The Asian financial crisis: The role of derivative securities trading and foreign investors in Korea

  • Published on

  • View

  • Download

Embed Size (px)


<ul><li><p>The Asian nancial crisis: The roleof derivative securities tradingand foreign investors in Korea</p><p>Eric Ghysels a,*, Junghoon Seon b,1</p><p>a Department of Economics, University of North Carolina and</p><p>the Department of Finance, KenaneFlagler School of Business, Chapel Hill,</p><p>NC 27599-3305, USAb Korea Securities Research Institute, 33 Youido-dong, Yongdungpo-gu,</p><p>Seoul 150-977, South Korea</p><p>Abstract</p><p>This paper is part of a larger research program pertaining to the role of derivatives duringnancial crisis and also part of the research pertaining to the causes of the Asian nancialcrisis. The Korean market is studied because of two reasons: (1) it is an important example of</p><p>the Asian nancial meltdown and (2) there is a detailed data set available of all transactions bydierent types of protagonists, including foreign investors. The results in this paper indicatethat futures markets and trading by foreign investors played a key role during the Korean</p><p>stock market turbulence in 1997. 2005 Elsevier Ltd. All rights reserved.</p><p>JEL classication: G1; F3</p><p>Keywords: Financial crisis; Foreign investors; Derivative securities; Herding</p><p>Journal of International Money and Finance</p><p>24 (2005) 607e630</p><p>* Corresponding author. Tel.: C1 919 966 5325; fax: C1 919 966 4986.E-mail addresses: (E. Ghysels), (J. Seon).</p><p>1 Tel.: C82 237710680; fax: C82 237710658.</p><p>0261-5606/$ - see front matter 2005 Elsevier Ltd. All rights reserved.doi:10.1016/j.jimonn.2005.03.002</p></li><li><p>1. Introduction</p><p>The role of derivatives during nancial market meltdowns is still not wellunderstood. The Brady Report, written by a Presidential task force in the wake of theOctober 1987 stock market crash, blamed trading in stock index futures for thenancial crisis. According to the report, portfolio insurers tried to cover their equityexposure by mechanical, price-insensitive selling stock market index futures. Theiractions drove down futures prices and created arbitrage opportunities which led indexarbitrageurs to implement reverse cash-and-carry strategies, i.e. they bought futuresand sold the underlying stocks. The dynamic interaction between portfolio insurersand index arbitrageurs was lethal and caused a spiral downward tumbling of prices.The events of October 1987 were characterized by the BradyReport as a vivid exampleof the dire consequences derivative securities can have during a nancial crisis.2</p><p>The Asian nancial crisis is certainly not a carbon copy of the October 1987 crash.There are similarities and dierences which make a comparison interesting. The mostblatant dierence is the role played by foreign investors. Several authors, includingRadelet and Sachs (1998) and Kim and Wei (1999), have put the blame for the Asiancrisis on foreign investors. In particular, Radelet and Sachs argue that the sudden pull-out of foreign investments exacerbated the crisis by causing a nancial panic combinedwith policy mistakes by Asian governments and the International Monetary Fund.Along the same lines, Kim and Wei argue that the nancial crisis in the East Asiancountries should, to a large degree, be ascribed to the panic reaction and herd behaviorof foreign investors rather than economic fragility of those countries. Choe et al.(1999) refute the idea of market destabilization by foreign investors. They takeadvantage of a unique data set covering transactions by domestic and foreigninvestors on the Korean Stock Exchange (henceforth KSE). The KSE trading systemis very similar to the Paris Bourse screen-driven market, studied by Biais et al. (1995).The pure order-driven market makes identication of foreign and domestic traderseasy, particularly since the former have to register before they are allowed to trade.</p><p>This paper examines two questions. First, what was the role of derivativesecurities in the Korean stock market crash. Second, to what extend did derivativestrading by either domestic or foreign investors, or both together, exert a destabilizinginuence during the KSE October 1997 crash. Despite the fact that futures andoptions markets are typically not as well developed in emerging economies, Koreahad very active futures trading. Established only 2 years prior to the crisis, themarket for the KOSPI 200 index futures contracts grew fast and established itselfquickly. Moreover, in the Korean case, futures contracts had the appealing featurethat foreign ownership restrictions were removed before the crisis.3 Futures and</p><p>2 Much has been written about the Brady Report, both supportive and critical. For further details see</p><p>Kleidon and Whaley (1992), Blume et al. (1989) and Santoni (1988).3 To be more precise, foreign ownership restrictions were removed for futures and options in July 1997,</p><p>i.e. 3 months before the crisis. More recently, namely in May 1998, they were also lifted for equities. Choe</p><p>et al. (1999) provide a detailed discussion of the restrictions which were in place for non-resident equity</p><p>608 E. Ghysels, J. Seon / Journal of International Money and Finance 24 (2005) 607e630holders.</p></li><li><p>options on the KOSPI 200 market index are traded on the KSE. The data collectedfrom the electronic trading system enables us to investigate the pattern and impact oftransactions by domestic (institutions and individuals) and foreign investors. Hence,similar to the equity trading study of Choe et al., we can identify the contractsbought and sold by domestic institutional and individual investors as well as non-residents.4 Choe et al. (1999) examine buy and sell strategies of domestic and non-resident traders thoroughly and do not nd any evidence supporting the idea thatforeigners had a destabilizing eect during the crisis. To make a direct comparisonbetween equity and futures trading we reexamine equity trading focusing exclusivelyon the 200 stocks of the KOSPI 200 index.5 Many of the arguments put forward bythe opponents of nancial liberalization apply to equity as well as derivativestrading. For instance, the argument that foreign investors act like a herd applies tostock buy and sell strategies as well as the long and short positions in futurescontracts.</p><p>This paper is part of a larger research program pertaining to the role ofderivatives during nancial crisis and also part of the research pertaining to thecauses of the Asian nancial crisis. Many of the key questions we try to address arethe same as in Radelet and Sachs (1998), Kim and Wei (1999) and Choe et al. (1999).Many of the techniques used in the paper are closely related to those presented inChoe et al. Our study is, however, not simply adding derivatives market data to theiranalysis. New results pertaining to equity trading are uncovered in conjunction withthe derivatives trading ndings. Namely, we examine extensively the intra-dailypatterns for each group of market participants. The KSE has dierent tradingsessions throughout the day which features dierent market microstructures andhence price discovery mechanism.</p><p>We rst document that foreign investors increased their presence in the futuresmarket. This is understandable since restrictions were lifted. Given this observation,we test several hypotheses pertaining to the role of stock index futures trading byforeign investors during the crisis. The rst hypothesis is whether herding and priceimpacts of foreign investors increased in the futures market. The second hypothesisis whether feedback trading strategies were used by foreign investors and werecontributed to the increase in their herding. The third and nal hypothesis is whetherstock index futures trading by foreign investors exerted a destabilizing inuence onthe stock market, i.e., selling pressures by foreign investors in the futures marketwere transmitted to the cash market inducing stock sales by domestic institutions.</p><p>The paper is organized as follows. In Section 2, we examine the role of the futuresmarket during the KSE crisis. In Section 3, we describe the trading mechanism of theKSE and the nature of our data set in order to dissect and separate the inuence offoreign investors. In Section 4, we analyze whether foreign and domestic traders herdand whether they pursue positive feedback trading strategies. For domestic market</p><p>4 We will use foreign investors and non-resident investors interchangeably. Strictly speaking, however,</p><p>there are foreigners who reside in Korea and have to register to trade.5 Choe et al. (1999) considered a larger set of stocks, namely 414 of the 762 stocks listed on the KSE at</p><p>609E. Ghysels, J. Seon / Journal of International Money and Finance 24 (2005) 607e630the end of November 1996.</p></li><li><p>participants we make a distinction between institutional and individual traders. Theimpact of foreign investors on prices and market returns in both the stock andfutures market and the linkage between futures and cash market is studied,respectively, in Sections 5 and 6. Finally, Section 7 concludes the paper.</p><p>2. Why Korea, why equities and futures?</p><p>The Asian nancial crisis can certainly not be conned to the crash of equitymarkets. Yet, the focus on the stock market is warranted given the chronology ofevents, as will be explained in this section. We also will explain why futures contractsare the focus of our attention as well when we want to study the inuence of foreigninvestors.</p><p>Korea, Indonesia, Malaysia and Thailand were among the countries hardest hitby the Asian crisis. By the end of 1997, the market capitalization in these fourcountries had fallen 70% to about $188 billion from $637 billion.6 Since Korea is theonly country among these four countries which has stock index futures and optionsmarkets during the Asian nancial crisis, we examine Korea as a representativeexample in studying the role of derivative securities trading and the foreign investors.</p><p>There is evidence that the stockmarket crash was not a result of currency crisis sincethe stock market crash preceded sharp increases in the won/dollar exchange rate andcorporate bond yields (AA). On October 23, 1997, the Korean governmentannounced that it would take over debt-ridden KIA Motors. As Standard &amp; Poorscut Koreas foreign debt rating in response to the news on the next day, the Koreanstockmarket started to show a dramatic downward trend. Over a span of 4 consecutivebusiness days, from October 24 to October 28, the KOSPI 200 market index lost 20.72of its value.7However, as can be seen fromFig. 1, the exchange rate ofKoreanwon andthe yield on 3-year corporate bonds (AA) had restricted movements until November21, when the government announced that it was discussing details of a bailout packagewith the IMF to cope with the rapid deterioration in its external nancial position.</p><p>At the time of the crisis, the KOSPI 200 futures contracts were traded on a liquidmarket with signicant daily volume and open interest, even though it was in-troduced 2 years earlier.8 The underlying asset for the contracts is the KOSPI 200index, which is composed of the 200 most actively traded stocks on the KSE. Thecontract months for futures are March, June, September and December, and thelongest maturity period is 1 year. The last trading day of each contract is the secondThursday of each expiration month. Fig. 2 plots monthly averages of daily volumefor nearest month KOSPI 200 contracts in 1997 and S&amp;P 500 futures contracts</p><p>6 For an elaborate discussion of the stock market crashes during the Asian nancial crisis, see for</p><p>instance Barth and Zhang (1999).7 KOSPI stands for Korean Stock Price Index, the KOSPI 200 index represents the 200 most actively</p><p>traded stocks in Seoul, which take up about 80% of the total market value.8 From its incubation the trading volume of futures contract increased by 203.4%. Comparable gures</p><p>for the Nikkei 225 futures contract are 0.49% (from 1988 to 1990), for the S&amp;P 500 futures contract</p><p>610 E. Ghysels, J. Seon / Journal of International Money and Finance 24 (2005) 607e630106.96% (from 1982 to 1984) and 191.82% for the Han-Seng futures contract (from 1986 to 1988).</p></li><li><p>in both 1987 and 1997 as a fraction of the trading volume of the underlying asset. Priorto July 1997 the KOSPI 200 futures-to-cash market volume was comparable to levelsfor the S&amp;P500 futures and cash market during the 1987 and 1997 crash years. AfterJuly 1997 the monthly averages in Korean markets soared far above levels observedfor S&amp;P 500 futures contracts. At the peak of the crisis the ratio exceeded 13%, morethan double the typical level found prior to the crisis (and levels applicable to S&amp;P500 futures markets). Fig. 3 shows the strong upward trend of KOSPI futurestransactions by foreign investors. The upward trend started in the third quarter of1997 and continued its path through the following year.</p><p>For several reasons, the trading of KOSPI 200 index futures will be the onlyrelevant market for our analysis, though it is among various types of derivativecontracts that were traded before, during and after the Korean nancial crisis. First,trading in options on the KOSPI 200 index, established in June 1997, was still in itsinfancy during the crisis roughly 4 months later. Second, speculative trades on equity</p><p>KOSP</p><p>I</p><p>KOSP</p><p>I 200</p><p>20.00</p><p>30.00</p><p>40.00</p><p>50.00</p><p>60.00</p><p>70.00</p><p>80.00</p><p>KOSPI KOSPI 200</p><p>Oct. 23, 1997Co</p><p>rpor</p><p>ate </p><p>Bond</p><p> Yie</p><p>ld (%</p><p>)</p><p>Exch</p><p>ange</p><p> Rat</p><p>e</p><p>3-year Corporate Bond Yield(AA-) Won / Dollar Exchange Rate</p><p>Nov. 21, 1997</p><p>300</p><p>400</p><p>500</p><p>600</p><p>700</p><p>800</p><p>901</p><p>906</p><p>912</p><p>922</p><p>927</p><p>1004</p><p>1010</p><p>1016</p><p>1022</p><p>1028</p><p>1103</p><p>1108</p><p>1114</p><p>1120</p><p>1126</p><p>1202</p><p>1208</p><p>1213</p><p>1220</p><p>1227</p><p>901</p><p>906</p><p>912</p><p>922</p><p>927</p><p>1004</p><p>1010</p><p>1016</p><p>1022</p><p>1028</p><p>1103</p><p>1108</p><p>1114</p><p>1120</p><p>1126</p><p>1202</p><p>1208</p><p>1213</p><p>1220</p><p>1227</p><p>5.00</p><p>10.00</p><p>15.00</p><p>20.00</p><p>25.00</p><p>30.00</p><p>35.00</p><p>600</p><p>800</p><p>1000</p><p>1200</p><p>1400</p><p>1600</p><p>1800</p><p>2000</p><p>2200</p><p>Fig. 1. Korean Stock Price Indexes, corporate bond yields and exchange rates.</p><p>611E. Ghysels, J. Seon / Journal of International Money and Finance 24 (2005) 607e630</p></li><li><p>swaps and individual stock options by foreign investors were prohibited by the lawof foreign exchange control at the time of the Korean nancial crisis. Third, currencyand interest rate swaps were seldom initiated by foreign nancial institutions as theywere possible only with real goods transactions.9</p><p>3. The market structure and the data</p><p>We can classify investors active on the KSE into the following three groups:Korean individuals, Korean institutions and foreign investors. Foreign investors arerequired to register with the Financial Supervisory Service. In 1997, there werea total of 6514 foreign investors from 66 countries registered, 4514 (69.3%) wereinstitutional and the remaining 2000 (30.7%) were individuals. Consequently, it willbe more meaningful to compare foreign investors with Korean institutional traders.Choe et al. (1999) discuss in detail the Korean government restrictions which applyto both individual and aggregate foreign investment in stocks and derivatives.According to the restrictions, stock holdings cannot exceed a certain percentage ofthe total outstanding shares of each company and futures cannot exceed the averagedaily open interest for the last 3 months for futu...</p></li></ul>


View more >