of their critics will more closely resemble the eventual economic developments. It is not yet clear whether foreign financial groups will actually change their strategy and limit their exposure, or whether such a move will, in fact, undermine the supply of credit necessary for economic recovery. Similarly, it is not clear whether Hungary is an exception or whether it sets a new trend in the relationship between foreign bankers and EU10 host countries. What this example clearly indicates, however, is that the distribution of the burdens created by the crisis is a contested issue that strains the relations of EU10 authorities and foreign-owned banks.

The policy space of the EU10 countries is constrained by the EU law defining the principles of the single market in financial services. This law provides protection to foreign investments in the financial sector and prevents EU members from imposing measures such as capital controls that could prevent the exit of foreign banks from any economy. Tax policy is one of the domains where member states prevented any substantive reduction of their sovereignty by EU law. It provides the most flexible policy tools for member governments wishing to impose a higher proportion of the crisis burden on foreign banks. However, there are more technical measures that can have a material impact on the distribution of burdens between foreign banks and the local population. An example of such a measure is bankruptcy legislation.

The Latvian parliament passed a new personal bankruptcy law at the same time as Hungary introduced the special bank tax. The law significantly strengthens the position of debtors relative to creditors and is thus bad news for Latvian banks that will have to absorb a higher proportion of losses stemming from the collapsed real estate bubble that they helped to fuel. The law allows borrowers declared bankrupt to write off their debt after one year if they repay 50 percent of what they owe, after two years if they repay 35 percent and after three year if they repay only 20 percent. In practical terms, debtors are only liable to pay back the market value of their collateral rather than the actual amount of the loan (Danske Bank 2010:1). The law also requires banks to cover the administrative costs of the personal bankruptcy procedure (Baltic Reports, July 27, 2010).
Unlike the Hungarian case, the law was prepared in consultation with parent banks as well as the IMF and EU representatives overseeing the Latvian stabilization program. An earlier version was passed by parliament in June 2010, but was vetoed by the Latvian president who considered some of the banks’ objections. The initial proposal would have allowed debt to be forgiven by paying 30 percent of a debtor’s income for two years (Bloomberg, July 27, 2010), and made no distinction between people who had taken loans for their own property and those who had speculated on the real estate market (Baltic Reports, July 27, 2010). The version signed into law is seen by Nordic parent banks as a compromise solution (Reuters, July 26, 2010). Although the law pushes a higher proportion of mortgage resolution costs onto banks, it is not expected to have any immediate effect on Latvian banks or their Nordic parents. Swedbank, SEB, Nordea and DnB NORD, which dominate the Latvian market, have already written off over €900 million in bad loans, including many of those covered by the new legislation (Bloomberg, July 27, 2010). However, the limits placed on the recovery process will make loans more risky for banks. They will require borrowers to put up more of their own funds in the future, which may prolong the ongoing credit crunch and thus the current economic recovery (Danske Bank 2010:1). At the same time, the new law will make a credit boom of the pre-crisis levels less likely in the distant future.

As in the case of the Hungarian special tax, the new Latvian bankruptcy legislation was passed nearly unanimously in the parliament and was voted in by both the government and opposition parties. This indicates that measures that shift the burdens of crisis to banks potentially enjoy support across the political spectrum, which may invite further emulation in other countries. To date there are merely two examples of such measures; but if the economic recovery proves to be slow and accompanied with persistent unemployment, with more people and firms going bankrupt, then similar legal or tax reforms may follow even in the less affected EU10 countries.

Conclusion

The crisis tested the relationship between foreign-owned banks and their host countries in the EU. During the pre-crisis period, the EU10 countries pursued the same model of financial integration with the old EU countries, but with varied intensity. The Baltic countries, Hungary, Bulgaria and Romania became dependent on external financing to a greater extent than the other countries, which made them more vulnerable to the consequences of the global financial crisis. Higher vulnerability translated into deeper recessions and greater risk of banking crises. The latter has been avoided through cooperation among the host authorities in the EU10 countries, the home authorities of the parent banks, parent banks themselves and the EU and international financial institutions. Although this cooperation was largely ad hoc, it was sufficient to prevent escalation into a systemic crisis during the acute period of crisis in late 2008 and early 2009.

The aftermath of the crisis presents new challenges, particularly in the most affected countries. As the examples of Hungary and Latvia have shown, the distribution of the burdens of the crisis often puts local parliaments on a collision course with foreign banks. Although banks and EU10 governments have shown a considerable capacity to cooperate before and during the crisis, protracted economic recession may force them to adopt less consensual policies. At the same time, neither the Hungarian tax nor the Latvian bankruptcy laws damages the interests of foreign banks to such an extent that they would reconsider their presence in EU10 markets. Their subsidiaries are more likely to adapt at the margin by reducing or slowing down certain auxiliary activities to reduce the impact of new measures on their financial results.

The post-crisis changes in the policy environment increase the political risk that foreign bankers face in the EU10 economies most affected by the crisis. The economic analysis implicitly presumes that the political risks are fully covered by the EU membership, but as the Hungarian and Latvian examples indicate there is a scope for more stringent policies. The post-crisis period of slow growth, high unemployment and fiscal austerity, creates political incentives for shifting more of the crisis related burdens onto foreign-owned banks. Whether these examples are mere outliers or whether they set a new trend of less consensual relationship between foreign banks and host authorities in the new member state remains to be seen.

Notes

[1]     The EU10 refers to the 10 new EU members viewed as emerging economies by the financial investors, i.e. Bulgaria, Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Romania, Slovakia and Slovenia.

[2]    The EU support for the bank privatization is easily observable from the pre-accession progress reports summarizing achievements and challenges for each candidate country during the 1999 to 2004 (2007 in case of Bulgaria and Romania) period.  Slovenia, which privatized its banks to a limited extent, was often criticized for this policy standpoint (Commission 2003:8).

[3]     The case of Hungary is somewhat specific. It started the process of macroeconomic consolidation caused by high fiscal deficits and debts already in 2006. Therefore, there was less of the pre-crisis overheating, which it shares with the group of less affected countries. On the other hand, the high proportion of foreign currency loans kept Hungary vulnerable.

[4]     Given that the Capital Requirements Directive implemented the Basel II capital standards, it points out to the global regulatory failure.

[5]     The unhedged currency risk born by households and firms with no income in foreign currencies means that if the value of the local currency falls against the euro, Swiss franc or yen in which the loan was taken, the outstanding debt repayments will rise overnight by the same proportion. Large currency swings can thus bankrupt many households and firms. Moreover, countries with high proportion of foreign currency loans cannot let their currencies devalue without many bankrupting households and firms, which is a serious constraint on macro-economic policies used to counter the crisis.

[6]     Three EU10 countries - Slovenia, Slovakia and Estonia - switched to euro by 2011.

[7]     An example of such defection on part of international banks was reliance of foreign banks on the IMF bail outs of East Asian economies during the 1997 crisis. These countries paid most of their liabilities towards international banks with IMF loans that they have to repay over time. This allowed the banks to escape any losses that were shifted fully on the local taxpayers (see Stiglitz 2002, for detailed argument). An example of defection on part of the domestic government would be punitive taxation or even outright nationalization of profitable foreign bank in order to raise revenue in the short term at the expense of the longer term financial integration with the rest of the world.

[8]     Hungary was the first to request IMF support in November 2008 due to a drying up of  domestic bond markets caused by the outflow of portfolio investments, which made it difficult to auction government securities and parallel difficulties encountered by banks in accessing the foreign exchange swap market (IMF 2010:50, 57). Latvia followed suit in December 2008, when its access to external finance fell very sharply due to the freezing of global markets, downgrades of its sovereign credit rating and the collapse of the second largest bank – the domestically owned - Parex Bank. Romania followed in May 2009 after a continued decline of its international reserves (IMF 2010:50).

[9]     Only in Bulgaria the credit growth remained positive although also falling by 62 percentage points from its peak rate of 64 percent in the first quarter on 2008 to its through rate of 2 percent in the second quarter of 2010.

[10]     The seriousness and fragility of the situation was well illustrated by several occasions when an offhand remark of a prominent economist such as Paul Krugman or faulty number in an IMF report triggered large sales of Austrian assets based on the theory that large exposure of Austrian banks to Central and Eastern Europe could bankrupt Austria in similar manner as Iceland. Cordero (2009:5) reviews some of these media exchanges that had material impact on the new and old EU members.

[11]     It did not help that the initial proposal also included support for non-EU Eastern European countries.

[12]     A more systematic response to the problems on EU periphery was adopted only year later, in response to the Greek crisis that lead the Council to create the European Financial Stability Mechanism.

[13]     The Commission adopted several communications in 2008 and 2009 clarifying application of state aid rules to measures supporting the EU banking sector (see summary in Commission 2010:3-4). All measures in EU10 were eventually approved by the DG Competition, although initially the bank aids were not properly notified to the Commission (see, for example, C (2010) 91 final.

[14]     In addition, the EU10 subsidiaries may help the parent banks to repay some of the state aid provided by their home-country governments. For example, the KBC plans to sell part of its stakes in Czech and Slovak bank to repay the aid to Belgian government.

[15]     The only large EU10 bank going under during the crisis is Latvian Parex Bank that was owned locally. Parex had liquidity problems and faced large deposit outflows that were partially motivated by the introduction of a blanket guarantee on all deposits in Swedish banks that made their Latvian subsidiaries much safer for depositors (Mayes 2009).

[16]     That not all profits from the EU10 economies were due to bubbles is demonstrated by the fact that banks were very profitable even in the countries such as the Czech Republic, Poland or Slovakia, which did not experience unsustainable credit booms before the crisis.

[17]     With the exception of the Lithuania and Latvia, the EU10 banking sectors regained profitability already in the Spring or summer of 2009 (MNB 2010:23).

[18]     Unlike the usual taxes on profits, the taxes on assets need to be paid regardless of whether the bank is making profits or losses in a given accounting period. Moreover, the tax at this rate would shift half of all returns generated by the banking sector to the public budget as average return on assets of Hungarian banks was about 1 percent at the end of 2009.

[19]    The largest Hungarian bank OTP is one of the few EU10 banks not owned by the foreign financial groups, but majority of its shares is held by dispersed group of foreign owners.

Bibliography

Arons, Steven (2009) Weathering the Storm. Corporate Finance in Emerging Europe 1(2): 6-10.

Bakker, Bas B.; Anne M. Gulde (2010) The Credit Boom in the EU New Member States: Bad Luck or Bad Policies? IMF Working Paper 10/130.

Berglof, Erik; Patrick Bolton (2002) The Great Divide and Beyond: Financial Architecture in Transition. Journal of Economic Perspectives (16)1: 77-100.

Commission (2003) Comprehensive Monitoring Report on Slovenia’s Preparations for Membership. Available at «http://ec.europa.eu/enlargement/archives/pdf/key_documents/2003/cmr_si_final_en.pdf»

Commission (2010) State Aid: Latest Scoreboard Shows Reduced Use of Crisis Support to Banks. Available at «http://europa.eu/rapid/pressReleasesAction.do?reference=IP/10/623&format=HTML&aged=0&language=EN&guiLanguage=en»

Cordero, Jose A. (2009) The IMF’s Stand-by Arrangements and the Economic Downturn in Eastern Europe: the Cases of Hungary, Latvia, and Ukraine. Washington: Center for Economic and Policy.

Danske Bank (2010) EMEA Daily. Danske Research, July 27.

Delloite (2009) International Tax and Business Guide: Hungary. Budapest: Deloitte Touche Tohmatsu.

EBRD (2009) Transition Report: Transition in Crisis? London: European Bank for Reconstruction and Development.

EBRD (2010a) Vienna Initiative. Available at «http://www.ebrd.com/downloads/research/factsheets/viennainitiative.pdf»

EBRD (2010b) Regional Economic Prospects in EBRD Countries of Operation: May 2010 Update. London: European Bank for Reconstruction and Development.

EBRD (2010c) Transition Report 2010: Recovery and Reform. Introductory Presentation by Chief Economist available at «http://www.ebrd.com/downloads/research/transition/tr10p.pdf»

Evans-Pritchard, Ambrose (2009) Failure to Save East Europe Will Lead to Worldwide Meltdown. Telegraph, February 14.

Fitch (2009) Major Western European Banks’ Exposure to Eastern Europe and the CIS - Downside Risk Contained? Fitch Ratings, April 16.

Gros, Daniel (2009) Collapse in Eastern Europe? The Rationale for a European Financial Stability. Voxeu.org, February 25.

Haeberle, Armin (2009) Not Exactly Stimulating. Corporate Finance in Emerging Europe 1(2): 11.

IMF (2010) Regional Economic Outlook: Europe -building Confidence. Washington: International Monetary Fund.

Kudrna, Zdenek (2010) Cross-border Resolution of Failed Banks in the EU: A Search for the Second-best Policies. Institute for European Integration Research Working Paper no. 8, November.

Kudrna, Zdenek; Stanley Nollen; Roman Pazdernik (2005) The Troubled Transition of Czech Banks to Competitive Markets. Post-Communist Economies 17(3): 363-380.

MNB (2010) Report on Financial Stability: Presentation. Budapest, Magyar Nemzeti Bank.

Mayes, David G. (2009) Did Recent Experience of a Financial Crisis Help in Coping with the Current Financial Turmoil? The Case of the Nordic Countries. Journal of Common Market Studies 47(5): 997-1015.

RGE (2010) Hungary's Controversial Bank Tax. Roubini Global Economics, October 26.

Stiglitz, Joseph (2002) Globalization and Its Discontents. New York: W.W. Norton and Company.

Stiglitz, Joseph (2010) The Stiglitz Report: Reforming the International Monetary and Financial Systems in the Wake of the Global Crisis. New York: The New Press.

Stokes, Mary (2009) Eastern Europe: The Makings of a Cross-Border Banking Nightmare? Roubini Global Economics, February 23.

Tuma, Zdenek (2009) Crisis Propagation in the Czech Economy and CEE Region. St Louis: NABE Annual Meetings presentation, 13 October.

Unicredit (2009a) Cross-Border Banking in Europe: What Regulation and Supervision? Available at«
www.unicreditgroup.eu/ucg-static/downloads/cross_border_banking_discussion_paper.pdf».

Unicredit (2009b) CEE Banking Outlook: Risk Appetite Crucial to Win the Upside. Vienna: Unicredit Group.

World Bank (2010) EU10 Regular Economic Report: Safeguarding Recovery. Washington: The World Bank.