SEVERITY OF THE CRISIS AND ITS TRANSMISSION CHANNELS Cesar Calderon and Tatiana Didier* 56752 December 20, 2009 Abstract The current global crisis, although initially circumscribed to the US housing market, spread rapidly across markets and borders. It has affected almost all countries through different reinforcing channels: the contraction in international trade, capital flows, remittances, and international commodity prices. The main goal of this note is to empirically analyze the mechanisms through which the financial crisis of 2007-2009 propagated throughout the world by characterizing the main factors behind the fall in GDP growth rates. Our findings indicate that a greater decline in the growth rate was registered in countries with higher de facto trade openness, less resilient domestic financial markets, and, to a lesser extent, improved macroeconomic frameworks. To complement this evidence, we construct an aggregate index of the severity of the crisis that captures the real and financial consequences in each country of this unprecedented global financial shock. This index provides evidence that LAC countries were indeed hit by the crisis but not as severely as industrial economies and Eastern European countries. 1. Introduction The current crisis affecting the world economy is of historical dimensions and is re-shaping the international economic and financial landscape. It started in the U.S. housing market and it was initially considered a relatively bounded problem. However, and after Lehman Brothers' collapse in September 2008, the turmoil spread rapidly across markets and borders. Losses at the epicenter of the storm, advanced economies, were massive: failures of financial institutions, considerable deleveraging, and a staggering collapse in asset values (some US$ 18 trillion in G-7 stock market capitalization has vanished relative to the admittedly overvalued pre-crisis peaks). At the periphery, the effects include but are not limited to the sharply reduced availability of international market finance (debt and equity), the deterioration in the terms of trade for net commodity exporters, the decline in remittances, and the pronounced contraction of external demand for emerging economies' goods and services. In an effort to restore confidence and stimulate aggregate demand, governments throughout the world have resorted to large-scale financial rescue and fiscal stimulus packages. Fortunately, a full meltdown of financial intermediation was averted and confidence began to stabilize, with an inflexion point materializing around March-April 2009. A broad menu of bold actions by central banks, led by the U.S. Federal Reserve Bank, played a key role in averting the catastrophic scenario. * For very helpful comments, we would like to thank Augusto de la Torre. We are grateful to Paula Pedro and Virginia Poggio, who provided excellent research assistance. The views expressed here are entirely those of the authors. They do not necessarily represent the views of the World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Despite the intensity of the shock and its unforeseen synchronized propagation, the transmission of the US crisis was heterogeneous across countries around the globe. In some countries, like Malaysia and Colombia, stock markets tumbled initially but started to recover just a couple of months later. Other countries, such as Chile and Hong Kong, saw steady declines in stock markets until March 2009, when US markets reached their bottom. The fall in economic activity was not completely symmetric across countries either. For instance, GDP contractions in 2009 are estimated for Japan at 5.7 percent and for Western Europe at 3.9 percent. Among emerging regions, Eastern Europe was hit the hardest, with 2009 growth expected at -5.6 percent (Figure 1). Real GDP among the East Asian tigers (EAP7) is envisioned to contract by 0.7 percent in 2009, whereas Asia (excluding China and India) may expand 1.64 percent. In other words, some countries displayed a higher degree of stability than others, which raises questions about the mechanisms behind the transmission of the crisis. This note aims to empirically investigate the mechanisms through which the financial crisis of 2007-2009 propagated throughout the world by analyzing the factors behind the fall in GDP growth rates. 2. Growth Collapses To better gauge the observed differences across economies and ascertain the factors behind them, it is worth focusing on the changes in growth rates, rather than on the growth rates themselves. This is because the distinction between trend and cycle in economic activity is crucial. By examining deviations from trend growth, it is possible to identify the drivers of the current downturn while controlling for initial differences in potential GDP. Unfortunately, the techniques used to calculate potential GDP are usually unreliable towards the end of the sample period due to an overweighting of the endpoint sample observations (Mise et al. 2005). Hence, in this note, we take an admittedly blunt shortcut by using GDP growth in 2007 as a proxy for trend growth before the crisis. Next we calculate the difference between real GDP growth rates between 2009 and 2007 and use the resulting "growth collapse" as a measure of deviation from trend. Thus, our measure of growth collapse indicates not only the actual deceleration in growth between 2007 and 2009 but also aims at capturing the difference between peak and trough during the current downturn. The average growth collapse in LAC between 2007 and 2009 was significant (approximately 6.8 percentage points). However, LAC does relatively well when comparing 2007-2009 growth collapses across other regions (Figure 2). The collapse in Eastern Europe and Central Asia was around 13 percentage points, and that for East Asian tigers reached more than 7 percentage points. Rich countries also experienced a collapse in growth (around 7 percentage points) that was higher than that of LAC. Moreover, LAC's collapse is comparable to that of East Asian tigers and industrial economies. However, smaller collapses were observed in Sub-Saharan Africa (3.5 p.p.), Middle East and North Africa (3.3 p.p.), and in South Asia (2.2 p.p.). LAC also contracted more than China, which is expected to decelerate from a real GDP growth rate of 13 percent in 2007 to 8.3 percent in 2009. In order to understand these differences in growth collapses across countries, we evaluate the factors behind the fall in GDP growth rates. Our sample of growth collapse episodes includes 140 countries, of which 22 are industrial economies and 19 are Latin American countries. The analysis considers pre-crisis (2000-2007) indicators as regressors. We run simple cross-sectional regressions with the change in GDP growth rates as our dependent variable. As independent variables, we consider more structural factors such as openness, trade and capital flow composition as well as some vulnerability indicators such as the resiliency of the banking system. 2 Take openness first. The evidence in Table 1 clearly suggests that emerging countries with greater de facto trade openness experienced a sharper drop in growth rates. Similarly, growth tended to decelerate less in countries with larger domestic demand, as measured by the share of private consumption in GDP. Simply put, and not surprisingly, as trade collapsed, countries that relied more heavily on it were hit harder. On the other hand, de facto financial openness (measured as country's total assets and liabilities as a percentage of GDP) did not seem to have mattered for the collapse in growth rates, despite the financial market origins of this crisis. It should be noted however that these results do not imply that protectionism measures to close the economy are a good idea. Countries with a greater degree of openness would show a higher degree of synchronization with the world business cycle.1 Therefore, they are more exposed to external shocks.2 However, the empirical evidence also suggests that higher trade volumes lead to a positive effect on a country's income.3 Consider next in Table 2 the composition of a country's trade flows (by trading partners and by product) and its structure of external assets and liabilities, respectively. Countries with a higher share of manufactures in total exports experienced a greater deceleration in growth than commodity exporters. However, the diversity in trading partners was not statistically significant to explain the variations in growth collapses across countries. Finally, in terms of the composition of capital inflows, given that debt flows to developing countries fell sharply while FDI remained relatively stable, countries with a greater share of equity (FDI and portfolio) liabilities in total external liabilities (which in LAC include Brazil, Chile, and Peru as opposed to Argentina, Uruguay, and Venezuela which rely more on debt flows) registered smaller declines in GDP growth rates. Similarly, countries with a greater share of debt flows experienced a larger growth collapse. Consider next the resiliency of financial systems in Table 3. We measure it mainly through the inverse of the loan-to-deposit ratio--that is, banking system resiliency rises to the extent that credit intermediation relies more on the mobilization of domestic financial savings and less on the mobilization of foreign funds. We also analyzed excess credit growth, which is measured by the difference in the real growth rates of credit in 2006-2007 vis-a-vis that of 2002-2005. For robustness purposes, we looked at average real credit growth between 2001 and 2007 as well. Typically, higher growth in domestic credit is associated with a lowering of lending standards which have lead to higher non-performing loans during past crises. Hence, this measure is also suggestive of more fragile financial systems. The regression results suggest that, after controlling for trade openness, emerging countries with less resilient financial systems faced significantly larger declines in their growth rates. The most notable case is that of several Eastern European countries. Fortunately, LAC entered the crisis with relatively safe and sound financial 1 Several empirical studies show that countries with deeper trade linkages would exhibit higher business cycle synchronization (Frankel and Rose, 1998). Also, the impact of trade on cycle correlation is higher among North- South countries than among South-South countries (Calderon et al. 2007). Furthermore, Di Giovanni and Levchenko (2010) show that bilateral international trade increases co-movement significantly more in cross-border industry pairs that use each other as intermediate inputs. 2 Di Giovanni and Levchenko (2009) find that countries that are more outward oriented tend to be more volatile. Using industry-level data on manufacturing production and trade, the authors find that trade may lead to higher vulnerability thanks to the interaction of the following channels: (a) trade is accompanied by increased specialization, and (b) sectors more open to international trade are more volatile. 3 See for example Frankel and Romer (1999). Recent evidence shows that: (a) trade has a significant effect on income and explains almost one-fifth of cross-country differences in income per capita over the long run (Feyrer, 2009), and (b) annual growth rate in countries that liberalized their trade regimes were 1.5 percentage points higher than before liberalization, with the investment being the most predominant channel of transmission (Wacziarg and Welch, 2008). 3 systems, owing to significant improvements in macroeconomic and financial policies and institutions. Moreover, reinforcing this argument, it should be noted that the level of credit in an economy is not significant in explaining differences in the decline of GDP growth rates in this crisis. Improved macroeconomic policy frameworks were also associated to some extent with smaller growth collapses (Table 4). For instance, countries that have pursued pro-cyclical fiscal policies before the crisis have experienced a greater fall in their real growth rates, other things equal. This can help us understand differences between Argentina, which had highly pro-cyclical policies, and Brazil and Paraguay. Still on the fiscal stance, countries with growing budget deficits between 2003 and 2007, i.e., countries in which government expenditures have been growing faster than government revenues, have also experienced greater growth collapses once we control for income level and trade openness. On the monetary front, countries with lower inflation rates have been associated with a smaller fall in their growth rates. However, exchange rate flexibility has not been associated with smaller growth collapses. Nevertheless, better monetary frameworks in LAC countries might have helped avert systemic damage while affording room for countercyclical responses in the monetary area. In effect, and in sharp contrast with its past, LAC did not experience domestic crises this time around. 3. The Severity of the Crisis: An aggregate index Section 2 characterized the different channels of transmission of the current global financial crisis by regressing the collapse in GDP growth rates on several structural features of the economy: trade and financial openness, composition of trade flows, structure of foreign liabilities, the resilience of the domestic financial sector, and the stance of macroeconomic policies. A more comprehensive measure of the severity of the current financial crisis goes beyond the collapse in GDP growth rates. It should also capture different developments on the financial side and on the real side of economic activity: massive stock sell-offs, surges in sovereign spreads, weakening of domestic currencies, and downturn in real economic activity (as approximated by collapse in GDP, but also by manufacturing production and foreign trade). Following Rose and Spiegel (2009a and 2009b), we use the conventional factor analysis in summarizing the information of six (6) financial and real indicators of the consequences of the crisis by estimating a single common factor. We consider three indicators that capture the real aspects of the crisis: (a) Growth in real GDP in 2009.Q1 (% year-on-year), (b) variation in industrial production index in 2009.Q1 (% year-on-year), and (c) growth in exports in 2009.Q1 (% year-on-year). On the other hand, the following financial indicators were also considered for our measure of the severity of the crisis: (a) variation in the real effective exchange rate in March 2009 (% year-on-year), (b) rate of change in the aggregate stock price index in March 2009 (% year-on-year), and (c) change in the country credit rating from Institutional Investor (IICCR, variation in March 2009 vis-ą-vis March 2008). Our measures of real and financial consequences of the crisis are not particularly closely associated (Table 5). The correlation among crisis variables is never greater than 0.5 (in fact, the largest degree of co-movement is between growth in GDP and IPI, which equals 0.44). The aggregate indicator of the severity of the crisis is then computed by extracting a common factor of the six indicators mentioned above for a sample of 65 countries.4 Table 6 reports the Top 20 countries 4 A principal component analysis involves a mathematical procedure that transforms a number of possibly correlated variables into a smaller number of uncorrelated variables called principal components. The first 4 hardest hit by the crisis. Twelve (12) out of the top 20 countries in our index of the severity of the crisis come from Eastern Europe, the region with the largest fiscal and external vulnerabilities prior to the crisis. According to our index, Ukraine is the countries that was hit the hardest by the crisis --as signaled by a decline of 14% YoY in GDP, a sharp drop of 30% of both manufacturing production and exports, stock prices plummeting and a downgrade in the institutional investor country rating (of more than 13 points). Latvia and Lithuania have also been adversely affected by the crisis: they experience a GDP contraction of more than 15%YoY (along with a decline of more than 30%YoY in exports), and a sharp drop in stock prices (of more than 60%YoY). Among industrial economies, Iceland, Ireland, Finland, Luxembourg and Japan have been hit the hardest. We should note that while Iceland and Ireland have been more affected by developments in financial sector (financial crises and housing market turbulence), the other countries have been affected by a sharp decline in real economic activity --i.e. exports and manufacturing production plummeted. Finally, we should add that two Latin American countries (Argentina and Mexico) are at the bottom of the top 20 countries most severely hit by the crisis. According to the latest Consensus Forecasts, Mexican GDP is expected to decline by 7% in 2009 while the contraction is Argentina will be of 2.7%. Figure 3 depicts the scatter plot of the real consequences vis-ą-vis the financial consequences of the crisis for our cross-section of 65 countries. The 45 degree line in the Figure signals whether the real and financial consequences of the crisis have been symmetric. The figure shows that, for instance, although the crisis has manifested in the deterioration of real and financial conditions in Iceland, Ukraine, Kazakhstan, Bulgaria, and Romania, the impact has been more adverse on the financial side. The real effects of the crisis have been more deleterious in Latvia, Lithuania, Estonia, Russia, Hungary and Serbia. Note that while in the former group, stock market prices and sovereign spreads plummeted more than in other countries; in the latter group, the contraction in GDP was significant. Note that the countries in the LAC region, with the exception of Argentina and Mexico, have not been as badly affected by the crisis. Finally, Figure 4 plots our index of economic consequences of the crisis and the level of income per capita. We clearly observe a negative association between both indicators. This implies that richer countries have been hit the hardest during the current global crisis. However, we should point out that several middle income countries with fiscal and external imbalances at the beginning of the crisis --say, ECA countries-- have also been severely affected. 4. Conclusion The global financial crisis has had deleterious effects on the financial and real conditions of economies around the globe. What started as a somewhat bounded turmoil in the U.S. housing market turned into a severe financial crisis at the heart of industrial economies and spread quickly to emerging market economies. This crisis left almost no country unharmed. Initially, stock market prices plummeted, sovereign and corporate spreads surged and exchange rates depreciated. The global credit crunch severely affected international payments and trade credit, thus leading to a collapse in exports and manufacturing production across the world. The global crisis affected well-behaved countries as well as countries that had macroeconomic imbalances in the run-up to the crisis. However, the impact has been heterogeneous across countries principal component, which is used here as a proxy for the severity of the crisis, accounts for as much of the variability in the data as possible. 5 both in terms of the severity of the crisis and on the channels of transmission through which the global shock affected these economies. This note attempted to characterize the structural factors that were more associated to the collapse in growth for developed and emerging market economies. Our findings indicate that a greater decline in the growth rate was registered in countries with: (a) higher de facto trade openness (and especially those with a larger share of manufacturing exports), (b) a structure of external liabilities biased towards higher debt rather than equity (FDI and portfolio equity holdings), (c) less resilient domestic financial markets, and, to a lesser extent, (d) improved macroeconomic frameworks. Finally, the collapse in the growth rate of economic activity is not the only manifestation of the current global crisis. We constructed an aggregate index of the severity of the crisis by summarizing information on the real and financial conditions of the economy using factoral analysis. This index extracted information on the consequences of the crisis on the real side of the economy (GDP growth, manufacturing production, and exports) and on the financial side (stock markets, real exchange rate, and country ratings). The evidence presented here showed that LAC countries were indeed hit by the crisis but not as severely as industrial economies and Eastern European countries. References Calderon, C., A. Chong, and E. Stein, 2007. "Trade Intensity and Business Cycle Synchronization: Are Developing Countries Any Different?" Journal of International Economics 71(1), 2-21. Cerra, V., and S. C. Saxena, 2008. "Growth Dynamics: The Myth of Economic Recovery." American Economic Review 98(1), 439-457. Di Giovanni, J., and A. Levchenko, 2009. "Trade Openness and Volatility." The Review of Economics and Statistics 91(3), 558-585. Di Giovanni, J., and A. Levchenko, 2009. "Putting the Parts Together: Trade, Vertical Linkages, and Business Cycle Comovement." American Economic Journal: Macroeconomics, forthcoming. Feyrer, J., 2009. "Trade and Income: Exploiting Time Series in Geography." NBER Working Papers 14910, April. Frankel, J.A., and A.K. Rose, 1998. "The Endogeneity of the Optimum Currency Area Criteria." The Economic Journal 108(449), 1009-25. Frankel, J.A., and D. Romer, 1999. "Does Trade Cause Growth?" American Economic Review 89(3), 379- 399. Mise, E., T.-H. Kimb, and P. Newbold, 2005. "On sub-optimality of the Hodrick­Prescott filter at time series endpoints." Journal of Macroeconomics 27(1), 53-67. Rose, A. K., and M. M. Spiegel, 2009a. "Cross-country causes and consequences of the 2008 crisis: Early Warning." NBER Working Paper Series 15357, September. Rose, A. K., and M. M. Spiegel, 2009b. "Cross-country causes and consequences of the 2008 crisis: International linkages and American exposure." NBER Working Paper Series 15358, September. Wacziarg, R., and K.H. Welch, 2008. "Trade Liberalization and Growth: New Evidence." The World Bank Economic Review 22(1), 187-231 6 Figure 1. Growth Outlook Across Regions This figure presents the annual growth rate of the real GDP in 2008 across regions and the GDP growth forecasts for 2009 and 2010. The data comes from Consensus Forecast (Oct. 2009) Recent Growth and Forecasts for 2009-10 Annual Growth Rate 7% 5% 3% 1% -1% -3% -5% 2008 2009 2010 -7% Japan Eastern EURO US Latin World Asia Pacific Europe America Figure 2. Growth Collapses This figure presents the growth collapse in 2009 (in percentage points). The collapse was defined as the difference between the GDP growth in 2007 and expected GDP growth for 2009. The source was the IMF's World Economic Outlook for the 2007 real GDP growth, and the 2009 GDP growth forecasts were taken from IMF's Regional Economic Outlook and Consensus Forecasts (as of Oct. 2009). Growth Collapses Across the World GDP Growth collapse, in p.p. (diff. between growth in 2007 and 2009) 0 -2 -4 -6 -8 -10 -12 -14 ECA EA IND LAC China SSA MENA SA Tigers Figure 3. Severity of the crisis: Financial and real aspects This figure decomposes the aggregate index of the severity of crisis into its real and financial components. As described further, the real index is composed by the variations in the GDP, industrial production and exports. The financial sub-index captures the variations in the REER, stock prices and the country credit rating. Our sample is composed by 65 countries. Sources of data: IMF's IFS, Institutional Investor and Bloomberg. Real and Financial Consequences of the Crisis 2 1 TUN COL CHL BRA MAR Index of Financial Turbulence ISR JOR MON MYS CHEPER CHN TWNDEU TUR ZAF PHL IDN SWE SVN CAN HKG FRA OMN IND MEXGBRDNK NZL NOR POL AUS 0 JPN FIN SGP CZE PRT ESP NLD THA USA KOR EGY ITA LUX BEL RUS SER AUT VNM HRV GRC HUN LVA ROM LTU EST IRL ARG BGR -1 KAZ PAK UKR -2 -3 ICE -4 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 Index of Real Turbulence 8 Figure 4 ­ Index of Economic Turbulence and Level of Income Per Capita This figure plots the scatter of the aggregate index of the severity of the crisis vis-ą-vis the level of GDP per capita. Sources of data: IMF's IFS, Bloomberg and Haver analytics GDP Per Capita and Economic Turbulence 2 CHN 1.5 Overall Index of Economic Turbulence IND TUN MAR COL JOR 1 IDN VNM OMN EGY BRA AUS PHL PER CHL ISR ECU 0.5 POL CHE ZAF KOR NOR MYS NZL CAN FRA PAK HKG USA GRC 0 THA TWN PRT DEUGBR SWE DNK TUR CZE ESP SGP BEL NLD MEX AUT ARG SVN ITA HRV -0.5 KAZ SER JPN FIN LUX RUS HUN IRL ROM BGR -1 -1.5 EST -2 LTU LVA UKR ICE -2.5 -3 0 10 20 30 40 50 60 70 80 90 100 GDP Per Capita (in US$ Thousands, 2004-2008 Average) Table 1. Trade Openness and Growth Collapses Dependent Variable: Change in GDP Growth Rates Between 2007 and 2009 (1) (2) (3) (4) (5) (6) De Facto Trade Oppeness 0.0402*** 0.0294*** (0.0114) (0.00997) Tot. Assets + Tot. Liab. (%GDP) 0.0160 -0.121 (0.107) (0.113) Private Consumption (%GDP) -0.0766* 0.00646 (0.0417) (0.0290) Real GDP Growth in 2007 1.090*** 1.003*** 0.904*** (0.182) (0.148) (0.170) Observations 108 108 94 94 131 131 R-squared 0.104 0.431 0.001 0.345 0.062 0.371 Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 Controls: GDP per capita 9 10 Table 2. Composition of Trade and Financial Flows and Growth Collapses Dependent Variable: Change in GDP Growth Rates Between 2007 and 2009 (1) (2) (3) (4) (5) (6) (7) (8) Export of Nat Resources/Tot. Exports -0.0413* -0.0406** (0.0221) (0.0202) Export of Manufacturings/Tot. Exports 0.0441** 0.0582** (0.0218) (0.0231) Herfindhal Index of Export Partners -4.235 -2.475 (3.266) (2.888) (FDI+Equity Assets+Liabilities)/GDP -0.509* -0.473* (0.295) (0.274) (Debt Assets+Liabilities)/GDP 0.489** 0.297 (0.219) (0.187) Total International Reserves/GDP 6.109** 2.787 (2.333) (3.406) Real GDP Growth in 2007 0.992*** 1.037*** 0.926*** 0.991*** (0.178) (0.186) (0.149) (0.150) Observations 95 95 95 95 140 140 92 92 R-squared 0.029 0.328 0.037 0.361 0.038 0.395 0.034 0.364 Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 Controls: GDP per capita Table 3. The Resilience of the Financial System and Growth Collapses Dependent Variable: Change in GDP Growth Rates Between 2007 and 2009 (1) (2) (3) (4) (5) (6) (7) (8) Loan-to-Deposit Ratio 4.881*** 3.760*** (1.704) (1.300) "Excess" Credit Growth: 19.96*** 11.46** Avg Credit Growth 07-06 vs Avg Credit Growth 02-05 (7.258) (5.331) Average Real Credit Growth 2001-2007 29.81*** 22.97*** (9.092) (8.192) Priv Credit/GDP -0.0253 -0.0115 (0.0195) (0.0149) Real GDP Growth in 2007 1.003*** 0.778*** 0.737*** 1.242*** (0.162) (0.150) (0.174) (0.401) Observations 104 104 50 50 50 50 69 69 R-squared 0.232 0.507 0.188 0.407 0.296 0.502 0.113 0.309 Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 Controls: GDP per capital, Openness 11 Table 4. The Role of Macroeconomic Framework Indicators and Growth Collapses Dependent Variable: Change in GDP Growth Rates Between 2007 and 2009 (1) (2) (3) (4) (5) (6) (7) (8) Fiscal Policy Indicators Fiscal Procyclicality 5.420*** 2.017 (1.892) (1.575) Differences in Growth Rates: 14.40*** 7.721* Govt. Expenditures - Govt. Revenues (5.222) (4.522) Monetary Policy Indicators Level of YoY CPI 0.486*** 0.205 (0.160) (0.129) Stable Flexible Exchnage Rate Regime between 03-07 1.135 0.144 (0.947) (0.825) Real GDP Growth in 2007 1.026*** 1.055*** 1.193*** 1.087*** (0.204) (0.203) (0.229) (0.190) Observations 97 97 100 100 102 102 108 108 R-squared 0.171 0.444 0.159 0.452 0.201 0.453 0.113 0.431 Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 Controls: GDP per capital, Openness Table 5. Correlation among Crisis Variables Correlation Analysis of Crisis Indicators GDP IPI Exports REER Stocks IICCR GDP 1 IPI 0.4395 1 Exports 0.3109 0.086 1 REER 0.2848 0.049 0.2198 1 Stocks 0.1204 0.0531 -0.0072 0.0788 1 IICCR 0.3008 0.2395 0.0925 0.1655 0.058 1 12 Table 6. Top 20 Countries Hardest Hit by the Crisis The ranking was based on the aggregate index of the severity of crisis Top-20 Countries Index of Economic Rank Region Country Turbulence 1 ECA Ukraine -2.39 2 IND Iceland -2.36 3 ECA Latvia -2.05 4 ECA Lithuania -2.03 5 ECA Estonia -1.65 6 ECA Bulgaria -0.82 7 ECA Romania -0.77 8 ECA Russia -0.70 9 IND Ireland -0.69 10 ECA Hungary -0.66 11 ECA Serbia -0.61 12 ECA Kazakhstan -0.57 13 IND Finland -0.56 14 IND Luxembourg -0.47 15 IND Japan -0.47 16 ECA Croatia -0.43 17 IND Italy -0.39 18 ECA Slovenia -0.38 19 AMER Argentina -0.36 20 AMER Mexico -0.28 13