Risks Rising in the Baltic States
Recent market volatility in Latvia on the back of market rumours of a possible devaluation served as a reminder of growing external vulnerabilities in the Baltic states. Fitch Ratings believes that delays to euro adoption, coupled with sustained inflationary pressures and rising external imbalances, act to constraint further rating upgrades for Estonia, Lithuania and Latvia; and a failure to address overheating could lead to future negative rating actions. The conclusions of this report are:
The Baltic states have enjoyed years of strong and sustained growth that has raised per capita income, hastening convergence with average EU levels. They have been among the four fastest-growing EU countries since 2001: each country’s real GDP growth rate averaged 8-9% over the five years to 2006. Nominal US dollar-denominated income has more than doubled since 2000.
However, Estonia, Latvia and Lithuania are, to varying degrees, facing rising inflation, tightening labour markets, rapid credit growth and substantial current account deficits, indicating that their economies are overheating. The deterioration in price stability has pushed back the euro adoption timetables for all three countries and Fitch expects inflationary pressures – particularly the requirement to adopt EU excise duties and raise administrative prices; and higher gas import prices – to be sustained in the medium term. High productivity growth in the tradable sector could also be contributing to higher inflation (the Balassa-Samuelson effect). Latvia and Estonia had the highest 12-month average inflation rates in the EU at end-2006 with rates of 6.6% and 4.4% respectively.
Furthermore, the Baltic countries’ currency pegs and the inflation differential with the rest of the EU contribute to ‘perverse’ pro-cyclical low real interest rates (as nominal interest rate convergence has largely been attained), spurring the economic and asset price boom in these countries. Fast-paced growth and labour emigration have driven unemployment rates to multi-year lows, adding to wage pressures. Estonia, Latvia and Lithuania had unemployment rates of 5.6%, 6.9% and 5.9% respectively at end-2006, all below the EU average of 7.9%.
Growth levels far above the EU average have been financed for the most part through the rapid rise of external liabilities, which has fuelled the increase in external imbalances. Fitch estimates Latvia’s current account deficit widened to almost 22% in 2006 from 13% in 2005, largely as a result of continued rapid import growth fuelled by the consumer boom coupled with a slowdown in exports. Foreign direct investment is still an important source of external financing for the Baltic countries, although FDI covered only 36% of Latvia’s current account deficit in 2006 and 41% of Lithuania’s (17% excluding privatisation revenues of US$0.8bn from the sale of the government’s 30.7% stake in Mazeikiu nafta to PKN Orlen). The bulk of the financing for the current account deficit is provided by financial flows from foreign parent banks to their Baltic subsidiaries.
All three countries’ gross external debt levels were above the A-range median in both 2005 and 2006. Latvia’s gross external debt burden rose to over 100% in 2006. The proportion of short-term external liabilities is also considerable, raising near-term financing risks. As a result, the Baltic countries’ near-term external position generally appears weaker than the A-range median and the Central and Eastern Europe (CEE) four: Hungary, Poland, Czech Republic and Slovakia. Furthermore, the Baltic countries’ deteriorating external imbalances are increasingly inviting comparisons with pre-crisis Mexico or Asia. The ratio of short-term external debt to foreign-exchange reserves was a good leading indicator of the Asian crisis and is comparatively high in the Baltic states, particularly in Latvia.
Nevertheless, Fitch believes there are important factors that mitigate the negative pressure that growing external imbalances place on the ratings. Crucially, the growth in external debt stocks is the result of private-sector rather than sovereign borrowing. All three Baltic countries have had public external debt levels below 20% of GDP since at least 1993 and all are net public external creditors. Low public debt levels (in addition to small budget deficits, or a budget surplus in the case of Estonia) also allow the Baltic countries scope to cushion any cyclical downturn. Estonian and Latvian banks and other private-sector enterprises, on the other hand, had built up gross external debt of almost 85% and 100% of GDP respectively in 2006, while it is lower at 60% of GDP in Lithuania.
Flexible labour and product markets help to ensure that countries can adjust to shocks. The Baltic countries have made significant progress in implementing structural reforms and their flexible private-sector-driven economies compare favourably with other transition countries as well as to a number of other EU members: the World Bank’s 2007 Doing Business report ranked Lithuania, Estonia and Latvia 16th, 17th and 24th respectively for ease of doing business. The Baltics also score well on the World Economic Forum’s Global Competitiveness Index: Estonia, Latvia and Lithuania rank 25th, 36th and 40th respectively compared with Italy (AA-; ranked 42nd) and Greece (A; ranked 47th).
Fitch notes that the ratio of short-term external liabilities to official reserves is particularly high in Latvia due to the large volume of non-resident deposits in the banking system (equivalent to 27% of GDP in Q306), reflecting the role of some banks as a conduit to transfer payments for corporates between Russia and other CIS countries and the West. Non-resident funding is very short-term and the outflow of these deposits in response to a shock (such as concerns over money-laundering) could prompt an abrupt correction to credit growth or difficulties for Latvia’s second-largest bank, Parex banka (BB+/stable), whose non-resident funding makes up over half of its deposit base. Nevertheless, the bulk of these deposits have proved stable in the past (for example, through the Russian crisis of the late 1990s) and Latvia’s attractiveness as a country for Russian and other CIS-based depositors to bank in – an EU country with many Russian-speakers – increases the likelihood that they remain so.
Furthermore, almost all of banking system assets in Estonia and Lithuania, and over half in Latvia are foreign owned, reducing the contingent liability to the sovereign. Much of the short-term external debt across the banking systems is inter-company loans from foreign parent banks and does not involve as high re-financing risk as market debt. Parent banks are likely to be long-term strategic investors who would only cut credit lines to their subsidiaries as a last resort (for example, SEB and Swedbank’s Baltic subsidiaries all enjoy support ratings of 1).
The large current account deficits in the Baltic countries are in part validated by strong growth and high rates of investment and are an unavoidable constituent of the real convergence story. Gross investment to GDP was 39% in Estonia and 38% in Latvia in 2006, the highest among A-range sovereigns with the exception of China. This ratio was 27% in Lithuania, above the A-range median of 25%. Nevertheless, Fitch notes that investment is not necessarily channelled to the tradables sector and that investment in residential housing, for instance, is ‘non-productive’ in that it does not boost exports or the country’s capacity to correct external imbalances and repay external debt. However, the Baltic countries do not appear to have lost export market share.
Another pressing argument for the sustainability of imbalances in the Baltics is the ‘exit strategy’ in the form of euro adoption. However long the delay, the Baltic countries are obliged through the requirements of EU membership to adopt the euro, suggesting that comparison with pre-euro Portugal, Spain and Greece rather than with pre-crisis Asia may be more appropriate. Fitch takes comfort from the fact that Lithuania and Estonia’s ratios of short-term external debt to foreign reserves have been consistently lower than Portugal’s during the latter’s run-up to euro adoption in 1999. Nevertheless, the agency notes that the support to stretched external finances from euro adoption is being weakened as eurozone entry dates slip further into the future.
Baltic banks’ foreign borrowing reflects the rapid pace of credit growth to the private sector. Lending to the private sector grew in nominal terms by over 50% in all three Baltic countries in 2005, the highest rate in the EU by a significant margin, and Fitch expects credit growth to have remained above 50% in 2006. The rapid pace of lending to the private sector has been sustained, averaging above 35% in each country from 2001-2006. The lending boom is being supported by the unlocking of credit constraints, which is amplifying a perverse real interest rate effect.
Fitch is concerned that the rapid growth in lending could lead to a rise in credit risk, which would only emerge in the event of a downturn. In addition, Baltic banks face market-risk-related credit risk as the high level of euroisation – spurred by stable currency regimes and anticipation of euro adoption – exposes unhedged borrowers to currency risk, increasing banks’ vulnerability to widespread default in the event of devaluation. In Latvia, loans denominated in euros grew by 65% in the first three quarters of 2006 compared with growth of just 18% in lats-denominated loans. Over three-quarters of outstanding loans in Latvia and Estonia, and just over half in Lithuania, are denominated in foreign currencies.
Furthermore, Fitch notes that mortgage lending is spearheading credit growth in the Baltics. While this type of lending typically has a more favourable risk profile, the Baltic countries have been experiencing a boom in property prices, raising the question of whether they are experiencing a real estate bubble. Property prices in Vilnius are more expensive in EUR/square metre than in Copenhagen, Stockholm or Berlin, while Riga is more expensive than Vienna or Frankfurt. A severe fall in property prices would weaken the contribution that secured lending makes to mitigating credit risk. Property prices in Lithuania stabilised following the EU Council’s decision on euro adoption but they have continued to grow rapidly in Latvia (by around 60% in 2006). The high level of foreign ownership in Baltic banks has helped improve the domestic banks’ standalone risk profile, particularly by improving their risk management systems.
Although the pace of credit growth appears unsustainably high, ascertaining the extent to which it deviates from trend is difficult in the Baltic countries, which have yet to experience a downturn in the credit cycle. Furthermore, judging whether the pace of growth is excessive is related to the equilibrium level of private credit to GDP. Credit to the private sector is expected to grow at a faster rate than nominal GDP in emerging markets. This process of ‘financial deepening’ could be considered part of the Baltic economies’ transition and convergence to average EU levels of income. Credit growth in the Baltics has been from a low base and credit-to-GDP levels are still below the A-range median and the EU15 average of 76% and 128% respectively. Nevertheless, some studies have estimated that Estonia and Latvia may be approaching the equilibrium level of private credit for their level of economic development.
Comparing the Baltic countries’ macroeconomic indicators with those of pre-crisis Asia leads to the consideration of whether they are vulnerable to the type of financial market contagion that contributed to the spread of the crisis in Asia. Investors who lose money in CEE or more specifically one Baltic country could look to close their positions across the Baltics, triggering a halt to or even an outflow of foreign funds from Baltic banks. This would be likely to lead to an abrupt slowdown in lending and consequently a severe correction in the growth rate and, potentially, asset prices.
‘Psychological’ contagion, by which markets draw parallels between economic and financial trends across countries and react accordingly – as seen in the Asian crisis and the emerging-markets sell-off in May 2006 – is a risk in the Baltics. Nevertheless, Fitch believes contagion risk from ‘direct’ trade or financial linkages is very limited. A brief survey of the volume of banking assets that the large and mostly western European banking groups hold in CEE and the Baltics reveals that these account for at most 30% of total group assets (with the exception of Raiffeisen Zentralbank Oesterreich). While Fitch notes that this ratio will rise as rapid credit growth continues in the Baltics, the relatively contained level of exposure to credit in transition economies is a source of reassurance.
While Swedbank owns over half of banking system assets in Estonia, for example, its total exposure to the Baltic countries made up just 13% of its assets at end-2006. Furthermore, the source of foreign ownership provides a degree of diversification: it is Scandinavian banks that appear to have invested in Baltic banking assets while western European banks (for example Erste Bank der osterreichischen Sparkassen, Raiffeisen Zentralbank Oesterreich and Unicredito) hold banking assets in CEE. This reduces the risk of contagion between the Baltics and other countries in CEE.
Adopting the euro would make the risk of external financing and currency crises negligible for the Baltics and as such is an ‘exit strategy’ from their exchange rate regimes (fixed peg to the euro in Latvia; euro-based currency board arrangements in Lithuania and Estonia). Under these currency regimes, a negative shock that leads to an outflow of foreign funds (such as foreign parent banks cutting credit lines to their subsidiaries) could put downward pressure on the exchange rate, which in turn could reduce the ability of FX-denominated debt-holders to repay, triggering a credit shock.
Joining the eurozone would eliminate this risk and the Baltics are benefiting from the market expectation that they will adopt the common currency. However, high inflation levels have forced all three Baltic nations to delay EMU membership while continued latent inflationary pressures within the economies led their governments to refrain from adopting a new concrete target date for joining the euro zone.
Lithuania’s bid to join EMU in 2007 was rejected by EU finance ministers in mid-2006 on the grounds that it had not fulfilled the Maastricht inflation criterion. Lithuania’s 12-month average inflation rate was 0.06% above the reference rate in March 2006, when the assessment was made. In October 2006, Estonia officially abandoned its target date of 2008 (Fitch revised the outlook on Estonia’s foreign currency issuer default rating (IDR) to stable from positive in August 2006 on the basis that Estonia would have to revise its euro adoption timetable).
The prime minister has stated that he ‘hopes’ the country will be able to adopt the common currency in 2010. Fitch revised the outlook on Latvia’s foreign currency IDR from Positive to Stable in September 2005 as rising inflation brought uncertainty to the government’s January 2008 target date for joining the eurozone. Latent inflationary pressures – particularly higher expected energy costsand the requirement to adopt EU excise duties – have prevented the Latvian and Lithuanian governments from announcing a new concrete target date for adopting the euro and Fitch does not expect them to join EMU before 2010.
The Baltic nations’ exchange rate regimes mean there is no independent exchange rate policy and only limited scope for monetary policy to tackle inflationary pressures. Central banks have responded to rising inflation by introducing tighter bank prudential measures: Latvia raised its policy rates by 100bp; Estonia increased the risk weights on mortgage lending; and both countries have raised their reserve requirements. But these have had only a temporary effect in stemming exuberant lending and can have unwanted consequences, such as increasing off-balance-sheet lending, encouraging further euroisation or the proliferation of less well-regulated non-bank financial institutions. Rising ECB rates will help contain inflationary pressures although euro-area tightening is likely to be insufficient to curb buoyant consumer demand in the Baltics.
Despite the authorities’ limited ability to use monetary policy tools and their relative impotence, Fitch believes there could be scope for raising the reference rate in Latvia if pressure on the lats persists. The agency also welcomes the Latvian authorities’ recent commitment to re-instate the limits on banks’ net open positions in euro.
The onus to tackle inflationary pressures therefore falls largely on fiscal policy, although here the Baltics do not have a much room to manoeuvre, as they already have low budget deficits (or a budget surplus in the case of Estonia). Furthermore, the desire to pursue real convergence with Europe is making the Baltic governments reluctant to take fiscal measures aimed at curbing consumer demand. Per capita income and wages in the Baltics are still among the lowest in the EU and there is a powerful belief among policymakers that growth should not be sacrificed for the sake of nominal convergence with the Maastricht criteria even at the cost of a delay to euro adoption. Furthermore, as new member states join the EU, the funds available to the Baltic countries are likely to diminish, making it politically difficult to constrain demand by slowing the implementation of EU-funded projects, as has been suggested by the IMF. Moreover, the risk of labour emigration also argues for pursuing rapid convergence.
Fitch notes that a proposal to reduce the personal income tax rate from 25% to 15% was shelved by Latvia’s re-elected government, although the government has also stated that it does not plan to balance its budget before 2010. However, in the light of recent developments with the currency, Fitch believes the working group set up to propose anti-inflationary measures has a better chance of prompting a shift in policy. The agency considers that the adoption of fiscal measures such as a tax on gains from real estate sales would send an important signal regarding the government’s commitment to tackling inflation.
The combination of the Baltic countries’ exchange rate regimes and their significant external imbalances leaves them vulnerable to an adjustment in capital and financial flows and slowdown in economic growth. This adjustment could involve a loss of market confidence, leading to an abrupt correction to financing, which would put downward pressure on the Baltic currencies and potentially create both liquidity and foreign-currency-related credit problems for the banking sector. Alternatively, and in contrast to what would appear to be a more typical emerging-markets crisis, the Baltics could be faced with a stalling of the convergence process similar to Portugal’s experience after 2000. The Portuguese economy went through a boom in the run-up to euro adoption. Sustained growth rates above the euro-area average were coupled with growing macroeconomic imbalances, pro-cyclical fiscal policy and rising inflation, which ultimately led to a loss of competitiveness. Portugal subsequently entered a lengthy period of sluggish growth to correct these imbalances with negative implications for the fiscal position.
Fitch notes several market factors that make the likelihood of a speculative attack on the Baltic currencies a weak prospect. Shallow markets imply that there are few local-currency assets to sell or liquid investments to short-sell. Most importantly, sustained fiscal discipline has delivered low levels of public debt for all three Baltic countries and the volume of local-currency debt held by foreigners is also low. Stock market capitalisation is also low, reducing the likelihood that this could be used as an avenue through which to attack the currency. Furthermore, currency board arrangements such as Lithuania’s and Estonia’s are arguably more robust than a fixed exchange rate as the domestic currency in circulation must be backed by foreign reserves. While Latvia does not have a currency board arrangement, the central bank is committed to backing the monetary base fully with foreign- exchange reserves.
Nevertheless, Fitch is not of the opinion that the Baltic currencies’ membership of ERM-II offers any protection against a speculative attack on the currency. While the European Central Bank’s reserves are much larger than those of the Baltics, it is not legally obliged to defend the central parity rates of ERM-II members but only the outer +/- 15% bands under certain conditions. Fitch does not therefore view the ECB’s reserves as a bulwark against currency volatility in the Baltic states. While an abrupt adjustment typical of an emerging- markets crisis would allow a more rapid return to equilibrium, the fact that the Baltic countries’ external imbalances are funded largely by financial flows from parent banks to their subsidiaries (in addition to not insignificant FDI flows) imply that they are less vulnerable to a loss in market confidence than countries which experience large volumes of short-term capital flows, such as Turkey. This also makes the probability of a slow adjustment process greater than that of an abrupt correction. However, a loss of confidence by residents remains a potential channel of downward pressure on exchange rates. An article in a Latvian newspaper in late February raised the possibility of a devaluation of the lats triggering a rumour of an imminent change in the value of the lats’ peg to the euro; the lats weakened to 0.7077LVL/EUR from a central parity of 0.702804LVL/EUR. While the currency did not violate the floor of its +/-1% band that would have necessitated the intervention of the central bank, the volatility caused a sharp spike in local interbank interest rates. The three-month RIGIBOR rate rose as high as 9% from 3.5%-4% at the beginning of the year. Although the rate had declined to 5.5% at the end of February, the volatility served as a reminder of growing vulnerabilities.
While all three Baltic countries are showing signs of overheating, Latvia and Lithuania respectively appear to be the most and least vulnerable to an abrupt adjustment to capital and financial flows and a slowdown in economic growth. Lithuania’s external debt levels are lower than in the other Baltics. The pace of credit growth has also been slower in Lithuania and the stock of bank credit to the private sector has been from a much lower base than in Estonia and Latvia. Furthermore, soaring property prices appear to have stabilised following the EU finance ministers’ decision on euro adoption, dampening speculation that could have been creating an asset price bubble. In short, Lithuania’s external finances compare favourably with its Baltic neighbours, implying that slippage to its euro adoption timetable is relatively less important as a rating driver. Owing to the economy’s standalone fundamentals, Fitch upgraded Lithuania’s FC IDR to A in October 2006 despite the EU Council’s rejection of Lithuania’s bid for EMU membership. Latvia was downgraded to BBB+ in August 2007. Its inflation rate is the highest in the EU and its external imbalances are the most over-stretched of the Baltic countries. Furthermore, the Latvian banking system has the lowest proportion of foreign ownership in the Baltics, indicating a relatively higher contingent liability for the sovereign.
Significant foreign ownership of banking system assets, the fact that rapid credit growth has been from a low base and the existence of an eventual ‘exit strategy’ in the form of euro adoption are crucial factors in arguing the case for the sustainability of the boom in the Baltics. Nonetheless, sustained inflationary pressures and rising external imbalances coupled with the lack of a credible and not-too-distant target for joining the euro zone increase the chance of an economic and financial correction. Fitch believes that although the lack of a monetary policy tool is an impediment to curbing runaway inflation, the Baltic governments could take steps to dampen consumer demand through fiscal policy measures. In the absence of such a resolve, external imbalances and financing risks could build and potentially lead to negative rating actions.