CAuSES Of fiNANCiAl CriSiS: ThE CASE Of lATviA

In this paper, we review how Latvia developed during the boom period and discuss the key structural features of the Latvian economy. We show that a combination of monetary and fiscal expansion contributed to a greater vulnerability to external shocks. We also show that the GDP growth was largely driven by capital deepening, while productivity gains played a significantly smaller role. As a result, one of the most important explanations for the exceptionally deep recession in Latvia should be distortions in the non-traded sectors of the economy. Finally, we give a brief analysis on policy measures that have been taken to correct the distortions and possible pros and cons of an external devaluation.

However, the key issue in this respect is whether a large CAD is constructively utilized to facilitate the catching-up process or whether it just reflects an excessive private spending.
As the growth in the Baltic States was largely driven by credit growth, tighter lending standards in the wake of the global credit crisis and falling house prices put an end to the expanding domestic demand.The Baltic economies have furthermore been hit particularly hard by lower capital inflows as capital markets froze completely, resulting in major economic contractions in the Baltic economies.Latvia was hardest hit among the Baltic states, although it was also the most imbalanced country compared to the other two Baltic states.overall, the stability of the entire region came under serious stress as well.Amongst the Baltic states, only Latvia suffered a financial strain as early as 2007.
With nearly 100 banks in the Baltics, owned by Western European counterparts, financial integration and borrowing in foreign currencies has been substantial.Financial integration brought some disadvantages by encouraging credit booms, excessive borrowing and a trend towards foreign currency borrowing (EBRD, 2009).The crisis has highlighted the need for policy measures to reduce dependency on external lending (especially in foreign exchange) and to manage the demand for credit more effectively.
Furthermore, the fact that the Baltic economies act under fixed exchange rate arrangements probably enhanced the buildup of excessive (through cheap) credit.A fixed exchange rate setup effectively disables monetary policy.Instead, fiscal policies should play the key role in short run stabilization.However, Latvian fiscal policy did not, as we will show, comply with a proper macroeconomic stabilization according to the Stability and Growth Pact (SGP).As our calculations concerning the cyclically adjusted primary budget balance suggest, fiscal policies were largely pro-cyclical during the boom years.
For a small open economy, the fixed exchange rate arrangement may be a reasonable choice.However, it requires a strong fiscal discipline.In our paper, we seek to show that a mix of poor monetary and fiscal policies contributed to the buildup of a large vulnerability towards external shocks.
Consequently, Part 2 of this article describes how Latvia developed during the boom period and discusses the key structural features of the Latvian economy.Part 3 focuses on the country's poor policy mix which significantly contributed to a greater vulnerability to external shocks.We discuss that the economic growth was mostly linked to a spurt of domestic demand, which resulted in a strong credit growth.There was a clear risk that too little of borrowed funds were diverted to productive activities.As a result, one of the most important explanations for the exceptionally deep recession in Latvia should be the distortions created in the non-traded sectors of the economy (mainly constructionrelated sectors).Part 3 analyzes also the importance of financial sector integration.It is obvious that the presence of western banks encouraged the excessive foreign currency borrowing.For the future, better ways of curbing foreign currency borrowing should be implemented to reduce and limit the excessive foreign currency borrowing.Due to the complexity of the issue, some problems touched upon by the authors are left for further discussions.In Part 4, we provide a discussion on the way out of the crisis together with the pros and cons of two policy options -internal and external devaluation.The socalled two 'Ds' (devaluation vs deflation) policy options have originated much discussion between economic analysts and politicians in the country.There are various options regarding the future of the Latvian economy, but none of them are painless.

Early warning signals -the signs of overheating
Between 2000 and 2007, the three Baltic economies recorded some of the highest growth rates in the world.The annual growth rate of the gross domestic product (GDP) was around 10% in Latvia between 2005 and 2007 and averaged almost 9% annually between 2000-2007.The decomposing growth accounting shows that capital deepening was the most important driving force behind the economic growth.Thus, the robust growth was mainly stimulated by the domestic demand, especially by private consumption and investments (see Fig. 1).The main drivers behind the fast growth of private consumption include the favourable monetary environment (negative real interest rates), strong income growth and increased employment.Indeed, the share of manufacturing in the total GDP has even exhibited negative trends.External trade contributes negatively, mostly reflecting a buoyant domestic demand and a strong import growth.A creditdriven boom in real estate prices plays an important role in fostering economic growth in all the three Baltic States.There is a clear argument that too little of the borrowed funds were directed to productive activities rather than construction activities and simple spending on goods.Thus, despite the dynamic investment developments, the productivity of investments was a clear matter of concern.A high proportion of investments was concentrated in the real estate sector and, therefore, did not contribute significantly to the overall medium-term efficiency gains.Further, as both corporate and household sector debt levels soared, this may impair the economy's ability to repay the external debt.It is notable, though, that the very

Private consumption
Net export gov.consumption gDP investments high growth can be partly explained by catching-up.However, even allowing for those effects, external imbalances and credit growth in the Baltics did reach unsustainably high levels which finally contributed to a significant economic downturn.
The Baltic economies have experienced the so-called financial deepening process when credit supply was boosted as a result of bank privatization and financial integration with the European Union.In general, we expect that the financial deepening process will play a positive role in spurring economic growth, -this is already backed by several literature reviews, Levine's (2004) being one of the most extensive in this area.However, there are many empirical surveys to show that the extraordinary pace of the financial deepening process could lead to a pressure in financial systems 1 .In 2005-06, in Latvia, lending to the corporate sector rose by 50% annually, while lending to households rose by more than 70% y/y (in the case of mortgage loans by more than 90% y/y).Thus, the pace of financial deepening in Latvia was extraordinary, even in the emerging market context (EBRD, 2009).The debt-to-equity ratio, a measure of the indebtedness in the corporate sector, rose sharply during 2005-2006 in Latvia.The Latvian household sector debt also rose sharply between 2000-2007. Between 2006and 2007, the household debt-to-income ratio jumped close to one.The high household borrowing constituted to some extent a rational response to favourable lending conditions and consumption smoothing; however, the level of indebtedness in Latvia at the end of 2007 exceeded that of many other economies with a relatively similar per capita income performance, such as Poland or Hungary.
This might reflect too high expectations about the future growth potential.The high private sector indebtedness and the current deep recession have already led to a sharp rise in the stock of overdue loans.Hence, the share of stock of overdue loans in the loan portfolio of the five biggest banks in Latvia jumped to almost 10% in September 2009.
The banking sectors of the Baltic States were dominated by an increased presence of foreign bank and their presence played a crucial stabilising role in the Baltic markets.The Latvian banking boom story was a bit different from that of the other Baltic economies for several reasons.Firstly, the Latvian banking system was less dominated by foreign banks.Locally owned banks, which made up 40% of the Latvian banking system, during the boom period borrowed heavily from abroad.Secondly, Latvian banks relied more on fragile forms of financing as almost half of total deposits in the Latvian market came through non-resident deposit banks.Hence, due to external funding constraints, the second largest bank -Parex (locally controlled) -suffered a crippling bank-run in the end of 2008 and finally was nationalized.This significantly deepened the economic crisis in Latvia in 2008-2009.
The high degree of economic and financial integration resulted in substantial economic benefits, but it also created important vulnerabilities linked to an increased exposure to foreign exchange liabilities, rising external indebtedness and bubbles in the property markets.

Overheated real estate market
In combination with excess borrowing, the real estate boom in the Baltic economies played a special role as it supported a sharp increase in domestic demand via a positive wealth effect.Better credit conditions from 2003-2004 in addition to the strong income growth, decreasing unemployment and increasing optimism, likely due to the growing expectations about the future in the wake of obtaining EU-membership in 2004, fuelled an explosive rise in property prices.The lack of housing supply at the early stage of price adjustments stimulated a surge in housing prices as well.A surge in demand, especially for new housing, was also determined by the need for better quality.Ten years ago, the majority of residential housing in the Baltics had been built in Soviet times and were characterized by bad thermal and sound insulation conditions.A number of measures can be employed to assess the valuation of the real estate market, typically including international comparisons, price-to-income ratio, the number of years needed to acquire housing, price-to-rent ratio and several other factors.one of the most popular measures used to assess housing market conditions is the price-toincome ratio.If this ratio rises above its long-term average, it could be an indication of overvaluation -and vice versa.In all Baltic countries, the price-to-income ratios 2 have risen substantially in the last few years and were way above the levels seen in some other countries.
The Latvian property market was the most overvalued in the Baltic region, and especially the Latvian capital, Riga, saw soaring prices.According to data provided by a local real estate developer, Latio, the price of a 70 sq metre apartment in Riga peaked at close to 12 times gross annual wages -significantly higher than in Estonia and Lithuania (see the figure below).In general, real estate prices in the Baltic states quickly closed the gap to those in Western Europe and were clearly too high for the current levels of incomes and rents.
The overheated housing market created strong indirectly increased tensions in the labour market as well.Workers of highly demanded professions, namely construction, real estate intermediation, some activities of manufacturing (furniture production), enjoyed very strong real wage increases, and spillovers of wage pressures were felt across the whole economy.This created a strong income effect, in fact much stronger than can be justified by productivity growth rates, which led to a further rise in domestic demand.This in turn led to a build-up of external imbalances such as a large C/A deficit.
Fiscal policies favoured speculative investment in real estate as well, due to the absence of real estate taxes and tax exemption schemes.In 2007, after a prolonged public debate, some tax measures were introduced (higher taxes were imposed for investments in the third consecutive real estate purchase for the same purpose).

2
Because of the lack of data, we use the average wage instead of disposable income per capita.

Sizeable external sector deficits
The fast growing environment created sizeable macroeconomic imbalances -first of all large C/A deficits -as was pointed out by Christensen and Rasmussen (2007).What can the macroeconomic indicators tell us about the current account deficit sustainability in the Baltic states?As import growth, due to a very high domestic demand, was considerably more buoyant than export growth in the last few years, external deficits rose markedly in the Baltic region.However, the C/A deficit was considerably higher in Latvia than in the other Baltic economies.In 2007, the CAD in Latvia reached a historically high level -almost 25% of GDP.The current account development and structural features across the Baltic states are similar and differences are mostly quantitative.As the privatization process came to an end, the importance of foreign direct investment (FDI) inflows in financing CAD has declined from more than 70% on average over the period from 2001 to 2002 to a bit more than 30% from 2003.However, this may indicate a slight decrease in the attractiveness of the region as a destination best suited for so-called "greenfield" investments.Financing CAD through loans (from local banks and other financial institutions) has picked up sharply in all three countries since 2005, enabling rising FX-reserves despite the massive CAD.However, that kind of financing resources are much more risky and volatile than FDI, and there is no doubt that the reversal of this component from H2 2007 and onwards contributed to the deepening of the current recessions.A strong import growth led to an exploding CAD, despite a relatively strong export performance implying that at least part of the external deterioration comes from the low interest rate environment and financial deepening.All three Baltic states have exhibited the alarming feature of an extremely rapidly increasing foreign debt which, moreover, was rather short-termed.Thus, the external vulnerability of the Baltic states was extremely high.
There are many sources of vulnerabilities that are common for the Baltic states during the current crisis, and a risky financing profile of external liabilities is certainly one of the most important ones.The relation of foreign debt to GDP might indicate the future problems of repayments if the figure displays high amounts.In the case of Latvia, the gross external debt surpassed GDP already in 2005.In 2008 it reached 130%, which seems to indicate a severe future burden.Turning to the assessment of the country's ability to service foreign debt in the medium to long run, it should be noted that according to the so-called Greenspan-Guidotti rule (GGR) (see olivier, Ranciere, 2007)3 , a country is considered prudent or is believed to be less vulnerable if it holds foreign reserves in the amount of its total external debt maturing within one year.The original GGR rule was to condition the amount of access to the IMF's (now defunct) Contingent Credit Line (CCL), i.e. access should be granted only if reserves plus CCL are higher than 12 months of debt being due.The Latvian short-term debt4 -to-reserves ratio exceeds the allowable level more than two times.The high ratio of external debt is a clear signal of Latvia's vulnerability resulting from an excessive short-term debt and exacerbated by an insufficient FX reserve.Thus, according to GGR, despite the falling trend, the short-term debt-to-reserves ratio exceeds the allowable level more than two times.What can the other indicators tell us about the current account deficit sustainability in the Baltic states?The current account balance can be described as the difference between national savings and national investment.
In the past few years of the boom period, investment picked up in all three countries, especially in Latvia where gross capital formation to GDP ratios rose well above 30% of GDP and exceeded historical averages by far, while the saving rates remain relatively stable.As the inter-temporal view on the CAD would suggest, a favourable interest rate environment provided a very strong stimulus to investment, while a strong domestic demand (and rising domestic prices) ensured a solid profit growth.At the same time, the impact of real interest rate dynamics on savings is not that univocal: risk-free assets were losing their attractiveness, whereas riskier alternatives provided attractive returns.Milesi-Ferretti and Razin (1996), following their theoretical and empirical studies, suggest that the current account imbalance is likely to be less sustainable if the deficit is caused by a reduction in savings rates rather than an increase in investment rates.They have pointed out that high levels of investment imply a higher future growth through the build up of large productive capacities.However, in Latvia, like in other Baltic economies, it was a clear over-investment process mostly in non-productive activities.
The Latvian economy moved into free fall very suddenly in H2 2007 as the domestic demand collapsed and the external environment deteriorated significantly.The tightening lending standards in the wake of the global credit crisis and falling house prices have put an end to expanding the domestic demand.Due to the global deleveraging process, the gap between investments and national savings narrowed significantly in 2008-2009, mostly through the scaling down of investments, but some increase in national savingsthrough the collapse of consumption -was observed as well.In fact, the C/A balance turned positive in all three Baltic states.The negative aspect of the rapid adjustment is that it reflects a very sharp contraction in domestic demand in the Baltic states, a contraction that certainly has been worsened by the fixed exchange rate setup.Hence, the negative cyclical effect has been significantly stronger in the Baltic states than in other Central East European economies where the depreciation of nominal currencies contributed to the re-balancing of C/A balances.In this context, the high level of indebtedness constrained the ability of the Baltic countries to respond to the crisis.

The role of national economic policies
The economic development in Latvia was more imbalanced than in the two neighbouring Baltic states.one of the reasons is economic policy failures.The combination of monetary and fiscal expansion contributed to an acceleration of GDP growth, which set the stage for the overheating of the Latvian economy in the years up to 2007.Turning to monetary policy, if a country's economic policy is based on a currency peg option, the burden of adjustment to shocks falls mostly on fiscal policy.However, when assessing Latvian economic policies, our studies show that policymakers failed to halt the excessive credit growth and did not effectively use fiscal policy tools.Fiscal policy in Latvia during the boom period was highly pro-cyclical.Thus, due to the poor fiscal policy management, the fiscal and monetary policy mix at that time was far from optimal.It was a favorable monetary policy condition with a negative and even declining real interest rate.The credit-fuelled domestic demand increased the inflationary pressure which lowered the real interest rate and stimulated credit growth.
Latvia's budgetary strategy during the boom period implied a significant fiscal loosening despite a very favourable macroeconomic environment.The cyclically-adjusted budget deficit rose significantly, suggesting that a tighter fiscal stance would have been more appropriate during the boom period which was characterized by strong demand pressures and large external imbalances.
In the context of international standards, the Latvian public debt was at a sustainable level before the crisis.It was well below the Maastricht criterion (60% of GDP) and IMF sustainable level for typical emerging economies -25% of GDP (IMF, 2003).However, the current credit crisis showed that even a low public debt could raise significant concern.Despite the relatively low debt-to-GDP ratio, the Latvian CDS (credit default swap) spread for five-year debt skyrocketed after the crisis.This clearly indicated a collapsed confidence due to a huge private external debt.Thus, financial markets were closed, and in the absence of any financial reserves the government was forced to ask the IMF for help.

Monetary policy
Regarding the monetary policy, the Latvian authorities started with a strong independent central bank with a flexible exchange rate arrangement.However, since February 1994 Latvia has had a de-facto peg to the IMF special drawing rights (SDR) basket of currencies, and its policies are quite similar to those of a currency board.on 1 May 2004, Latvia, together with nine other Central and Eastern European states, joined the European Union (EU).Latvia joined ERM II in May 2005 and chose to maintain the exchange rate against euro within the range ±1%.The fixed exchange rate arrangement influenced the Latvian monetary policy and was supposed to restrict its fiscal policy.The monetary policy became almost insignificant, and the fiscal policy had all economy-regulating levers.
Empirical evidence shows that a fixed exchange rate policy is associated with a lower inflation (see Ghosh et al., 2000).Indeed, currency boards in Lithuania and Estonia, until mid-2007, were remarkably successful at maintaining price stability, thereby contributing to an outstanding economic performance in the last decade.However, curbing inflation was less successful in Latvia.On the other hand, it is clear that if a fixed exchange rate regime is not supported by sound fiscal policies, imbalances can build, possibly leading to overvalued real effective exchange rates, which in turn lower the growth potential.Thus, due to the increasing inflation rate and rising labour cost, the Latvian real exchange rate (REER) appreciated, especially when based on the unit labour costs (ULC).This indicates that export competitiveness was deteriorating sharply.
Regarding the exchange rate development in the long run, it is expected that all three countries will finally adopt the euro.But there is still a long way to go, and we think that fixed exchange rate regimes could be a risky policy option in the preparation for euro adoption.In the face of nominal convergence and positive capital inflows, as was seen between 2002-2007, the fixed exchange rate prevents nominal appreciation and thus creates excess liquidity, which lead to bubbles in asset markets.Therefore, price and current account deficit (CAD) pressures emerged.
Although the currency pegs have encouraged price stability and restored the stability of financial markets over the past periods, the extent of the recession has called attention to the drawbacks of the pegged currencies.First of all, the reliance on fiscal adjustments was the only policy tool available to restore competitiveness and growth.

The role of ULC development
There are many empirical studies that have examined the role of unit labour costs (ULC) in shaping inflation and competiveness dynamics.The evolution of ULC is a key determinant of inflation and competitiveness.The high ULC growth can be explained by 1) intensive wage growth, 2) low productivity growth, or by the combination of both factors.Thus, despite a strong productivity growth, a significant increase in wages can result in a high ULC growth -and it did in Latvia as we will see below.It can be noted that in Latvia, as well as in other Baltic economies, labour cost growth started to accelerate soon after the EU accession.
In the case of the Baltics, the elevated growth in ULC was driven by a strong domestic demand in non-traded sectors, especially in the construction and real estate sectors.Further, a strong labour force emigration, both internally and externally, added to tightness in labour markets, which in turn pushed up wage growth excessively.This undermined the external competitiveness in the Baltic states, especially in Latvia where the share of high-tech and medium-high-tech products of total exports is relatively low.
When discussing the issue of competitiveness, one might address the Balassa-Samuelson (BS) effect.If productivity growth in the traded goods sector is higher than in the non-traded goods sector, this leads to higher wages in the traded goods sector, but also in the non-traded goods sector, even if not justified by productivity gains in that sector.This effect assumes full employment and perfect labour mobility between sectors.one standard definition of the real exchange rate is the ratio of prices between non-traded and traded goods.Hence, a natural extension of the above argument is that real appreciation can follow in cases where the home country experiences higher productivity growth in the traded sector compared with the non-traded sectors.
Some studies address the case of the adverse BS effect, which can occur when for example an increase in demand in the non-traded sector pushes up non-traded wages which, through labour mobility, force the traded sectors to increase productivity and wages.This effect was discovered by Grafe and Wyplosz (1997) for transitional econo-    mies.For many transitional economies, the non-traded sector (mainly services) is in general underdeveloped, providing a large productivity potential compared to the traded goods sectors.Hence a relative price increase of non-traded to traded goods is caused by initial "distortions" in relative prices (housing, health care, education) rather than by a faster growth of productivity in manufacturing than in services.
In order to support these arguments, we have calculated labour productivity and real wage growth rates in the traded and non-traded sectors.In the traded sector, we included manufacturing, agriculture, mining, transportation and communications and hotels, while the non-traded sector includes energy, construction, wholesale and retail trade, real estate and business services, education, health and personal services.overall, we see a limited empirical evidence of the Balassa-Samuelson effect at play in Latvia during [2005][2006][2007][2008].First of all, when we take the ratio of traded-to-non-traded productivity growth, we detect almost no change in relative productivity since 2005 as both sectors experienced a comparable productivity increase during this period.However, we see some mechanisms underlying the adverse Balassa-Samuelson effect to have some impact, as it looks like a real wage increase in non-traded sectors in 2005-2006 spilled into manufacturing and raised wages there.However, the wage increase exceeded the productivity growth in both sectors in general, and this put upward pressure on ULC and inflation, ultimately hurting the competition.

Fiscal policy stance
Over the period of the economic boom, the fiscal policy tended to be pro-cyclical in all Baltic states and Latvia was certainly no exception.Hence, fiscal policies enhanced the boom rather than acting in a counter-cyclical manner.
However, the deficit of the consolidated budget was compliant with the criteria of the Maastricht Treaty as it was below 3% of GDP and the gross debt was below 60% of GDP.However, the actual deficit does not show whether the fiscal policy matches the economic conditions.Normal assessments regarding the soundness and sustainability of public finances are usually based on cyclically adjusted budget balances.These focus purely on the development of the structural budget balance which largely represents the result of discretionary fiscal policy.The cyclically adjusted budget balance represents the budget balance on a normal growth path.Adjusting budget balances for cyclical effects makes it possible to analyze the development of structural budget balances and to assess the effects of discretionary fiscal policy.In addition to the nominal Maastricht Treaty criterion, the SGP introduced a commitment to reach a medium-term budgetary position that would allow dealing with "normal" cyclical fluctuations while keeping the budget deficit below 3% of GDP over a business cycle.This measure is used as an assessment of whether or not the government finances in the EU countries comply with the stipulations of the Stability and Growth Pact.According to this pact, the medium-term fiscal policy strategy must be targeted to a balance, and the anti-cyclic fiscal policy costs should not exceed 3% of GDP.The 2005 reform of the SGP includes a revision of the criteria for the determination of mediumterm budgetary objectives.They are allowed to be differentiated across countries reflecting differences in debt levels, sustainability and potential growth.
The deficit of a cyclically adjusted budget does not reveal the impact of the fiscal policy on the economy.In this case, one should assess automatic budget stabilizers mitigating the influence of economic fluctuations on state finances.Following the expert evaluation of the results (elasticity indices) and other factors (revenue and expenditure structures), it may be stated that the automatic stabilizer does not work for spending, and the one for revenue should be quite considerable taking into account the rather high dependency of indirect taxes on private consumption.This allows for the assumption that during a recession a lower deficit may be expected, if it is clear that no targeted policy to increase spending is going to be undertaken.However, the reality was quite the opposite due to the significant effect of the discretionary fiscal policy.The cyclical influence on the fiscal position during the boom period in the Baltics was positive.However, cyclically adjusted or not, the structural balance deteriorated during the boom time.In the context of strong demand pressures and external imbalances, contractive fiscal policy would be very desirable in Latvia.
Calculations5 show that Latvia's nominal budget almost balanced during the boom period time.This was primarily due to the positive impact of the business cycle; fluctuations of macroeconomic variables do play a significant role.The dynamics of a cyclically adjusted primary balance is very close to that of a cyclically adjusted budget balance, as changes in interest rate payments relative to GDP were relatively moderate.It is apparent that the cyclically adjusted primary fiscal balance has deteriorated substantially in the last five years (see Figs. 19,20).The largest deterioration was observed in 2006-2007.It should be mentioned that the budgetary deterioration in 2006 mostly appeared due to parliamentary elections.The main fiscal measures for 2006 include an increase in personal income tax-exempt income and tax relief, as well as extra increases in public expenditure.
Thus, the risk to the government deficit might exceed the 3% of GDP reference value in the event of falling economic growth rates raised already in 2006.A signifcant deterioration of the cyclically adjusted balance has occurred in accelerated GDP growth conditions, which indicates that during the boom time the fiscal policy in Latvia shows strong pro-cyclical tendencies.

The ways out of the crisis -internal devaluation option
A response to a crisis aimed at countering the drop in domestic demand depends upon a country's ability to implement counter-cyclical fiscal and/or monetary policies.The Baltic states, however, do not have any leeway for expansionary policies.First of all, the countries have been running budget deficits (except Estonia) in the past ten years.Thus, in this case, any attempt to raise domestic demand by running fiscal deficits will only widen trade deficits and current account imbalances and further stimulate imbalances in the economy.The current crisis has even elevated the need for painful adjustments in this region, with relatively adverse economic impacts.Thus, a key policy choice for Latvia in this respect was the so-called internal devaluation when domestic wages and prices are adjusted down to restore competitiveness.The Latvian labour market is described as being very flexible, especially in terms of downward wage flexibility.Besides, unemployment changes are relatively well correlated with employment changes.The relatively high labour market flexibility has allowed it to adjust more readily to external shocks.
It looks like the wage/productivity gap is finally beginning to close as of H2 09, and the real effective exchange rate (REER) is thus correcting.However, a stronger adjustment is observed in the REER deflated by CPI than ULC, which might suggest a need for a continuation of wage disinflation for some time to come.It should be mentioned that internal devaluation particularly affected the distribution of social transfers -and in some cases with significant consequences.Most of the population is hit by these measures, but especially those with the lowest income levels.Latvia has already experienced a jump in unemployment, and the ongoing internal deflation is likely to force unemployment up even higher in the short run.
In total, in the long-run, the internal devaluation is supposed to result in lower prices (costs) in the economy and will finally make local products more attractive to export markets as well as more affordable for local consumers.An increase in exports should revive other sectors of business and lead to increased consumption and domestic demand.
However, this policy option would lead to deteriorating labour market indicators.

External devaluation, its pros and cons
"When prices and wages adjust slowly, the immediate effect of devaluation is to improve international competitiveness" (Begg et al., 2001).The simplest arguments in favour of external devaluation: Probably the quickest way of improving competitiveness.1.
Should lead to lower real interest rates, if devaluation is carried out in a credible 2.
manner.A decline in real interest rates along with improved trade prospects, in turn, would stimulate economic activity and reduce unemployment.
Higher inflation would help boost government tax revenues and improve nominal 3.
fiscal position.
Lower relative prices would promote tourism.4. of course, there are arguments against these policy options: Higher inflation would hurt the purchasing power of the households.

1.
Negative balance sheet effect (as most of Latvian liabilities are denominated in 2.
euro, devaluation of the national currency would lead to a redistribution of wealth to creditors from debtors.However, the inflation that would follow devaluation would somewhat mitigate this wealth redistribution, although it would bring its own set of problems.Devaluation would put a substantial pressure on the Latvian financial system.

3.
The higher costs of servicing the Latvian government's foreign debts following 4.
devaluation would increase the public budget deficit as well.
It should be noted that imports have a large impact on the Latvian economy; especially the importance of energy resources in imports is high as these goods are necessary both for domestic consumption and for manufacturing exportable products (intermediate and capital goods account for 60% of Latvian imports).In the event of devaluation, the price of imports will rise significantly over the medium term, and this could cancel out the positive effect of devaluation.
From the theoretical point of view, depreciation of the exchange rate is thought to promote exports and decrease imports, leading to an improving trade balance; this holds true only if the sum of the price elasticity of export and import demand is higher than one -the Marshall-Lerner condition (Krugman, obstfeld, 2006).In the simplest version of the Marshall-Lerner condition (MLC), a real devaluation (or real appreciation) of the currency will improve the trade balance if the sum of elasticities (in absolute value) of the demand for import and exports with respect to the real exchange rate is greater than one.So, what the MLC states is that, in the event of real devaluation, if each elasticity is less than one but the sum is greater than one, the increase in imports will be more than offset by the increase in export.
Thus, in general, the functioning of the MLC is based on a full exchange rate passthrough to import prices and zero pass-through to export prices (Coricelli et al., 2006).only in this case we might expect that the valuation effect in the case of devaluation would overwhelm the quantity effect.Trade sector structure in terms of the response of import/export to changes in relative prices is an important issue.The Latvian import structure indicated a relatively high share of imported inputs.Taking into account that the domestic substitute of those goods (for example, energy products) is narrow, we might see that the country's industries would be exposed to exchange rate movements through the cost of imported inputs into production rather than through their export markets.Latvia has a relatively high import share of material-intensive goods, and we might expect to see only a "partial" aggregate exchange rate pass-through.In summary, the Latvian external trade structure provides some pessimism about foreign trade elasticity given such import/export structure.
What about the impact of devaluation on domestic cost and as a consequence on exports?Domestic costs are affected by devaluation through two channels.First of all, it causes the rise of the cost of imported intermediate inputs and thus an increase in domestic cost.The significance of this effect will depend on the share of imported inputs in the unitary cost of the production and the elasticity of substitution between domestic and imported goods.Second, devaluation would reduce the real prices of domestic inputlabour -in terms of importable goods.Thus, devaluation would improve the competiveness more in the sectors with a high labour share in unitary costs compared with those with high imported intermediate inputs.
A relatively high concentration on low-and medium-tech products in total Latvian exports is associated with a relatively high labour intensity.However, a deeper analysis of the Latvian manufacturing sector is needed.2006) draws a connection here between the exchange rate channel and the credit channel.The effect of the credit channel on the economic activity of firms constrained by external finance is exacerbated if these firms are in low mark-up sectors.If changes in the exchange rate significantly impact the firms' balance sheets, they will influence these firms' capacity to borrow externally (broad lending channel).A collapse in investment due to an exchange rate-induced fall in firms' net wealth may outweigh competitiveness gains, provided the MLC is verified.Consequently, individuals and corporations would prefer to hold their liquidity in foreign currency -keeping it safe at home or in a foreign bank.This was clearly demonstrated by concerns rising within the community about the stability of the Latvian lats in H1 09 and affecting the behaviour of the population.The overall bank deposits in lats decreased by almost a tenth.However, as these concerns abated, this trend reversed.
The MLC is verified in the transition countries of Central and Eastern Europe (Coricelli et al., 2006), and we have no strong evidence that the MLC is not met in Latvia.There are many complications to these conditions; according to empirical evidence, the exchange rate pass through is typically not equal to 1, and many other things might occur at the same time, such as exchange rate devaluation, and finally outweigh the positive/negative effects.

Conclusions
Relieving credit constraints and cheap money lead to an excessive bank lending and property bubbles in Latvia.Besides, such credit-stimulated domestic demand creates inflationary pressures which, in turn, further support credit growth at lower real interest rates.All those factors increase concerns about the Baltics overheating well before the global credit crunch (Christensen, Rasmussen, 2007).
The presence of sizeable macroeconomic imbalances -including strong tensions in the labour market, rises in unit labour costs, large current account deficits, rising inflationary pressure, excessive borrowing and potential overvaluation -highlight the need for adjustment.
The biggest challenge was that a combination of monetary and fiscal expansion contributed to an acceleration of credit and GDP growth, which set the stage for the overheating of the Latvian economy in the years up to 2007. on the one side, monetary authorities cannot effectively control the supply (or the price) of credit due to fixed exchange rate arrangements.This poses a challenge as to the first-best and second-best policies geared to contain the credit growth.The pro-cyclical fiscal policy fuelled the demand side as well.
Tightening lending standards in the wake of the global credit crisis and falling house prices have put an end to the expanding domestic demand.However, the current crisis has elevated the need for painful adjustments.In the absence of exchange rate flexibility, the response to shocks will have to be mitigated only by a tight fiscal policy as well as by policies enhancing structural flexibility.The real exchange rate appreciation, credit restraint and fragile recovery in external demand are making those adjustments even more challenging.
The other policy options that would imply an immediate positive effect to exports caused much discussion among experts, both domestic and foreign.So far, according to our analysts, both deflation and devaluation would stimulate the export of the Latvian economy.However, with no strong empirical evidence confirming the Marshall-Lerner condition, which is crucial for the expansionary effect of devaluation, we cannot be sure whether the negative effects on import prices do not outweigh the positive effects on exports.This issue should be examined further.
Both deflation and devaluation would further improve the external account.On the other hand, the foreign account is not just an external competitiveness issue, especially when there is a huge external debt denominated in foreign currencies.Even if devaluation improves the trade account, the external debt will be revalued.Thus, the net effect on the foreign account will be uncertain.
A downside to devaluation policy option is the risk of a high inflation or even hyperinflation.However, due to a large negative output gap, we might expect a relatively low risk of this effect in Latvia.
Finally, there are greatly large negative balance sheet effects.on the other hand, banks have already set aside reserves for loan loss provisions.Besides, foreign banks in Latvia have announced that the size of losses is not affected by devaluation.only the timing of losses is affected.
It is also a fact that the Latvian currency is still acutely overvalued, and this reduces capital inflows to the country as well.Devaluation would probably lead to a rise in capital inflows.This was the case in Argentina after the default and devaluation in January 2002.
Finally, again we admit that there is no certainty that, under the given structure of the Latvian economy, devaluation would work as an output expansionary policy option.However, it should be pointed out that despite the expected positive outcome in the long run, in a short run the internal devaluation will be a socially unfriendly course of policy.Thus, it is still an open question as to whether the right policy has been chosen.Deflation and the deteriorating labour market have caused a long-term recession of the Latvian economy, which would be difficult to recover from, even if the global environment improved.There are various options regarding the future of the Latvian economy, but none of them are simple and painless to society.What is clear is that in order to restore a balanced economic development and competitiveness, more complex policy options need to be implemented.

Fig. 8 .
Fig. 8. ratio of apartment prices to gross annual wages Sources: Statistics Lithuania, Estonia, Latio real estate agency data.

Fig. 11 .
Fig. 11.latvia: some indicators of sustainability of C/A Sources: Latvian central bank, authors' calculations.