The effect of 1997 financial crisis on the economicgGrowth of Asian countries

(1)

Department of Economics Faculty of Economics and Business Universiteit van Amsterdam

Bachelor program: International Economics Specialization: Macroeconomics

The Effect of 1997 Financial Crisis

on the Economic Growth of Asian Countries

Bachelor Thesis (10 EC) Author: Vidita Vergia Verena Student number: 6121934

Supervisor: Gonul Dogan Date: August 17, 2011

(2)

ABSTRACT

From the 1970s, and especially in the early and mid-1990s, there was growing international recognition of the sustained rapid economic growth and industrialization of the East and South East Asian region. However, in 1997, some Asian countries – Indonesia, Thailand, South Korea, and Philippines – experienced sharp banking and currency crisis. The shortfalls of GDP growth were intense in comparison with four other East Asian countries that were less affected by the financial crisis. This thesis suggests that in the short run, Asian financial crisis in 1997 leads to higher economic growth to related countries, especially to countries that were injured by the crisis the most. However, panel analysis shows that, in long run, a combined banking and currency crisis in Asia generally leads to relatively lower growth rates even though there is evidence shows countries that hurt more by the crisis have a relatively higher growth than countries that do not. This thesis also points out some important lessons that can be drawn from the crisis to prevent a similar problem in the future.

(3)

CHAPTER 1 Introduction

During the last few decades, economists and policymakers around the world have closely followed the economic success of the countries in East and Southeast Asia. From 1965 to 1996, the gross domestic product (GDP) of the countries in this region grew at the rate at least twice as high as the growth rate in any other comparable region in the world. After the early success of Japan, most of the recent achievement of this success can be attributed to the growth performance of the four countries (Hong Kong, Singapore, South Korea, and Taiwan) and the three newly industrialized countries of Southeast Asia (Indonesia, Malaysia, and Thailand). In the latter three countries, average income increased more than quadrupled from 1965 to 1996, and income in South Korea rose sevenfold. This region attracted almost half of total private capital inflows to developing countries, which totaled to US$100 billion in 1996.

Besides generating economic growth, these aforementioned countries have also been successful in increasing life expectancy, extending education, and reducing poverty. A set of common factors can be identified as the source of their remarkable performance. These include outward-oriented, market-friendly government policies complemented with macroeconomic stability, agricultural development, investment in human resources, mobilization of savings, high rates of productive in investment, industrial policy designed to close the technological gap, and a strong emphasis on egalitarian policies. The most important and influential document which attempted to explain the rapid growth, structural change and industrialization of much of East Asia is The East Asian Miracle study (EAM) published by the World Bank in 1993.

Against this background, it is reasonable to say that the scale and depth of the Asian economic and financial crisis in 1997 to 1998 was quite surprising. Some economists regarded the claims of an Asian economic miracle as overstated, arguing that these countries were bound to run into diminishing returns eventually. However, no one had anticipated the magnitude of

(5)

events. Instead of a gradual slowdown in economic growth, these countries experienced the collapse of domestic asset markets, widespread bank failures, and bankruptcies of business firms.

In this thesis, I mainly analyze the crisis’ impact on the economic growth of the Asian countries. Firstly, I describe the cause of Asian crisis and how it spread throughout the region, and then the third chapter explains some economic theories about the relationship between financial crisis and economic growth. The thesis, afterwards, defines two groups: control group and experimental group, and compare their economic growth. A panel regression is then used to show the long run effect of the crisis to economic performance of treatment group. Lastly, the thesis concludes with a discussion on the effect of financial crises on economic growth and some lessons for crisis prevention.

(6)

CHAPTER 2 The Asian Crisis

While the East Asian economies continued to achieve rapid economic growth in the 1990s, there were growing imbalances and weakness at both the microeconomics and macroeconomic levels. Most importantly, there was a rapid buildup of short-term external debt into weak financial systems. This was possible because of East Asia’s successful track record of attracting foreign credits, and because partial financial market liberalization in the region opened new channels for the entry of foreign capital. The inflows led to appreciating real exchange rates, a rapid expansion of bank lending, and in particular, increasing vulnerability to a reversal in capital flows. When capital inflows did wane in late 1996 and early 1997, a series of missteps by Asian governments, market participants, the IMF, and the international community resulted in a financial panic, which led to crisis.

The magnitude and suddenness of the financial reversal can be seen in Table 1 in Appendix, which records net capital flows to the five East Asian crisis economies: Indonesia, Korea, Malaysia, the Philippines, and Thailand. According to Table 1, private net inflows to these five countries soared, rising from $40.5 billion in 1994 to $93.0 billion in 1996. However, in 1997 the long period of inflow abruptly reversed, with a net outflow of around $12.1 billion. The remarkable and unexpected swing of capital flows of $105 billion (from $93 billion inflow to $12 billion outflow) represents around 11 percent of the pre-crisis dollar GDP of these five countries. Direct investment remained constant at around $7 billion. The rest of the decline has come from a $24 billion fall of portfolio equity and a $5 billion decline in non-bank lending.

Reflecting the macroeconomic conditions in the region, national stock markets started to drop and currencies came under speculative pressures in the first months of 1997. The first currency to come under attack in the spring was the Thai baht. Once the baht started to depreciate in July 1997, the currencies that came under speculative pressure were those of countries with economic fundamentals and export structure similar to the ones of Thailand such as Malaysia, Indonesia, and the Philippines.

(7)

By the end of the summer, the combined effective devaluation of about 30% of the currencies of Thailand, Indonesia, the Philippines, and Malaysia had a strong negative impact on other currencies in the region. The Singaporean currency that was formally on a float started to depreciate because of the sharp deterioration of Malaysian ringgit. Relatively small depreciations occurred in the Singaporean dollar, starting in August, and the New Taiwan dollar, starting in October. The South Korean won depreciated substantially starting in November. Japan also had a moderate devaluation between July 1997 and January 1998. No significant devaluations took place in China, which has remained relatively insulated from world financial markets, and Hong Kong, which maintained a currency board linked to the U.S. dollar.

During 1998, forecast of the economic slowdown in the crisis countries have been steadily revised downward. The economic recession in East Asia was spreading from the crisis countries (Korea, Indonesia, Thailand, and Malaysia) to Hong King, Singapore, the Philippines, and Taiwan. The Indian subcontinent was fragile, Pakistan was having serious external balance and debt problems, and India was facing economic difficulties. More crucially, the economic conditions in Japan, the leading economy in the region, had deteriorated, and this country was in need of difficult banking and structural reforms, let alone an effective macroeconomic policy to recover from the long period of stagnation. Policy failures leading to a further weakening of the yen could undermine the stability of the currencies of China and Hong King, triggering a further round of stagflationary competitive devaluations in the entire Asian region. Different countries were hit to different degrees, and were adopting different policies to try to avoid a crisis. China and Hong Kong maintained their pegged exchange rates and did not devaluate their currencies. After the crisis, Malaysia decided to control capital flow (especially outflow) while South Korea encouraged foreign capital inflow (especially foreign direct investment). On the other hand, Thailand, Indonesia, and South Korea were three countries that received IMF aid packages after the crisis.

The Asian financial crisis came as a surprise to policymakers, investors, and academics alike. Yet, many agreed not only that the crisis could have been expected, but also that, to a great extent, it might have been avoided. Investors and policymakers missed some warning signs of unsustainable lending booms, such as high corporate debt-to-equity ratios. In 1996, those ratios were respectively 310% in Indonesia and 518% in Korea. High ratios of short-term debt to

(8)

central bank reserves, an important measure of a country’s overall external foreign currency liquidity, were another red flag. In 1996, this ratio was 177% in Indonesia and 193% in Korea. However, some symptoms common in previous crises, such as excessive current account and budget deficits, were missing. Importantly, prior to the Asian financial crisis, early warning systems focused on government external finances and ignored private debt stocks that could become public liabilities because of implicit guarantees. For these reasons, the early warning systems did not sound alarms.

(9)

CHAPTER 3 Theoretical Background

3.1 Cerra and Saxena (2008)

Cerra and Saxena systematically document the behavior of output following financial and political crises in a large set of 190 countries. They try to formally analyze the impact of financial and political shocks on output in a broad set of countries, particularly whether output losses are recovered. Financial shocks comprise currency, banking, and twin financial crises. For political shocks, they examine civil wars, deterioration in the quality of political governance, and twin political crises comprising both shocks.

In the attempt to examine the impact of financial and political crises on output, Cerra and Saxena construct qualitative indicators of financial and political crises and estimate impulse response functions to the shock. They test the statistical relationship between growth and the shock by estimating impulse response functions for each different type of shock. In particular, they estimate a univariate autoregressive model in growth rates, which account for the nonstationarity of output and for serial correlation in growth rates. They estimate the following univariate autoregressive (AR(4)) model in growth rates:

where g is percentage of change in real GDP, D is dummy variable indicating a financial or political crisis, and coefficient δs measures the impact of crises on g.

This paper documents that the large output loss associated with financial crises and some types of political crises is highly persistent. Impulse response functions show that less than 1 percentage point of the deepest output loss is regained by the end of ten years following a currency crisis, banking crisis, deterioration in political governance, or twin financial crises. Of the large negative shocks examined, a partial rebound in output is observed only for civil wars.

(10)

Moreover, the magnitude of persistent output loss ranges from around 4 percent to 16 percent for the various shocks.

The paper also provides some suggestive evidence of causality. Financial crises are associated with growth optimism. Forecasts of economic growth, whether measured by projections from a univariate autoregressive model or by consensus forecasts of financial experts, tend to be higher than actual growth outturns. This evidence cannot, however, rule out the possibility of a third factor that precipitates a crisis and leads to a reversal of growth optimism.

3.2 Ranciere, Tornell, and Westermann (2008)

Ranciere et. al suggest that on average, countries with occasional financial crises have grown faster than countries with stable financial conditions. They measure the incidence by the skewness of credit growth since financial crises are realizations of downside risk. Unlike variance, negative skewness isolates the impact of the large, infrequent, and abrupt credit busts associated with crises. They find a robust negative link between skewness and GDP growth, that means there are a positive effect of systemic risk on growth.

To explain this finding, they present a model in which contract enforceability problems generate borrowing constraints and impede growth. In financially liberalized economies with moderate contract enforceability, systemic risk taking is encouraged and increased investment. This leads to higher mean growth but also to greater incidence of crises. In the data, the link between skewness and growth is indeed strongest in such economies.

Their finding that fast-growing countries tend to experience occasional crises sheds light on contrasting views of financial liberalization. In one hand, financial liberalization leads to excessive risk taking, an increasing volatility, and more frequent crises. However, liberalization strengthens financial development and contributes to higher long run growth. Ranciere et. al also indicate that although liberalization does lead to systemic risk taking and occasional crises, it also raises growth rates, even when the costs of crises are taken into account. Nevertheless, the fact that systemic risk can be good for growth does not mean that it is necessarily good for welfare. Systemic risk taking is not a strategy for increasing growth that can be pursued in the very long run. Once a country becomes rich enough, it must shift to a safe path.

(11)

CHAPTER 4 Prior Empirical Experience

In this thesis, I consider the behavior of economic growth in China, Hong Kong, Indonesia, Japan, South Korea, Philippines, Singapore, and Thailand. These eight economies break down naturally into two groups depending on the extent to which they were impacted by the Asian financial crisis of 1997. The first group of four countries – Indonesia, Philippines, South Korea, and Thailand– experienced nominal currency depreciations of more than 50% from July 1997 to early 1998 and their nominal interest rates (determined primarily by forward exchange rates) reached at least 25%. Subsequently, I refer to this group as treatment group. The other four East Asian economies, that I refer to control group, experienced nominal depreciation of less than 25% and their nominal interest rates remained below 20%.

4.1 Control Group

Figure 1 on Appendix shows the annual GDP growth rate per capita for each economies of the control group from 1961 to 2009. Countries of control group were not really affected by the sharp economic contractions in 1997-1998. Per capita growth during 1998 was 6% in China, -6% in Hong Kong, -2% in Japan, and -1% in Singapore.

In 1999-2000, economic recoveries occurred, and the per capita growth rates were positive in all economies. For control group countries, the rates were 7% in China, 3% in Hong Kong, 0.2% in Japan, and 7% in Singapore. In the longer period after Asian crisis, the growth rates are quite stable at around 11% in China, 5.5% in Hong Kong, 2% in Japan, and 8% in Singapore, until another crisis occurred in 2008.

4.2 Treatment Group

Countries in treatment group were a lot more affected by the hit. As shown in Figure 2, real GDP per capita fell by 18% in Indonesia, 11% in South Korea, 6% in Philippines, and 9% in Thailand. After economies recovered, the annualized per capita growth rates among the four

(12)

crisis countries were 1% in Indonesia, 9% in South Korea, 3% in Philippines, and 4% in Thailand. Then, all growth rates had a small decrease during 2001. However, between 2002 and 2007, the economies for all countries tend to slightly increase to about 4%-6%.

4.3 Cross Country Analyses of Economics Outcome

Panel Regression for Growth Rate

Variables Model 1 Model 2

ln interest rate 0.60629 0.883007 (0.177) (0.035) treatment 3.742656 3.20361 (0.043) (0.069) time -2.33721 -1.55169 (0.039) (0.070) treatment * time 1.07809 - (0.235) - ln interest * treatment -1.49464 -1.49464 (0.026) (0.027) ln interest * time 1.012407 1.458974 (0.086) (0.161) constant 0.8003992 0.4076392 (0.316) (0.577) R-square 0.6967 0.6422 Adj R-square 0.4945 0.4632

I run two panel regressions with GDP growth rate as dependent variable. The first explanatory variable is ln interest rate, which is the logarithm of interest rate, as one parameter of the financial crisis. Moreover, logarithm here is used to specify non-linear relationship between interest rate and economic growth (Stock and Watson 2007). The next two variables are dummies. The first one is treatment, which has a value of 1 if the country is part of treatment group and 0 otherwise, and the second one is time, that has a value of 1 if the growth rate is after 1997 and 0 otherwise. Next in Model 1, there is treatment*time, which is multiplication of treatment and time variables. The last two variables study the interaction between interest rate and those two dummy variables.

(13)

Result shows that interest rate has a positive coefficient related to economic growth. Yet, this does not mean causality that an increase in interest rate gives a rise in economic growth. As contractionary monetary policy explains, by increasing interest rate under its control, a monetary authority can reduce money supply because higher interest rates encourage savings and discourage lending. Both of these effects decrease the size of the money supply and eventually output growth. However, this regression result suggests that a high level of nominal interest rate during 1997 was associated with a positive effect on growth. A high level of interest rate is associated with currency depreciation in country, which then attracts foreign investors to invest so eventually leads to a positive effect on economic growth. However, this argument is not strong because an increase in growth only relying on currency depreciation is only temporary. This implies that a higher interest rate in a country shows that the country is hurt more by the crisis but not significantly. Thus, this finding is in line with the next independent variable, treatment, which also has a positive coefficient. This means a country from treatment group tends to grow faster than a country from control group because it allows more reforms to take place. This result is in accordance with the finding of Ranciere et al (2008). The time variable is also a binary variable. The negative sign means countries tend to have a dropped growth rate after the crisis since crisis increase volatility in the long term. Meanwhile, treatment*time, which is multiplication of treatment and time variables, has a positive coefficient, which means countries in treatment group have a faster growth after crisis in comparison to growth rate before crisis and growth rate of countries in control group. However, this variable is not significant, so I will drop it later and run another regression in Model 2. The next two variables is multiplication of interest rate and two dummy variables, treatment and time. The first variable is significantly negative for growth indicating that countries in treatment group that have higher interest rate are likely to gain lower level of growth rate compared to countries in control group. On the other hand, the last variable has a marginally positive effect on growth, which suggests that a higher level of interest rate after crisis would lead to a higher economic growth.

After dropping an insignificant variable (in Model 2), treatment*time, there is a slight change in other variables. The R-square drops a little, and so does adjusted R-square. Some variables like interest rate, time dummy, and multiplication of interest rate and time dummy, have bigger coefficients than before, while treatment dummy variable has a decreased coefficient

(14)

For example, coefficient for interest rate times treatment is -1.49. This means that before crisis there is a significant difference in the effect of interest rate between treatment and control group countries. Treatment group countries have a significantly lower growth rate, which leads to a negative overall effect of interest rate (0.88 - 1.49 = -0.61) in comparison to control group countries that only have the interest rate effect as much as 0.88. However, after crisis, according to Model 2, the overall effect is positive for both groups. This is shown by the coefficient of interaction between interest rate and time variable that has a coefficient of 1.45, which means interest rate is significantly higher after crisis and the overall effect is positive (0.88 + 1.46 = 2.34 for control countries, and 0.88 + 1.46 – 1.49 = 0.85 for treatment countries). This result also suggests that the effect of interest rate after crisis is more significant in control group than in treatment group since for control countries 1% increase in interest rate leads to 2.34% rise in growth rate, which is much higher than a 0.85% increase in growth rate for treatment countries.

To answer whether crisis affect treatment and control group differently, I investigate all coefficients related to treatment variables. Treatment variable itself has a significantly positive coefficient, 3.20. This implies that, in general, countries in treatment group significantly grow faster than the ones in control group. On the other hand, the effect of crisis in general is negative. The coefficient of -1.55 in time variable suggests that after crisis countries have significantly lower growth rate than before.

To make it clearer, I will try to estimate the growth rate of Indonesia (treatment group country) and Japan (control group country). Assume interest rate is xi for simplicity. Before

crisis, growth rate of Indonesia is 0.4 + 0.88logxI + 3.20 - 1.49logxI = -0.61logxI + 3.60, while

growth rate of Japan is 0.88logxJ + 0.40. After crisis, growth rate of Indonesia is 0.4 + 0.88logxI -

1.55 + 3.20 - 1.49logxI + 1.46logxI = 0.85logxI + 2.05, and growth rate of Japan is 0.4 + 0.88logxJ

- 0.10 = 0.88logxJ + 0.30. This suggest that the net effect of crisis in treatment group is 0.85xI +

2.05 - 0.61logxI - 3.60 = 0.24logxI - 1.55 and in control group is 0.85logxI + 2.05 - 0.88logxJ -

0.40 = 0.03logxJ – 1.65. Meanwhile, the overall effect of classification of countries into two

categories, treatment and control groups, is 0.85logxI + 2.05 - 0.88logxJ - 0.30 = 0.85logxI -

0.88logxJ + 1.75 after crisis. Assuming interest rate is the same in both groups, countries in

treatment group still have a relatively higher growth than countries in control group (0.85- 0.88 + 1.75 = 1.72). This might also happen in the long term because control group countries like

(15)

Japan are relatively more stable than treatment group countries like Indonesia, so they relatively have less economic growth than crisis countries. However in reality, interest rate is not the same in both groups so this finding might not be applicable in all case of countries. For example, China and Singapore are part of control group countries, but their growth rates are still higher than treatment group countries. For China, this might happen because it has an abundant number of labors, so the production cost is much lower compared to other countries. Thus, China can stimulate its economic growth through domestic production even though their interest rate is relatively smaller. Meanwhile for Singapore, it has been a center of international market in Asia because it has an excellent infrastructure and safe environment, so it will still always attract investors and therefore have a high level of growth rate anyway.

Hence, this regression shows that in the long term, growth rate fails to rebound and crisis hurts economy. However, countries in treatment group will still grow relatively faster than countries in control group if they have the same level of interest rates.

4.4 Lessons for crisis prevention

There are some lessons we could learn from Asian financial crisis to prevent a repetition of another Asianflu-type crisis.

First lesson concerns banking supervision and financial-sector infrastructure. Bank regulators were encouraged to require greater transparency and supervise lending activity more strictly, playing particular attention to currency and maturity mismatches. Moreover, highly leveraged institutions are required to improve risk assessment and reduce leverage ratios. There are also needs to lengthen the maturity of capital control and alter the composition of foreign capital inflows so that more investment came in as equity and less as debt.

Next, central banks in emerging market countries have only a very limited ability to extricate their countries from financial crisis. Indeed, a speedy recovery is likely to require foreign assistance because liquidity provided from foreign sources does not lead to any of the undesirable consequences that result from the provision of liquidity by domestic authorities. Thus, an international lender of last resort was needed to resolve crises; in this Asian crisis case, IMF had an important role. However, we should be careful because without appropriate

(16)

conditionality for the lending, moral hazard created by the operation of an international lender of last resort can promote financial instability.

Third, there is a danger of pegging exchange rates. When a successful speculative attack occurs, the decline in the value of domestic currency is larger, more rapid, and more unanticipated than when depreciation occurs under a floating exchange-rate regime. Therefore, pegged exchange rate regime may increase financial instability in emerging market and make financial crises more likely.

(17)

CHAPTER 5 Concluding Remarks

In this thesis, I suggest the effect of 1997 Asian financial crisis on the economic growth of its countries. The Asian financial crisis was associated with a sharp reduction of economic growth in East and Southeast Asia, especially in the four countries that were most directly affected by the crisis. Rates of economic growth in Asian countries have rebounded in 1999-2000, but there is no evidence showing the permanence of this recovery. The failure of the growth rates to rebound significantly in the countries in general suggests that the crisis had a long term adverse effect. However, crisis countries grow relatively faster than others assuming all countries have same level of interest rate.

The crisis has not only been disastrous for the economies of countries in this region, but it has also put the global financial system under tremendous stress. Several lessons can be derived from the crisis. First lesson concerns bank regulation and financial-sector infrastructure. Second, there is a strong rationale for an international lender of last resort. And last, pegged exchange rate regimes are dangerous strategy for emerging market countries and make financial crises more likely.

(18)

BIBLIOGRAPHY

Barro, R. (2001) “Economic Growth in East Asia Before and After Financial Crisis” NBER Working Paper 8330,6.

Blejer, M. (2006) "Economic Growth and the Stability and Efficiency of the Financial Sector." Journal of Banking & Finance 30.12: 3429-432.

Cerra, V., and S. C. Saxena. (2008) "Growth Dynamics: The Myth of Economic Recovery." American Economic Review 98.1: 439-57.

Chowdhury, A. R. (1999) “The Asian Currency Crisis: Origins, Lessons, and Future Outlook.” UNU World Institute for Development Economics Research. Action 47.

Corsetti, G., P. Pesenti and N. Roubini. (1998) “What Caused the Asian Currency and Financial Crisis? Part II: The Policy Debate.” NBER Working Paper 6834.12.

Crafts, N. (1999) “East Asian Growth Before and After Crisis” IMF Staff Papers 46.2: 139-66. Kroszner, R., L. Laeven, and D. Klingebiel. (2007) "Banking Crises, Financial Dependence, and

Growth." Journal of Financial Economics 84.1: 187-228.

Mishkin, F. S. (1999) "Lessons from the Asian Crisis." Journal of International Money and Finance 18.4: 709-23.

Ranciere, R., A. Tornell, and F. Westermann. (2006) "Decomposing the Effects of Financial Liberalization: Crises vs. Growth." Journal of Banking & Finance 30.12: 3331-348. Ranciere, R., A. Tornell and F. Westermann. (2008) “Systemic Crises and Growth.” Quarterly

Journal of Economics 123.1: 359-406.

Stock, J. H., and Watson, M. W. (2007) “Introduction to Econometrics” Pearson Education, Inc. 2nd edition: 267-274.

(19)

APPEDIX

Figure 1. Control Group

GDP Growth Rate per Capita in China

(20)

GDP Growth Rate per Capita in Japan

(21)

Figure 2. Treatment Group

GDP Growth Rate per Capita in Indonesia

(22)

GDP Growth Rate per Capita in South Korea

(23)

The effect of 1997 financial crisis on the economicgGrowth of Asian countries