1

TWN Global Economy Series

Hot Money and Capital Controlsin Thailand

THAWATCHAI JITTRAPANUN AND SUTHY PRASARTSET

TWNThird World Network

15

3

Hot Money and Capital Controls in Thailand

THAWATCHAI JITTRAPANUN AND SUTHY PRASARTSET

TWNThird World Network

4

Hot Money and Capital Controls in Thailandis published by

Third World Network131 Jalan Macalister

10400 Penang, Malaysia.Website: www.twnside.org.sg

© Thawatchai Jittrapanun and Suthy Prasartset 2009

Printed by Jutaprint2 Solok Sungei Pinang 3, Sg. Pinang

11600 Penang, Malaysia.

ISBN: 978-983-2729-79-2

5

CONTENTS

1. BACKGROUND 1

2. THAILAND’S CAPITAL ACCOUNT LIBERALIZATIONAND CAPITAL FLOWS 3

3. FACTORS CONTRIBUTING TO THE 1997 FINANCIALCRISIS 8a. Explosion of private short-term dollar debt 8b. Weak financial system 9c. Misuse of capital 9d. The state’s ineptitude under global financialization 10e. Money politics – the crisis’s trigger 12f. The outbreak of the crisis 12

4. STABILIZATION AFTER THE 1997 FINANCIAL CRISIS 14a. The current account 15b. Foreign reserves and short-term debt 16c. External indebtedness 18d. Fiscal sustainability 21e. Financial restructuring 22

5. RECOVERY AND SURGE IN CAPITAL INFLOWSAND THE POLICY RESPONSE 23a. Hot money returns 23b. Sterilization 25c. Capital account measures 28

6. A QUANTITATIVE ASSESSMENT OF THEEFFECTIVENESS OF CAPITAL CONTROLS 31a. Capital control index 31b. Short-term capital flows defined 35c. Empirical estimates 37

6

7. CONCLUSIONS 48Policy implications 49

REFERENCES 51

APPENDIX 1: Summary of regulations on the measures toprevent Thai baht speculation 54

APPENDIX 2: The Bank of Thailand’s capital account measures 56

7

ACKNOWLEDGEMENTS

This paper was prepared as part of the Third World Network research projecton financial policies in Asia. The authors are grateful to Yilmaz Akyüz for hisadvice, critical comments and coordination; Professor Teerana Bhongmakapatof Chulalongkorn University for his valuable suggestions; KitskornCharoenpanij for his diligent assistance in compiling the Bank of Thailand’snotifications on capital control measures which were used to calculate thecapital control index; and Lean Ka-Min for his patience and thorough editingskills.

1

WITH increased financial liberalization since the early 1990s, Asian econo-mies have become attractive locations for international capital flows. Theinvestment atmosphere of East Asia has been attractive for many reasons,including spectacular growth rates which have often led to high interest ratesand lured foreign investors in search of high returns, giving rise to assetbubbles. These inflows, together with weak macroeconomic fundamentalsand a fragile financial sector, led to a financial crisis in 1997, producingcollapse of currencies, stock markets and asset prices in several East Asiancountries. Many businesses went under and, as a consequence, millions ofpeople lost their jobs, pushing a significant part of the population below thepoverty line.

As Asian economies started to recover from the crisis in the early yearsof this decade, liquidity in the US began to dry up from dotcom problems.The US policy response through sharp cuts in interest rates and aggressivemonetary expansion helped create a boom in property lending which, even-tually, pushed the US economy into a critical stage with the subprime mort-gage crisis, resulting in a significant weakening of the dollar. As excess li-quidity spilled over to emerging markets, currencies came under upwardpressure. The Thai baht (THB) rose by 16.6% against the dollar in 2006 atthe same time the US subprime problem started to unfold (Figure 1). Even-tually, the Bank of Thailand (BOT), already under heavy pressure from ex-porters, resorted to severe capital controls in December 2006 in order tothwart the dollar influx. The controls shocked financial markets and causedeconomic and political turmoil. The incident served as a wake-up call in

Chapter 1

BACKGROUND

2

alerting all sectors to the spectre of short-term capital flows that had wroughthavoc in Thailand a decade earlier.

The objective of this paper is to analyze the evolution of the capitalaccount regime in Thailand, paying special attention to the effectiveness ofmore recent capital control measures in slowing down the THB appreciationduring 2003-2007. The vector autoregression method is used to investigatethe reaction of hot money to control measures as well as other key factorsthat traditionally interact with capital flows such as covered interest paritydifferentials, the real effective exchange rate, and relative returns in stockmarkets between Thailand and the US.

Figure 1. THB/USD exchange rate

Source: Bank of Thailand

Baht/US dollar

15 Jul

33.49

45

42

39

36

33

30

Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr2005 2006 2007 2008

3

IN the 1980s Thailand adopted a liberal treatment of foreign direct invest-ment (FDI) and portfolio investment although repatriation of the principalof portfolio investment was subject to exchange control. Cross-border bor-rowing by Thai residents was allowed subject to authorization by the BOT.Current account items were mostly liberalized at about the same time, ex-cept income payments such as interest and dividends. In the second half ofthe decade, the authorities reduced tax impediments to inbound portfolioinvestments, especially for purchasing Thai mutual funds. In 1988, Thailandbegan to receive large amounts of capital including short-term funds.

In order to increase access of domestic financial institutions to interna-tional financial markets, the BOT raised the limit on banks’ and financecompanies’ net foreign exchange positions from 20% to 25% of capital in1990. Subsequently, the ceiling on short-term deposits was eliminated. Asthe need for external capital grew, three mutual funds were established inorder to attract foreign portfolio investment. The government also imple-mented a comprehensive financial reform plan in order to restructure thefinancial system with the stated purpose of “coordinating [and] synchroniz-ing several aspects of the reform with the ultimate objectives to enhancecompetitiveness, flexibility, efficiency, and stability of the financial sector.”The BOT “allowed non-residents to hold THB-denominated deposits andresidents to hold foreign currency deposits (subject to certain conditions).”The Investment Promotion Act was revised to attract FDI. Additional mea-sures were also announced that allowed multinational enterprises to hold a100% ownership stake in export firms. In order to enhance financial market

Chapter 2

THAILAND’S CAPITAL ACCOUNTLIBERALIZATION AND CAPITAL FLOWS

4

competition, the ceilings on commercial banks, finance and credit fonciercompanies were removed for deposit and lending rates (Vichyanond, 2001).

The turning point in the Thai liberalization process came with the es-tablishment of the Bangkok International Banking Facilities (BIBF) to “fa-cilitate international borrowing and to encourage capital inflows as a meansof financing the current account deficits” (Traisorat, 1998: 531). It was anambitious policy instrument to promote Thailand as a regional financial centrefor mainland Southeast Asia. Other steps towards liberalization were alsotaken, including creation of non-resident baht accounts at domestic com-mercial banks, lowering reserve requirements and eliminating conditionsfor purchases and transfer abroad of foreign currency by residents. Underthe BIBF financial institutions based in Thailand, both Thai and foreign,were permitted to conduct offshore business in foreign currencies. The twomajor categories of banking transactions, the Out-Out (offshore lending)and the Out-In (local lending of borrowed foreign capital), were allowed toaccept deposits or borrow from abroad in foreign currencies from natural orjuristic persons without establishment or business in Thailand. While theOut-Out businesses could use these funds to extend loans in foreign mar-kets, the Out-In businesses lend these foreign currencies to local markets.

The BOT stipulated that the BIBF must be separate from other bank-ing business. Further, the BIBF business in offshore lending must be sepa-rate from that in local lending. Those in the local lending business could nottransfer or raise funds for offshore accounts. In order to attract capital in-flows, tax incentives were offered for the BIBF businesses, with a corporatetax rate of 10% as compared to 30% for local commercial banking.

However, soon it became clear that the objective of making the BIBFthe financial centre of mainland Southeast Asia would be unrealized as itwas the Out-In business that predominated the BIBF. The non-resident bahtaccounts also contributed to an unprecedented surge in capital inflows alongwith the BIBF transactions and rose from almost nil in 1992 to THB261.1billion in 1994, about half of foreign loan inflows, before tapering off toTHB115.9 billion six months prior to the 1997 crisis (see BOT website,Table EX_XT_014). Concerned about the huge influx of foreign capital,the BOT took additional measures in 1993 to moderate the surge by reduc-

5

ing the ceilings on the net foreign exchange exposure of commercial banksfrom 25% of capital to 20% on the overbought and 15% on the oversold or$5 million, whichever was greater.

However, by the end of 1994, Thailand had fully liberalized its currentaccount transactions and allowed greater freedom for foreign investmenteven though it still maintained certain restrictions on foreign ownership,such as prohibition of foreign companies from being listed on the StockExchange of Thailand (SET) and restriction over foreign ownership in thereal estate and banking sectors (Sangsubhan, 2008). The continued relianceon capital inflows and shortening of maturities of external debt made Thai-land highly vulnerable to instability in global capital flows. The liberaliza-tion also led to increased leverage in the corporate sector (Figure 2) andraised risks in the portfolios of financial institutions.

Figure 2. Composition of capital flows in the BOT’s capital(financial) account

Source: Bank of Thailand

Million USD

30,000

25,000

20,000

15,000

10,000

5,000

0

-5,000

-10,000

-15,000

-20,000

Direct investment

Loans

Equity securities

Currency and deposits

Debt securities

Other liabilities

Trade credits

Financial account

Periods

6

Concerned over the upsurge of Out-In transactions, the BOT intro-duced a variety of measures aimed at moderating foreign-financed lending.In April 1996, for offshore borrowing by domestic financial institutions withmaturities of less than one year, it imposed a minimum 7% reserve require-ment in the form of non-remunerated deposits at the BOT, except for tradeand investment-related loans (Chomthongdi, 2002 and Reyes-Cantos, 2002).However, these did not make much impact on the maturity profile of exter-nal obligations. The share of short-term loans increased from 22% of allforeign loans in 1990 to almost 50% in 1995/96. This was greatly encour-aged by a fixed exchange rate regime whereby the THB was pegged to thedollar at a rate of THB25 to the USD (TDRI, 1997).

From the late 1980s the BOT had followed a policy of absorbing thesurge in capital inflows into foreign reserves. Before 1997, net foreign re-serves grew on average by 54% per annum, almost doubling in some years.The mounting pressures from capital inflows together with rising interestrates started to cause trouble for the Thai economy. Monetary policy during1993-96 was not effective in checking the overheating of the economy re-sulting from capital inflows. The combination of a fixed exchange rate re-gime and free capital mobility placed insurmountable pressures on the BOT,whose reckless defence of the THB paved the way for the 1997 crisis.

When the dollar started appreciating in late 1995, the THB becameovervalued. Along with other factors such as China’s devaluation of therenminbi, Thai exports lost competitiveness in major foreign markets. Thus,the drastic drop in exports and widened current account deficits (Figure 3),together with asset bubbles, maturity mismatches and domestic financialtroubles, started to erode investors’ confidence. As a consequence, the THBeasily fell prey to speculative attacks from February 1997 onwards. Theattempt by the BOT to separate domestic and offshore spot foreign exchangein an effort to stop speculation was not successful. The BOT had to give inand turned to a managed float regime on 2 July 1997. By 31 July 1997, theBOT’s net international reserves had fallen to a paltry $1.1 billion from alevel of over $30 billion a year ago. A few weeks later, Thailand had to signa Stand-by Agreement with the International Monetary Fund (IMF) securinga credit of $17 billion (FSRA, 2002: 649-53). When the THB was left to

7

float, it overshot to THB53.7 per dollar in January 1998. It recovered andstabilized subsequently, moving within a range of THB36-43 per dollar be-tween April 1998 and 2001 (FSRA, 2002).

Source: Bank of Thailand

Figure 3. Current account, capital account and balance of payments(BOP)

BOP & Components

25,000.00

20,000.00

15,000.00

10,000.00

5,000.00

0.00

(5,000.00)

(10,000.00)

(15,000.00)

(20,000.00)

mn

US

D

Err & omiss Curr Acct Cap Acct Overall Bal

8

IN spite of the vast body of literature on the Thai experience, it would beuseful to discuss briefly here the factors that contributed to the 1997 crisis.This would help in an assessment of whether appropriate lessons had beendrawn and put into practice in the management of the more recent surge incapital flows to Thailand.

a. Explosion of private short-term dollar debt

At the centre of attention in discussions on the crisis is the role ofshort-term capital flows. With the establishment of the BIBF in 1993, Thaifirms could have direct access to international financial markets. A notablefeature of capital inflows was not only the acceleration of foreign loans butalso a rising share of short-term loans, especially portfolio investment, non-resident deposits and trade credits. The total foreign debt of Thailand stoodat $29 billion in 1990 but jumped to $65 billion in 1994 and further shot upto $92 billion in 1997, mostly incurred by the private sector. This is a more-than-threefold increase in seven years. The ratio of foreign debt to GDPgrew considerably from 34% to 49% and the share of private debt rose fromabout 60% to around 80% before tapering off to about 73% in 1997. Evenmore frightening is the fact that these loans were not hedged as the THB wasfixed by the monetary authority. A high proportion of short-term unhedgeddollar loans in Thai foreign debt rendered the economy highly vulnerable tospeculative attacks.

Chapter 3

FACTORS CONTRIBUTING TO THE 1997FINANCIAL CRISIS

9

b. Weak financial system

Lacking sound and prudential financial institutions and strong regula-tory bodies, the Thai economy was unable to withstand the heavy load of theinflux of short-term foreign capital thanks to deregulation and financial lib-eralization. During the lending boom, asset quality constantly declined be-cause of shortcomings in banks’ ability to screen projects. Before the 1997crisis, the Thai financial institutions were operating under an illusion of greatsuccess when both banks and finance companies had managed to grow byabout 30% per annum. In fact, such a lending boom entailed considerablevulnerability for the banking sector (Sachs et al., 1996: 162).

The rapid growth of lending was particularly notable among financecompanies, especially in real estate and car loans (with a share of more thanhalf of total loans). Several banks also participated in financing real estateprojects and heavy industries such as petrochemicals, iron and steel and ce-ment which already faced substantial excess capacity, as well as margin loansfor the stock market. When the real estate sector finally succumbed underthe mounting weight of the excess supply and stock market crash (from apeak of 1700 at the end of 1993 to about 830 at the end of 1996), both banksand finance companies were plunged into a distressed state (Vongvipanond,1997).

The disaster can be traced back to imprudence and lack of proper dis-cipline for good banking practices. In addition, the failure of regulatoryauthorities, especially the BOT, was equally responsible for the debacle.The BOT failed to grasp the judicious policy advice that “countries shouldnot rely on foreign financial markets to fill large prolonged shortfalls be-tween domestic investment and savings” (Felix, 1996: 77).

c. Misuse of capital

As the Thai state failed to guide or regulate the use of capital, massiveamounts of capital inflows had been allocated to unproductive and specula-tive activities such as real estate, luxury resorts, golf courses, facilities forentertainment and luxury services. “Thai industrialists” with limited techno-

10

logical capability were building up excess capacities, constructing luxuryoffices and investing in real estate projects while Thai financiers were heavilyinvolved in a flurry of lending. It was a common practice for the banks andbig finance and securities houses to set up subsidiaries in real estate andleasing activities. In several cases, close relatives of owners of parent firmswere in charge of the subsidiaries.

The huge funds from the Thai stock market also contributed to “over-investment in several sectors, namely real estate, petrochemicals, iron andcement. Naturally, this over-investment resulted in excess capacity”(Prachachat Thurakit, 29 August-1 September 1996: 24). According to areliable source, the excess supply in relation to market demand in some keyindustries was estimated to be 300% in private hospitals, 200% in housing inmetropolitan Bangkok, 192% in motor cars, 195% in petrochemicals and150% in iron and steel (Prasartset, 1998: 7). Over-investment, of course,played a major role in rising incremental capital-output ratios, as noted byKrugman (1994).

d. The state’s ineptitude under global financialization

Global financialization requires a new mode of operation for state regu-latory agencies. But this process has not taken place adequately in Thailand.The gigantic size of global market players poses a grave threat to monetaryauthorities. In such a situation, any small central bank would eventuallylose its capacity to stabilize its currency. To put things in perspective, totalreserves of all the central banks combined in 1996 amounted to $640 billion,no more than a third of the daily volume of trading in currency markets(Lietaer, 1997).

We can now identify several policy failures committed by the BOTwithout which the crisis might have been averted or scaled down. One suchmistake “made by the authorities was the decision to open up the capitalaccount” by the introduction of the BIBF in March 1993. The Thai govern-ment did not put a break on such huge capital inflows, about half of whichwere of a short-term nature (Siamwalla, 1997).

11

The second mistake was that the BOT retained a fixed exchange rateregime alongside an open capital account (Siamwalla, 1997). Critics sug-gested that the BOT should widen the exchange rate band, but this was op-posed by interested groups, especially bankers and financiers, who weretrapped heavily in dollar debt. The BOT officials, in the same vein, pointedto a high level of foreign reserves and sound economic fundamentals ascounter-arguments. However, they were blind to the fact that such highforeign reserves were largely accounted for by short-term inflows (borrowedreserves), not by the earning capability of Thai exports or overseas invest-ment (earned reserves). The reserves were basically short-term and debt-induced reserves with a high degree of volatility.

Another misstep made by the BOT was its attempt to defend the THBbecause of its confidence in strong fundamentals. In November 1996, Feb-ruary and May 1997, the THB suffered a few rounds of speculative attacks.“The reserve depletion from these attacks was hidden from the public by theforward sale of its dollars to support the THB” (Siamwalla, 1997). It wasreported that about $23 billion was sold forward in swap transactions todefend the THB. The intervention generated a loss of about $8 billion, com-pared to about $10 billion used by the British government to defend thepound in 1991.

By ignoring the so-called impossible trinity, the Thai monetary author-ity (BOT) was committed to a pegged exchange rate regime and free capitalmobility at the same time. The BOT therefore had depleted its foreign re-serves in defending the fixed exchange rate. Without capital controls in place,the mixture of a fixed exchange rate regime and free mobility of capital onlyhastened the arrival of a currency crisis.

Another error by the BOT was to maintain high interest rates in orderto boost domestic savings and counter inflationary pressures. The wideninginterest rate spread attracted further capital inflows of the worst kind – hotmoney in pursuit of a quick profit. As the Thai economy got trapped in hugeforeign loans, policy makers were unable to cut the interest rates for fear ofigniting a reversal. In the aftermath of the 2 July 1997 devaluation of theTHB, Thailand was under IMF conditionality which prescribed tight mon-etary policy with high interest rates in order to prevent severe outflows and

12

avert possible speculative attacks on the THB despite its severe contractionaryimpacts on the economy.1

e. Money politics – the crisis’s trigger

In addition to the above factors, the dubious involvement of certainpoliticians in certain financial institutions also played a major role in theloss of confidence in the monetary authorities. Indeed, the 1997 financialdebacle in Thailand cannot be said to be free from a process of money poli-tics. The now extinct Bangkok Bank of Commerce (BBC) was only onecase among several others. It was reported that about 40 politicians wereinvolved in some way in its shady deals for personal gain (The Nation, 2March 1998). We agree with the view that the BBC scandal “set in motion achain of events that ultimately caused a loss of confidence in the integrity ofthe BOT’s supervision of financial institutions and brought about the finan-cial crisis.” Therefore, “the BOT had inexcusably mismanaged the failedBangkok Bank of Commerce” (The Nation, 2 March 1998).

In the context of money politics, bankers and financiers have main-tained close relations with political parties through their allied politicians.The former are therefore quite influential in their working relationship withthe regulating agency (BOT). It is not surprising that the BOT had failed toprevent or suppress cases of fraudulent behaviour, cronyism and imprudentlending practices in several financial institutions – as shown in failed banksand 56 failed finance companies.

f. The outbreak of the crisis

As a result of rapid growth and high interest rates, Thailand and manyother East Asian countries received a large inflow of hot money and experi-

1 Interestingly, the recent subprime crisis saw the US turn the IMF formula on its head.The US Federal Reserve has pumped huge amounts of liquidity into the market at theonset to thwart potential market collapse. In hindsight, this only adds to doubts on thereal intention behind the IMF prescription for financially distressed Third World countries.

13

enced a dramatic boom in asset prices. Fixed exchange rates also encour-aged external borrowing and led to excessive exposure to foreign exchangerisk in both the financial and corporate sectors. A fixed exchange rate re-gime, free capital mobility, inept regulatory authorities, a weak financialsector as well as reckless behaviour of both borrowers and lenders all com-bined to render the Thai economy extremely vulnerable to external financialshocks. In 1996, the global economic environment started to change. As theUS economy recovered from a recession in the early 1990s, the FederalReserve began to raise interest rates to head off inflation, making the UnitedStates relatively more attractive as an investment destination. At the sametime, Thai export growth slowed down sharply, deteriorating its current ac-count position. Loss of confidence in Thailand’s economic and financialsystem as well as the sustainability of the THB prompted speculators toattack the currency in 1996 and mid-1997.

The rapid reversal of hot money flows often leaves little time for mon-etary authorities to respond, particularly among emerging economies whosefinancial market institutions are still at the budding stage. This process inbroad sense can also be explained by the Frenkel-Neftci cycle (Taylor, 1999).During the crisis, large debt repayments demanded by creditors exerted sub-stantial pressure on the THB. The BOT, caught off-guard, squandered sub-stantial amounts of international reserves to defend the THB before givingup and adopting a floating exchange rate regime on 2 July 1997.

In the aftermath, the THB lost half of its value and reached its lowestpoint of THB55.9 per dollar on 13 January 1998. The stock market droppedby 75% in 1997 alone. Massive devaluation of the THB, hikes in interestrates and a credit crunch resulted in widespread bankruptcies and mountingnon-performing loans (NPLs) in the financial sector, with attendant conse-quences for the real economy. Output contracted 10% in 1998 with drasticdrops in investment and consumption followed by massive lay-offs andsharply rising levels of unemployment. The episode is credited as the mostsevere post-war economic crisis in Thailand.

14

WITH the outbreak of the 1997 crisis, Thailand came under an IMFprogramme with a stand-by credit of $17 billion from August 1997 to July2003, but was able to complete repayment of outstanding loans six monthsbefore the due date. The IMF prescribed a harsh stabilization policy to achievethe current account surplus needed to repay external creditors. The severeeconomic contraction that followed prompted Thai policy makers to seekroom for an expansionary fiscal stance and lower interest rates to boost theeconomy. The deepened crisis eventually led the IMF to substantially relaxits strict fiscal and monetary policy prescriptions. Consequently, counter-cyclical measures on both fronts were implemented in ways consistent withthe economic realities. However, since the 1997 crisis was so severe, it tookThailand about four years to regain the 1996 level of real GDP. From 1997to 2001, the Thai economy stagnated, with annual average growth of -0.1%.Burdened by twin crises in the balance of payments and the banking sector,Thailand took 18 quarters to overcome the capital account crisis (Mecagniet al., 2007). Between 2002 and 2004 the Thai economy recovered strongly,with average growth rising to 6.2% and moderating subsequently (Figure 4).

From 2006 Thailand suffered from a bout of political setbacks withwaves of political protests against then Prime Minister Thaksin Shinawatra,followed by a military coup at the end of 2006 and elections in 2008. How-ever, the emergence of the new government, which has strong ties to Thaksin,has rekindled another round of protests by the People’s Alliance for Democ-racy (PAD). In spite of these difficulties, the Thai economy has shown strong

Chapter 4

STABILIZATION AFTER THE1997 FINANCIAL CRISIS

15

resilience as demonstrated by sustained macroeconomic and financial sta-bility.

Gross Domestic Product(1988 Prices)

y-o-y Growth Rate (%)

Figure 4. GDP growth

Source: Office of the National Economic and Social Development Board

a. The current account

In general, a marked improvement in the current account from a highdeficit to a strong surplus is considered as a positive sign. Usually, persis-tent high levels of current account deficits signal macroeconomic vulner-ability, especially when coupled with fragile capital account positions. Onthe other hand, a string of high current account surpluses during a period ofeconomic contraction brought about by austere policy measures in responseto a currency and financial crisis does not reflect a viable situation either.This was the case for Thailand during 1998 to 2000 when the economy suf-

10.0

GDP Growth

5.0

0.0

-5.0

-10.0

-15.0

-20.0

16

fered negative growth and the current account was in surplus, at a rate of7.6% to 12.7% of GDP.

The surplus was due mostly to a drastic decline in imports of goodsand services. After the crisis, it took Thailand six years to reach the 1996level of merchandise imports, and seven years for services payments (BOTwebsite, Table ECONOMIC_EI_027). From 2003 onwards the Thai economyhas resumed its normal growth path with a relatively stable current accountposition. The GDP growth ranged from 7.1% in 2003 to 4.8% in 2007. Dur-ing these five years, the current account balance moved from a deficit of4.4% of GDP to a surplus of 6.0%.

b. Foreign reserves and short-term debt

One crucial indicator which was ignored by the Thai monetary author-ity before the 1997 crisis was the ratio of short-term debt to foreign reserves.In fact, for quite a while the distinction between short-term and long-termdebt was not even made in the compilation of financial data at the BOT. Inthe run-up to the 1997 crisis, foreign reserves of the BOT did not cover theamount of short-term debt.

After the crisis, the BOT took several measures in order to change thematurity structure of Thai external debt in favour of long-term debt. Thiswould also help correct maturity mismatches in firms’ balance sheets. Short-term debt accounted for around 50% of total external debt in 1995 but de-clined to 34% in 2007 (Sangsubhan, 2008). Concomitantly, with economicrevival around 2003 and recovery of capital inflows, in both direct and port-folio investment, the ratio of foreign reserves to short-term debt improvedconstantly, from around 0.7 in 1997 to 3.8 in 2003, reaching 4.1 in 2007(Figure 6). In 2003 foreign reserves exceeded the level of 1996 – $42.1billion compared to $38.7 billion. During the last three years foreign re-serves registered rapid growth and averaged around six months of imports.Moreover, if net forward positions from the BOT interventions are takeninto account, the total reserve accumulation of the BOT will now be higherby approximately 15% since late 2007. The increase in reserves relative toshort-term debt suggests that the Thai economy has built considerable resil-

17

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

4.5

Figure 5. Total external debt (mil USD)

Source: BOT_EC_XT_032: External Debt (US$)

Figure 6. Reserves to short-term debt

Source: BOT_EC_XT_032: External Debt (US$)

1992

120,000.00

100,000.00

80,000.00

60,000.00

40,000.00

20,000.00

0.00 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

Long-term Short-term

4

3.5

3

2.5

2

1.5

1

0.5

0

18

ience against external shocks of the kind that devastated the economy in the1990s.

In general, holding reserves at a level above three months of importsincurs unnecessary cost for the public. However, under capital account open-ness with many new risks in global financial markets, there would be groundsfor keeping a substantially higher level of foreign reserves, especially asself-insurance against possible speculative attacks (Akyüz, 2008: 30-31).Also, such a level of reserves may serve as a “broader signal of robust mac-roeconomic fundamentals to the market” (Mendoza, 2004: 75-76). How-ever, for the signal to be real, the accumulated reserves must not be of theborrowed type (Akyüz, 2008: 31-32).

c. External indebtedness

A high level of foreign currency debt compared to the GDP could sig-nal a high degree of external vulnerability. Thailand suffered greatly frommaintaining sizeable foreign debt both before and during the crisis years. Itsexternal debt as a percentage of GDP rose from 69% in 1995 to about 73% in1999. Since then the situation has improved markedly, to about 40% in2003 and under 30% in 2007 (Figure 7).

In a similar vein, the debt service ratio also declined appreciably fromits height in 1998 of 21.4% to 8.5% in 2004 and 11.0% in 2007 (Figure 8).Therefore, export receipts can now be allocated to productive activities ratherthan to repayment of accumulated external debt.

The BOT also took measures to discourage foreign short-term borrow-ing. As seen in Figure 9, the share of short-term loans in the total externaldebt declined consistently from around 47% during the three years beforethe 1997 crisis to its lowest three-year average of 19% during 2000-2002before rising again to 35% in 2007.

Non-bank firms have also reduced their foreign exchange risks by bor-rowing in local currency. A BOT survey, in the third quarter of 2007, of non-bank firms’ foreign liabilities classified by currency shows that while dollarliabilities still remained dominant at 44.9% and the yen at 13.4%, THB li-

19

Figure 7. External debt to GDP (%)

Source: BOT_EC_XT_032: External Debt (US$)

Figure 8. Debt service ratio (%)

Source: BOT_EC_XT_032: External Debt (US$)

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

80.00

70.00

60.00

50.00

40.00

30.00

20.00

10.00

0.00

25.00

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

10.00

5.00

0.00

20.00

15.00

20

Figure 9. Total external debt

Figure 10. Ratio of public debt to GDP

Source: BOT_EC_XT_032: External Debt (US$)

Source: BOT, Economic and Monetary Conditions in 2007

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

Public Debt

Long-term Short-term19

85

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

(D

ec)

Government limit (50% of GDP)

Government Debt

FIDF Debt

Public Debt: GDP

State Enterprises Debt

Other state organizations

Fiscal Year

24.5%

11.2%

2.0%0.3%

38.0%

Dec 2008

% GDP

70

60

50

40

30

20

10

0

21

abilities accounted for about 36.2% of the total. The upward trend in thelatter has been evident since 2000 (BOT Statistics 2007, Q4: 24). Develop-ment of the baht bond market is expected to reduce the relative importanceof foreign currencies in residents’ liabilities.

d. Fiscal sustainability

Public debt played little role in the 1997 crisis as the Thai governmenthad followed a conservative fiscal policy in the years leading to the crisis(Figure 10). Public debt as a share of GDP was above 40% during the eco-nomic boom in the late 1980s, but declined to around 15% in 1996. From1998 to 2002, this figure averaged more than 50%. Apart from direct gov-ernment debt (22.1% of GDP in 2000), major items include guaranteed andnon-guaranteed debt of non-financial state enterprises (18.1%) and debt aris-ing from bailout operations (15.5%) implemented by the Financial Institu-tions Development Fund (FIDF) (BOT Report 2000: 2.13.3).

It is important to note that during 1998-2002, public debt rose dramati-cally because the state had to absorb the losses incurred by private financialfirms. Thus, vulnerability in the financial and corporate sectors had beentransferred to the state and could jeopardize fiscal sustainability. It was outof concern over this that the Ministry of Finance and the BOT have intro-duced measures to achieve fiscal sustainability in the aftermath of the 1997crisis. Since November 2001, the Thai government has established a frame-work for sustainable fiscal policy by setting a 50% maximum limit for thepublic-debt-to-GDP ratio. Debt payments are not to exceed 15% of totalexpenditure while capital spending is to reach at least 25% of the budget(Worasangasilp, 2008: 33).

Such a policy framework seems to have worked well in reducing pub-lic debt relative to GDP; in fact, this ratio went down to around 38% in 2008.However, in assessing state expenditures in recent years, one should alsofocus on quasi-fiscal activities which are not part of national budgeting. Suchactivities include public-private partnership (PPP), securitization via specialpurpose vehicle (SPV), and the so-called populist projects (Pinto et al., 2007).These off-budget items accounted for about 10% of national expenditure

22

during 2006 and 2007. These could turn into a big budgetary burden if notproperly managed. However, monitoring such spending relies on adminis-trative decisions which might be prone to cronyism. Fiscal sustainabilityneeds greater fiscal discipline and measures to overcome structural mis-matches between revenues and spending.2

e. Financial restructuring

Not only the balance of payments but also the financial sector has be-come more resilient to shocks. The BOT has improved its supervision andprudential regulations and initiated a Financial Stability Master Plan tostrengthen financial institutions and enhance financial market efficiency. Thebanking sector has been restructured to reduce NPLs, enrich asset quality,and improve capitalization as well as management of risks. The Thai finan-cial sector is now moving towards the full implementation of the Basel IIframework.

2 Tax revenues tend to lag behind spending due to free trade agreements with severalcountries and regions; dependence on taxes which are highly pro-cyclical, such as theVAT and corporate tax; and highly liberal tax deduction allowances – see Pinto et al.(2007: 38-39).

23

a. Hot money returns

AS recovery in Thailand was well under way, capital flows started to moveback in the early years of this decade, helped also by favourable global fi-nancial conditions, and accelerated during 2006 as problems linked tosubprime lending became visible in the United States (Figure 11). This cre-ated serious challenges for the BOT in realizing its twin objectives of ex-change rate stability and inflation targeting. When the exchange rate hasshown excessive and persistent volatility which appears to be inconsistentwith fundamental changes, the BOT has intervened in the market to stabilizethe currency, mainly through outright spot transactions by buying/sellingthe THB against the dollar. Sometimes this has been supplemented by for-eign exchange swap transactions to influence the liquidity condition in theoffshore market in order to make it more costly to fund their speculativeposition. The timing and style of intervention differ and sometimes the BOTintervenes discreetly in order to “retain an element of surprise, employingboth verbal and/or actual interventions” (BOT, 2004).

By late 2004, Thailand’s international reserves accumulated throughsuch interventions amounted to $46 billion (BOT, 2004). Lately, with thepersistent surge in capital inflows, especially since 2006, intervention hasled to unprecedented reserve accumulation, which is consistent with the re-gional trend (Akyüz, 2008: 30-31). As the reserve accumulation continued,reserves finally hit the $100 billion mark around October 2007 (Figure 12).

Chapter 5

RECOVERY AND SURGE IN CAPITAL INFLOWSAND THE POLICY RESPONSE

24

Figure 11. Composition of capital flows after the 1997 crisis(million US$)

Source: Bank of Thailand

Figure 12. The BOT reserves (million US$)

Source: Bank of Thailand

Gross Reserves Net Forward Position Net Reserves

Capital account and components 25,000.00

20,000.00

15,000.00

10,000.00

5,000.00

0.00

(5,000.00)

(10,000.00)

(15,000.00)

(20,000.00)

140,000.00

120,000.00

100,000.00

80,000.00

60,000.00

40,000.00

20,000.00

0.00

-20,000.00

-40,000.00

FDI Other investment Equity securitiesDebt securities Port Outbound Cap Acct

Gross Reserves Net Forward Position Net Reserves

25

Despite heavy intervention, however, the THB appreciated by almost 17%in 2006 compared to its level in 2005 (Table 1).

Table 1. Percentage change of exchange rate betweenDecember 2005-2006

Currency THB TWD CNY MYR PHP SGD IDR KRW YEN EUR GBP

(2006)Percentage 16.6 0.8 3.1 6.4 7.8 7.8 8.3 9.5 -0.1 10.5 13.6

change from

Dec 2005Source: Bank of Thailand

b. Sterilization

Most countries which intervene in the foreign exchange market to sta-bilize the currency in the face of strong capital inflows find it necessary tosterilize the impact of the intervention on domestic liquidity in order to avoidinflationary pressures in product and/or asset markets. Although steriliza-tion can contain the expansion of the monetary base, it also leads to thereduction of central banks’ net domestic assets. It is also costly since interestpaid on debt issued for sterilization often exceeds the return earned on for-eign exchange assets.

The BOT has indeed used a combination of monetary instruments forsterilizing the impact of intervention on domestic liquidity, including repos,bond issuance and foreign exchange swaps, on the grounds that “interven-tion and sterilization are necessary policy tools under the inflation targetingframework with managed-float exchange rate system and therefore steril-ization is part of the appropriate management of liquidity in the money mar-ket.”

To what extent has the BOT been successful in sterilization? This canbe estimated empirically by regressing the change in the central bank’s netdomestic assets (NDA) on the change in the central bank’s net foreign assets(NFA) (shown in Figure 13) to obtain the sterilization index. The index is

26

based on the coefficient βi,t of the following OLS (Ordinary Least Squares)

model:

ΔNDAi,t,m

= αi,t + β

i,t*ΔNFA

i,t,m + u

i,t,m

where

ΔNDAi,t,m

= monthly change in country i’s central bank’s net domesticassets during month m of year t (consisting of “claims on other depositorycorp.”, “net claims on central government”, “contra-entry: coins in circula-tion” and “claims on other sectors” from Table 5 on the BOT website)

ΔNFAi,t,m

= monthly change in country i’s central bank’s net foreignassets during month m of year t.

The sterilization index (βi,t) or the slope of the regression line reflects

how much domestic credit changes in response to a change in internationalreserves. This number estimates the central bank’s effort to offset the expan-sion of the monetary base (domestic liquidity) associated with the accumu-lation of foreign reserves. For simplicity, the slope coefficient is multipliedby -1 so that an estimated value of the sterilization index equal to 1 impliesfull sterilization, whereas a value of 0 represents no sterilization – that is,expansion of the monetary base by the amount of intervention in the cur-rency market. In Table 2 the coefficient (β

i,t) takes a value of 0.33, indicating

that the BOT was only partially successful in sterilization during 2005-2007.Indeed, the BOT resorted to open market operations several times dur-

ing this period. From 2005-2007, it issued bonds worth about THB1.5 tril-

Variable Coefficient Std. Error t-Statistic Prob.

C 6605.602 6281.452 1.051604 0.3004

ΔNFA -0.331272*** 0.109956 -3.012758 0.0049

Table 2Dependent Variable: ΔNDA

27

mil

baht

Figure 13. The movement of NDA and NFA from January 2005 toDecember 2007

lion (approximately $40 billion) to absorb liquidity from the financial sys-tem as a result of its currency intervention (Suwanaporn, 2007). However,its ability to do so was somewhat constrained by limited availability of in-struments as well as the absorptive capacity of the Thai financial markets.

We also checked the sterilization index separately for each year. Theresults reported in Table 3 suggest that the sterilization effort was the mostintense in 2007 when net domestic assets declined by almost 50 cents forevery dollar increase in net foreign assets. In 2007, the BOT reported lossesfrom its foreign exchange intervention of about THB60 billion (about $1.7billion) based on the Revaluation of Average Cost of Reserves Method. Of

2005 2006 2007 June 07 – June 080.1 0.362 0.497 0.118

(0.41509) (-2.30242) (-2.44365) (-0.53187)

Table 3. Sterilization index

Numbers in the parentheses are t-statistics

NFA

NDA

28

this amount, THB52.5 billion (about $1.5 billion) could be attributed to thecost incurred on bonds issued by the BOT (Suwanaporn, 2007).

c. Capital account measures

The prospect of a massive surge in capital inflows and difficulties instabilizing the currency has caused serious concern among Thai policy mak-ers, as pointed out by a senior official from the Thai Ministry of Finance:

Thailand’s last experience with extensive liberalization of the capital

account in the early 1990s shows how easily a small open economy

without adequate safeguards can fall victim to the massive speculative

inflows. Sound macroeconomic fundamentals and a well-regulated fi-

nancial system are no longer sufficient conditions for financial stability

if all funds can freely flow into and out of the country. (Sangsubhan,

2008: 16)

This concern also echoes the cautious view on capital account liberal-ization expressed by a former chairman of the United States Federal ReserveSystem, Paul Volcker. On the heels of the Asian financial crisis, Mr. Volckerhad this to say:

What we are inclined to forget is how small [these countries] are in

financial terms. You can add up the whole banking system of most of

those emerging economies and they will not equal one mid-sized re-

gional bank in the United States. These US banks have concluded they

are too small to thrive or to survive in turbulent markets. For the emerging

economies, which had opened their financial markets to the world, their

entire banking system is to be subject to the volatility of international

financial markets. (Volcker, 1999: 424)

Experience in emerging markets has shown that exposure to unrestrainedcapital flows can easily cause serious vulnerabilities which might eventu-ally lead to sudden stops and reversals, hence precipitating currency and

29

banking crises. It is these concerns which underlie various measures of con-trol over capital inflows, notably short-term capital inflows, imposed by theBOT even though these measures have occasionally been half-hearted orinadequately planned.

According to the BOT, measures to prevent THB speculation includefour types of measures summarized in Appendix 1: those designed to limitTHB liquidity available to non-residents (NRs), to limit capital inflows, andmeasures on NRBA (non-resident baht account) and NRBS (non-residentbaht account for securities) accounts and non-deliverable forward (NDF).Since 1998 the BOT has taken various steps in these areas in order to limitthe scope for THB speculation; these are listed in Appendix 2. With thesurge in capital inflows after 2004, the attention of the BOT has increasinglyfocussed on liberalization of resident outflows and restrictions over short-term inflows in order to limit the upward pressure on the currency.

With the liberalization of capital outflows, whose aim is said to be “topromote more balanced capital movements, and to increase flexibility forThai businesses in managing their foreign currencies in order to enhancetheir competitiveness in the world market”, the BOT enlarged the scope andraised the limits over amounts that could be transferred abroad for the pur-pose of direct and portfolio investment, acquisition of real estate and varioustypes of remittances. In April 2005 the maximum amount that could beinvested by certain types of institutional investors in debt securities issuedby sovereign or quasi-sovereign entities was raised from $500 million to$1.5 billion.

The BOT attempted to check massive capital inflows twice by tighten-ing foreign exchange controls in November and December 2006. The mea-sures in November included prohibiting financial institutions from issuingand selling bills of exchange in THB to NRs. On 6 December 2006, theBOT imposed a stronger measure to curb foreign exchange influx and toprevent rapid appreciation of the THB by prohibiting the selling and buyingof all types of debt securities through sell-and-buy-back transactions for allmaturities. Still, the THB kept on appreciating. On 18 December 2006, theBOT decided to impose a 30% unremunerated reserve requirement (URR)on all capital inflows. This required all financial institutions to withhold

30

30% of foreign currencies bought or exchanged against the THB for anytransactions, except those related to trade in goods and services, repatriationof investments abroad by residents, and foreign direct investment. Investorsstood to lose one-third of the reserves if they repatriated their funds within ayear.

Figure 14. THB/USD exchange rate, onshore and offshore

Source: Bank of Thailand

Foreign capital flowsinto SET and M&A

Coup d’etaton Sep 19, 06

Abandon URRon Mar 3, 08

Rule URR onDec 18, 06

Offshore

Onshore

Jan

05F

eb 0

5M

ar 0

5A

pr 0

5M

ay 0

5Ju

n 05

Jul 0

5A

ug 0

5S

ep 0

5O

ct 0

5N

ov 0

5D

ec 0

5Ja

n 06

Feb

06

Mar

06

Apr

06

May

06

Jun

06Ju

l 06

Aug

06

Sep

06

Oct

06

Nov

06

Dec

06

Jan

07F

eb 0

7M

ar 0

7A

pr 0

7M

ay 0

7Ju

n 07

Jul 0

7A

ug 0

7S

ep 0

7O

ct 0

7N

ov 0

7D

ec 0

7Ja

n 08

Feb

08

43

The imposition of the URR triggered an immediate sharp (20%) de-cline in stock prices and created a relatively wide margin between onshoreand offshore exchange rates of the THB against the dollar (Figure 14). On19 December, the Minister of Finance had to exempt the equity market fromthe URR. However, it applied to debt securities until March 2008 when thenew government finally abolished it.

42

41

40

39

38

37

36

35

34

33

32

31

30

29

31

a. Capital control index

CAPITAL control measures at times of strong surges in inflows may be usedfor several reasons, including prevention of currency and maturity mismatchesin private balance sheets, restricting liquidity expansion and asset bubbles,reducing cost of sterilization and avoiding unsustainable currency apprecia-tions and payments positions (Akyüz, 2008). In Thailand, the main reasonexplicitly given by the BOT for introducing various capital control mea-sures is “the wide range and persistent departures of exchange rate from itsfundamental values, rather than short-term fluctuation” (BOT, 2004). Hotmoney was believed to be the main reason for the strength of the THB,particularly during 2005-2007, and the main motive behind the introductionof the URR.

The experience in Latin America, notably Chile, shows that as a mar-ket-based measure, the URR tends to lengthen the maturity of foreign debt,but is not very effective in achieving other objectives (Edwards, 1999). Ad-vances in telecommunication and increasingly sophisticated financial engi-neering devices often help investors to circumvent control measures, ren-dering them ineffective. Controls on outflows are also known to be ineffec-tive when underlying macroeconomic policies and conditions are unsustain-able. At times of strong inflows, removal of restrictions over resident in-vestment abroad does not necessarily lead to outflows and relieve the pres-sure on the currency.

Chapter 6

A QUANTITATIVE ASSESSMENT OF THEEFFECTIVENESS OF CAPITAL CONTROLS

32

In assessing the overall effectiveness of capital control measures intro-duced by the BOT in attaining their objectives, we start by building a quan-titative index, a capital control index (CCI), for measuring the intensity ofoverall controls and its variation over time. While there are several concep-tual and empirical problems in designing such an aggregate measure, it cannevertheless provide some indications as to the adequacy and appropriate-ness of the instruments of control adopted and the degree of their implemen-tation. Such indices are often constructed on the basis of information con-tained in the Annual Report on Exchange Arrangements and Exchange Re-strictions published by the IMF. Several empirical papers have experimentedwith quantifying capital controls (Alesina et al., 1994 and Rodrik, 1998) byrelying on a dummy variable representing such controls, assigning it a valueof “1” where controls are in place and “0” otherwise. However, for ourpurpose, an index based on yearly data would not be very helpful since theissue here is the impact of control measures on the exchange rate of the THBthrough their impact on short-term capital flows or the so-called hot money.

We have thus developed a CCI from information contained in notifica-tions, press releases and speeches related to foreign exchange measures andpublished formally on the BOT website. All measures of control are as-sumed to be in force on the effective date, regardless of the announcementdate. Since the data are not confined to annual reports and many new mea-sures have been frequently introduced, sometimes in a matter of days, theCCI developed and used here contains significantly more detailed informa-tion than traditional measures based on the IMF data.

The aggregate CCI score is reached through a systematic applicationof “+1” and “-1” to each announced measure of control. “+1” is assigned toany measure that restricts inflows or facilitates outflows while “-1” is as-signed to any measure that relaxes restrictions over inflows or tightens thoseon outflows. In the former case, pressure for currency appreciation would bereduced and reserve accumulation would slow down, while the oppositewould be the case when restrictions over inflows are relaxed and outflowstightened.

Next, the numbers are scaled by different weights based on direct andindirect impact criteria. The weight’s range is from 0.5 to 2.0, depending on

33

the severity of the measure. Rules and regulations directly relating to capitalflows are assigned greater weight. For example, measures designed to di-rectly block or relax capital inflows greater than $1 million are weighted by“2”. For each announced measure that indirectly controls capital flows, aweight of “1” is assigned; these include measures that allow or prohibit do-mestic financial institutions from dealing with foreign exchange transac-tions, measures related to hedging or derivative transactions, and those asso-ciated with transactions of NRs in amounts less than $1 million. Lastly,minor indirect restrictions or regulations on capital flows are weighted by“0.5”. Needless to say, this procedure leaves considerable room for judge-ment over the nature of the measures introduced.

Once every measure has been assigned its number and weight, the num-bers are sequentially accumulated over time to arrive at the CCI score. Theresult is shown in Figure 15.

Figure 15. Capital control index

Source: developed by authors

CCI35

97 98 99 00 01 02 03 04 05 06 07

30

25

20

15

10

5

0

-5

34

Figure 16. Capital control index for inflows

CCIINF

Source: developed by authors

Figure 17. Capital control index for outflows

Source: developed by authors

CCIOUTF

In addition, separate indices of control are constructed for inflows andoutflows, according to the objectives pursued by the announced measures(CCIINF and CCIOUTF respectively; Figures 16 and 17). The CCIINF, inparticular, is used to test the response of short-term capital inflows. The

110

108

106

104

102

100

98

96

97 98 99 00 01 02 03 04 05 06 07

104

100

96

92

88

84

80

76

72

68 97 98 99 00 01 02 03 04 05 06 07

35

spike in the CCIINF at the end of 2006 reflects the 30% URR to stem thesurge of hot money inflows. Similarly, the steep decline of the CCIOUTFillustrates the intense push by the Thai authorities to relax resident outflowsin order to relieve the pressure on the THB.

b. Short-term capital flows defined

For the purpose at hand, we focus our analysis on Net Flow of PrivateFinancial Account (BOT website, Table EC_XT_014) from the BOT whichcontains different categories of private capital flows, including bank andnon-bank flows. The latter consist of direct investment, loans, equity secu-rities, debt securities and so on. Bank flows include other foreign transac-tions such as foreign currency deposits. Sample data are shown in Table 4.Notes 2/ and 4/ in the table mean that before 2005 NRBA was not classifiedinto other items in the account but after January 2005 it was classified ac-cording to its present form. To maintain the consistency and integrity of thedata set, we therefore select a data period from January 2005 to December2007 for our analysis.

Given the components of the Net Flow of Private Financial Accountand the stated emphasis of the BOT on hot money flows as the main cause ofthe appreciation of the THB during 2005-2007, attention is here focussed oncapital inflows to equity securities (INFL_ES) and debt securities(INFL_DS).3 As seen in Table 5, these two are the most volatile flows com-pared to FDI (INFL_FDI), foreign loans (INFL_FORLOAN) and trade credit(INFL_TC). In Figure 18, FDI, foreign loans and trade credit exhibit rela-tively low volatility.4 Equity securities inflows represented a very smallvolume after the 1997 crisis but started to increase at the same time thedollar began weakening, reflecting the reaction of markets in turning awayfrom dollar assets.

3 Except capital invested in government and monetary authority securities.4 Direct investment is defined as investment by an incorporated or unincorporated enterprise

in which a direct investor who is resident in another economy owns 10% or more of theordinary shares or voting power (for an incorporated enterprise) or the equivalent (foran unincorporated enterprise).

36

Net Flow of Private Financial Account1 (THB million)

DEC 2007 NOV 2007 OCT 20071 Bank2 100,293.74 61,075.84 (11,915.37)2 Commercial banks 100,293.74 61,075.84 (11,915.37)3 Change in assets 92,803.98 37,718.83 (11,741.94)4 Change in liabilities 7,489.75 23,357.01 (173.43)5 of which: Recapitalization 20,985.57 0.00 0.00 6 BIBF 0.00 0.00 0.00 7 Change in assets 0.00 0.00 0.00 8 Change in liabilities 0.00 0.00 0.00 9 Non-bank (49,849.47) (96,616.20) (36,741.74)10 Direct investment3 (3,960.39) 19,233.79 15,391.19 11 Foreign direct investment 4,107.32 21,276.76 18,749.17 12 Inflow 59,705.99 42,864.56 60,473.92 13 Outflow (55,598.66) (21,587.79) (41,724.75)14 Thai direct investment abroad (8,067.71) (2,042.96) (3,357.97)15 Inflow 4,105.90 1,850.56 1,549.98 16 Outflow (12,173.62) (3,893.53) (4,907.96)17 Loans 28,289.41 (3,456.75) 3,039.35 18 Foreign loans 24,440.50 (2,679.35) 3,348.00 19 Inflow 43,875.50 17,499.55 54,859.74 20 Outflow (19,434.99) (20,178.91) (51,511.73)21 Thai loans 3,848.90 (777.40) (308.65)22 Inflow 3,910.17 140.14 654.15 23 Outflow (61.26) (917.54) (962.81)24 Portfolio investment (72,620.68) (97,056.89) (39,828.46)25 Equity securities (3,915.09) (11,648.53) 5,681.61 26 Inflow 95,361.47 121,150.88 157,071.89 27 Outflow (99,276.56) (132,799.41) (151,390.28)28 Debt securities (68,705.59) (85,408.35) (45,510.08)29 Inflow 10,248.43 3,292.12 6,759.51 30 Outflow (78,954.02) (88,700.48) (52,269.60)31 Trade credits (1,201.10) (14,555.98) (15,890.60)32 Inflow 125,627.82 118,152.64 120,705.59 33 Outflow (126,828.93) (132,708.62) (136,596.20)34 Others4 (356.71) (780.36) 546.78 35 Inflow 969.44 518.06 1,515.62 36 Outflow (1,326.15) (1,298.42) (968.84)37 Total 50,444.26 (35,540.36) (48,657.12)

Notes:1/ From April 2004 onwards, inputs for private financial flow data have been obtained throughdata sets electronically submitted by financial institutions.2/ Since January 2005, there has been a change in data source and transactions in non-resident baht account (NRBA). The NRBA has been reclassified to be included under thebanking sector, in accordance with balance-of-payments compilation guidelines set by the IMF.3/ Reinvested earnings have been recorded as part of direct investment in the financial ac-count, and its contra entry recorded as investment income in the current account. The serieshas been revised back to 2001.4/ Prior to the year 2005, transactions in non-resident baht accounts were included under“other capital”.

Table 4

Source: Table EC_XT_014, Bank of Thailand

37

c. Empirical estimates

To investigate the effectiveness of capital controls, we employ the vec-tor autoregression (VAR) model and impulse response functions to analyzethe interaction between hot money flows and the CCI. We also include othercontrol variables that have been shown in the literature to have influenceover short-term money flows. The data set used is monthly (in THB) andcovers the period from January 2005 to December 2007.5 The VAR equa-tion contains the following endogenous variables:

• LINFLPORT: Logarithm of portfolio inflows as a percentage of GDP• CIPD: Covered interest parity differential (between Thai and US 3-

month Treasury Bill rates)• LREER: Logarithm of the ratio of real effective exchange rate and a

series trend constructed by applying the Hodrick-Prescott filter to esti-mate the long-term trend component of a series

• ROS: Different returns between the Thai (SET) and US (S&P500) stockmarkets

INFL_DS INFL_ES INFL_FDI INFL_FORLOAN INFL_TC

Mean 1.0692 14.4388 9.2723 5.0284 16.2401

Std. dev. 1.1939 17.0054 3.4781 2.3423 2.9631

Coefficient of variation 1.1166 1.1778 0.3751 0.4658 0.1825

Table 5. Mean, standard deviation and coefficient of variation of eachtype of inflow (2005-2007)

Note: All numbers are shown as a percentage of GDP

5 Since capital flow and GDP data are on a quarterly basis, we have to convert them intomonthly data. This is accomplished using the quadratic-match average method in theEViews computer program. The method fits a local quadratic polynomial for eachobservation of the low-frequency series. The polynomial is then used to fill in allobservations of the high-frequency series associated with the period (see EViews formore detail).

38

INFL_FDIINFL_FORLOANINFL_DS

• LCCIINF: Logarithm of the CCI for inflows (constructed by the au-thors)

• LUS1M: Logarithm of the THB/USD one-month forward rate.

The VAR estimation using one lagged length is shown in Table 6.

Figure 18. Inflows as percentage of GDP, 2000-2007

Source: Bank of Thailand

70

60

50

40

2000 2001 2002 2003 2004 2005 2006 2007

INFL_TCINFL_ES

30

20

10

0

39

Vector Autoregression Estimates Date: 08/08/08 Time: 15:52 Sample: 2005M01 2007M12 Included observations: 36 Standard errors in ( ) and t-statistics in [ ]

LINFLPORT D(CIPD) D(LREER) D(LCCIINF) ROS D(LUS1M)

LINFLPORT(-1) 0.234332 0.001607 -0.001687 0.007475 -0.042756 0.000745 (0.12208) (0.00547) (0.00430) (0.00734) (0.01741) (0.00228)[ 1.91947] [ 0.29373] [-0.39192] [ 1.01851] [-2.45637] [ 0.32689]

D(CIPD(-1)) 10.67272 -0.408602 0.122812 -0.562094 2.002761 0.160280 (6.12158) (0.27437) (0.21586) (0.36802) (0.87280) (0.11435)[ 1.74346] [-1.48925] [ 0.56893] [-1.52736] [ 2.29463] [ 1.40161]

D(LREER(-1)) 20.23033 -0.729806 0.329997 -0.242687 1.490346 0.199615 (7.26709) (0.32571) (0.25626) (0.43688) (1.03613) (0.13575)[ 2.78383] [-2.24067] [ 1.28775] [-0.55550] [ 1.43838] [ 1.47043]

D(LCCIINF(-1)) 0.667479 -0.005552 -0.049265 0.035066 0.049111 0.080594 (3.36970) (0.15103) (0.11882) (0.20258) (0.48044) (0.06295)[ 0.19808] [-0.03676] [-0.41460] [ 0.17310] [ 0.10222] [ 1.28034]

ROS(-1) 0.980535 -0.073254 -0.005385 -0.047794 0.177583 0.021030 (1.36344) (0.06111) (0.04808) (0.08197) (0.19440) (0.02547)[ 0.71916] [-1.19874] [-0.11201] [-0.58309] [ 0.91351] [ 0.82567]

D(LUS1M(-1)) 31.83617 -0.292786 0.312374 -0.375498 1.481201 0.544042 (10.2798) (0.46074) (0.36249) (0.61800) (1.46567) (0.19203)[ 3.09697] [-0.63547] [ 0.86174] [-0.60761] [ 1.01060] [ 2.83308]

C 2.699408 -0.011637 0.006060 -0.027198 0.158759 -0.001490 (0.42461) (0.01903) (0.01497) (0.02553) (0.06054) (0.00793)[ 6.35737] [-0.61146] [ 0.40471] [-1.06547] [ 2.62238] [-0.18788]

R-squared 0.516489 0.188801 0.095435 0.143142 0.304408 0.314658 Adj. R-squared 0.416452 0.020966 -0.091716 -0.034139 0.160492 0.172863 Sum sq. resides 2.883174 0.005792 0.003585 0.010420 0.058610 0.001006 S.E. equation 0.315309 0.014132 0.011119 0.018956 0.044956 0.005890 F-statistic 5.162996 1.124924 0.509935 0.807432 2.115184 2.219107 Log likelihood -5.638495 106.1453 114.7789 95.57371 64.48471 137.6513

Table 6. VAR1_Portfolio Investment Capital Inflows

40

Akaike AIC 0.702139 -5.508073 -5.987719 -4.920762 -3.193595 -7.258405 Schwarz SC 1.010045 -5.200166 -5.679812 -4.612855 -2.885688 -6.950498 Mean dependent 3.502563 -0.004596 -3.07E-05 0.001100 0.002551 0.001092 S.D. dependent 0.412761 0.014283 0.010641 0.018640 0.049066 0.006476

Determinant resid covariance (dof adj.) 1.15E-20 Determinant resid covariance 3.16E-21 Log likelihood 543.2000 Akaike information criterion -27.84445 Schwarz criterion -25.99701

The impulse response function provides the information on the effectof a one-time shock on current and future values of the endogenous vari-ables in the system. Figure 19 gives the impulse response of portfolio invest-ment inflows (LINFLPORT) to measures of restriction introduced by theBOT as measured by LCCIINF. It shows that capital controls, through di-rect restrictions on portfolio inflows, do cause portfolio inflows to decline,with the greatest effect in the second month of the implementation. How-ever, similar to the results from other studies, it is seen here that the restric-tions’ effectiveness rapidly diminishes within six months of the implemen-tation, reflecting market adjustment to circumvent the controls.

Figure 19

Response of LINFLPORT to CholeskyOne S.D. D(LCCIINF) Innovation

1 2 3 4 5 6 7 8 9 10

.10

.05

.00

-.05

-.10

-.15

-.20

41

Figure 20 reveals that capital control effort intensified as large amountsof portfolio investment started to enter Thailand during 2005-2007. Again,this finding is also consistent with experiences of other countries employingcapital controls to discourage short-term foreign capital influx (Cardoso andGoldfajn, 1997).

Figure 20

Response of D(LCCIINF) to CholeskyOne S.D. LINFLPORT Innovation

Figure 21 shows the effect of foreign portfolio inflows on the real ef-fective exchange rate (REER). The response function confirms that with theacceleration of inflows, especially equity portfolio investment (see Figure22), the THB relative to Thailand’s trading partners strengthened but thislasted for only about two months when intervention by the BOT eased theupward pressure on the currency. The BOT has closely monitored real andnominal effective exchange rates as indicators of medium-term and long-term competitiveness of the Thai economy compared to the rest of the worldand intervened accordingly. Figure 22 shows that jumps in real and nominaleffective exchange rates during 2005-2007 coincided with the influx of for-eign portfolio inflows during the same period.

1 2 3 4 5 6 7 8 9 10

.014

.012

.010

.008

.006

.004

.002

.000

-.002

-.004

42

Figure 21

Response of D(LREER) to CholeskyOne S.D. LINFLPORT Innovation

Figure 22. Nominal and real effective exchange rates(trade-weight, 1994 = 100)

1 2 3 4 5 6 7 8 9 10

.008

.006

.004

.002

.000

-.002

-.004

115.00

105.00

95.00

85.00

75.00

65.00

55.00

Jan-

90

Jan-

91

Jan-

92

Jan-

93

Jan-

94

Jan-

95

Jan-

96

Jan-

97

Jan-

98

Jan-

99

Jan-

00

Jan-

01

Jan-

02

Jan-

03

Jan-

04

Jan-

05

Jan-

06

Jan-

07

Jan-

08

NEER

REER

43

Figures 23 and 24 confirm the impact of control measures on incomingportfolio investment, which was by far the most important component ofcapital inflows during the period under study. The impact is particularlypronounced for equity portfolio investment. The introduction of the 30%URR was certainly effective, as judged also by the spread between onshoreand offshore rates and the sharp fall in equity portfolio inflows.

Again there appears to be a strong relation between outbound portfolioinvestment and outflow CCI (Figure 25). There was a hasty relaxation ofrestrictions over resident outflows during 2005-07 for mutual funds. Eventhough the types and classes of securities that Thai mutual funds are allowedto acquire abroad have continued to be restricted, there has been a rapidincrease in portfolio investment abroad.

Figure 23. Private foreign portfolio inflows and capital control indexon inflows (as represented by CCIINF – monthly)

2000 2001 2002 2003 2004 2005 2006 2007

80

60

40

20

0

112

108

104

100

96

CCIINF

INFL_TC

INFL_DS

INFL_ES

44

Source: CEIC

Figure 24. Total foreign portfolio inflows and capital control index oninflows (annual)

Figure 25. Outbound portfolio investment and capital controlindex on outflows

45

Figure 26 compares the behaviour of the overall control measure indexwith that of the THB exchange rate. It reveals that the BOT intensified itsefforts to control capital flows as the THB gained against the dollar, remov-ing restrictions over outflows and tightening or introducing new restrictionsover inflows. Throughout the BOT also continued to intervene heavily inboth spot and forward markets. One week before abolishing the URR, theBOT’s net forward position amounted to $24.1 billion or about 25% of theinternational reserves.

We find no impact of the overall CCI, including control measures per-taining to both inflows and outflows, on the behaviour of the exchange rateeven though control measures for inflows and outflows were effective in theclasses of capital flows to which they applied. This may be a reflection ofthe limitations of such an aggregate measurement. It also suggests that theremay be strong counteracting influences between inflows and outflows. Justas increased freedom for resident outflows may bring back flight capital,

Figure 26. Relation between THB and CCI (for both inflows andoutflows)

Source: CEIC

Capital control measures have been intensified as the on and offshore THB got stronger

60.00

Jan-

97

Jul-9

7

Jan-

98

Jul-9

8

Jan-

99

Jul-9

9

Jan-

00

Jul-0

0

Jan-

01

Jul-0

1

Jan-

02

Jul-0

2

Jan-

03

Jul-0

3

Jan-

04

Jul-0

4

Jan-

05

Jul-0

5

Jan-

06

Jul-0

6

Jan-

07

Jul-0

7

Jan-

08

50.00

40.00

30.00

20.00

10.00

0.00

35

30

20

15

10

5

0

-5

-10

Onshore

Offshore

CCI

Bah

t

25

46

tightened restrictions over inflows may make non-resident investors morereluctant to repatriate their profits and investment.6

Indeed, Figures 27, 28 and 29 reveal that the amount of inflows waspositively related to outflows for equity (INFL_ES vs. OUTFL_ES) anddebt (INFL_DS vs. OUTFL_DS) as well as trade credit (INFL_TC vs.OUTFL_TC). As inflows slowed down (accelerated), so did the outflows.This suggests that in emerging markets closely integrated into global finan-cial markets, and particularly with a high degree of presence of non-resi-dents in domestic financial markets, it would be difficult to manage overallcapital flows and the exchange rate on the basis of counter-cyclical and par-tial control measures.

Figure 27

6 IMF (2007, Chap. 3: 10) has argued that: “Overall, the studies suggest that controls oninflows did not affect the volume of net flows in most countries, although it seems thatthe controls were able to temporarily tilt the composition toward longer maturities in afew cases. Even in cases in which a narrow range of objectives were met, controls hadonly temporary effects as market participants eventually found ways to circumvent them.”

2000 2001 2002 2003 2004 2005 2006 2007

80

60

40

20

0

-20

-40

-60

INFL_ES

OUTFL_ES

47

Figure 28

Figure 29

INFL_DS

OUTFL_DS

2000 2001 2002 2003 2004 2005 2006 2007

8

4

0

-4

-8

-12

2000 2001 2002 2003 2004 2005 2006 2007

30

20

10

0

-10

-20

-30

INFL_TC

OUTFL_TC

48

THE challenges that the Thai economy has been facing in recent years arequite different from those in the 1990s. Large capital inflows in the run-upto the 1997 crisis were allowed to result in significant appreciation of thecurrency, and hence were associated with a growing current account deficit.By contrast, there has been considerable awareness in recent years aboutvulnerability to external financial shocks and a determination not to allow arepeat of the crisis. Consequently, the policy approach to the recent surge incapital flows was dominated by a desire to maintain a competitive exchangerate to counter the upward pressures on the currency from the surge in port-folio inflows and relatively sizeable current account surpluses, greatly helpedby the boom in world commodity markets. Various means have been used,including sterilized intervention and accumulation of large stocks of reserves,measures to discourage short-term inflows as well as liberalization of resi-dent outflows for a class of investors. The appreciation of the THB over andabove the extent desired by the BOT suggests that these measures have hadmixed success, but in the absence of such interventions and controls, theexchange rate outcome could have been much more unfavourable. Con-versely, a more judicious and better-planned mix of controls over inflowswould have, in all likelihood, produced a better outcome in terms of cur-rency stability.

The problems that Thailand has faced in recent years in managing itscurrency are not very much different from those in other emerging marketssince there has been a strong surge in capital flows towards emerging mar-kets in search of relatively high yields because of rapid liquidity expansion

Chapter 7

CONCLUSIONS

49

and limited investment opportunities. Some countries chose to ignore theirimpact on exchange rates and the balance of payments, allowing their cur-rencies to float upward. Others have tried to limit their adverse impact onthe exchange rate and balance of payments through similar measures to thoseused in Thailand. But no country, including those with the most restrictivecapital account regimes, such as China and India, has been fully successfulin delinking their exchange rates and domestic financial conditions fromglobal capital markets and international capital flows because of closer inte-gration into the global economic system through various channels includingtrade and investment.7 This makes it all the more necessary to focus regionaland international cooperation on monitoring and managing capital flows andbringing multilateral discipline over policies of major countries which havea disproportionately large impact on global financial conditions.

Policy implications

The following are the policy implications to be drawn from the Thaiexperience:

1. Capital control measures should be applied in a counter-cyclical way.They are particularly important for checking a sudden upsurge in capi-tal flows. This is similar to the anti-surge mechanism which must beable to stem harmful short-term inflows and prevent a build-up of ex-ternal financial fragility.

2. The monetary authority should devise an appropriate mechanism forthe monitoring of capital outflows. As current liberalization of capitaloutflows eased pressures for a sharp rise in the THB value, it is criticalto note that the channels which facilitate normal outflows may alsoturn into those for a stampede when market confidence changes. It issuggested that such a mechanism needs to be equipped with certain

7 For the Asian experience in general, see Akyüz (2008); and for the specific experienceof China and India, see Yu (2009) and Chandrasekhar (2008) respectively.

50

trigger points which automatically become restrictive enough to pre-vent sudden outflows and capital flight.

3. Long-run measures to reduce volatility in the exchange rate should bethe focus of macroeconomic policy. Strengthening macroeconomic fun-damentals and raising competitiveness entail measures to promote in-dustrial upgrading, improve national technological capability and cor-responding human resource development. This is in addition to ensur-ing a strong and prudential financial system as well as efficient regula-tory agencies.

4. There is a need to actively engage in monitoring and take prompt ac-tions to prevent cases of “financial innovations” which seek to circum-vent or neutralize the impact of control measures. This is aimed spe-cifically at plugging loopholes and making the control measures aseffective as possible. Towards this end, sharing of experiences amongmonetary authorities of Third World countries should be instituted.

5. Regarding international cooperation, two issues are of particular con-cern. First, there is an urgent need to strengthen international pruden-tial supervision and regulation of global financial actors other than thebanking system. Hedge funds, mutual funds and other institutional in-vestors must be placed under BIS (Bank for International Settlements)-type reporting and “prudential regulation”. Of concern are the hedgefunds’ “flexible investment mandates and a relative absence of regula-tory constraints”, which give them “the freedom of employing varioussophisticated investment techniques” (ECB, 2007: 52). They can stillpose serious threats to a national financial system as before. Second,regional financial cooperation as outlined in the Chiang Mai Initiativeshould be strengthened to make use of the huge common pool of re-gional financial resources for urgent support to member countries withpayments difficulties arising from disruptive hot money flows(Sangsubhan, 2005).

51

References

Akyüz, Yilmaz (1995). “Taming International Finance”, in Jonathan Michie andJohn G. Smith (eds.), Managing the Global Economy (Oxford: Oxford Uni-versity Press).

Akyüz, Yilmaz (2008). The Current Global Financial Turmoil and Asian Develop-ing Countries, TWN Global Economy Series No. 11 (Penang: Third WorldNetwork).

Alesina, Alberto, Grilli, Vittorio and Milesi-Ferretti, Gian Maria (1994). “The Po-litical Economy of Capital Controls”, in Leonardo Leiderman and Assaf Razin(eds.), Capital Mobility: The Impact on Consumption, Investment and Growth(Cambridge: Cambridge University Press).

BOT (Bank of Thailand) (2004). “Foreign Exchange Policy and Intervention inThailand”, presented to BIS Deputy Governors’ meeting on “Forex Interven-tion: Motives, Techniques and Implications in Emerging Markets”, BIS, Swit-zerland, 2-3 December 2004.

BOT Report (various issues). Annual Report on Economic and Monetary Condi-tions.

BOT Statistics (various issues). Economic and Financial Statistics (Quarterly) (inThai).

BOT Statistics Tables (various years).Cardoso, E. and Goldfajn, H. (1997). “Capital Flows to Brazil: The Endogeneity of

Capital Controls”, IMF Working Paper No. 97/115, September.Chandrasekhar, C.P. (2008). Financial Liberalization and the New Dynamics of

Growth in India, TWN Global Economy Series No. 13 (Penang: Third WorldNetwork).

Chomthongdi, Jacques-Chai (2002). “Tobin Tax: The Debates in East and SouthEast Asia: Country Case Studies: Thailand”, in Asian Regional Exchange forNew Alternatives (ARENA), A Reader on Short Term Capital Flows: Arrest-ing Speculation and Volatility.

Classens, S., Dooley, M. and Warner, A. (1995). “Portfolio capital flows: hot orcold?”, The World Bank Economic Review, 9: 153-174.

ECB (European Central Bank) (2007). Financial Stability Review, June.Edwards, Sebastian (1999). “How Effective Are Capital Controls?”, NBER Work-

ing Paper No. 7413.Felix, David (1996). “Financial Globalisation versus Free Trade: The Case for the

Tobin Tax”, UNCTAD Review.FSRA (Financial Sector Restructuring Authority) (2002). “Financial Crisis and

Resolutions in Thailand 1997-2002, Roles of the Financial Sector Restructur-ing Authority” (Bangkok: Research and Information Office, FSRA).

52

Griffith-Jones, S. (1999). “Stabilizing Capital Flows to Developing Countries”, IDSBulletin, Vol. 30.

IMF (International Monetary Fund) (2007). World Economic Outlook: Globaliza-tion and Inequality, October.

Jomo, K.S. (ed.) (2001). Malaysian Eclipse: Economic Crisis and Recovery (Lon-don: Zed Books).

Klein, Michael W. and Olivei, Giovanni (1999). “Capital Account Liberalization,Financial Depth and Economic Growth”, NBER Working Paper No. 7384.

Krugman, Paul (1994). “The Myth of Asia’s Miracle”, Foreign Affairs, December.Lietaer, Bernard (1997). “From the Real Economy to the Speculative”, Third World

Network Features, December.Mecagni, Mauro, Atoyan, Ruben, Hofman, David and Tzanninis, Dimitri (2007).

“The Duration of Capital Account Crises – An Empirical Analysis”, IMFWorking Paper No. 07/258.

Mendoza, Ronal U. (2004). “International reserve-holding in the developing world:self-insurance in a crisis-prone era?”, Emerging Markets Review, Vol. 5: 61-82.

Pinto, S., Chuched, T. and Manuichachai, B. (2007). “Assessing Fiscal Vulnerabil-ity in Thailand: Fiscal Risks and Policy Implication”, BOT Discussion PaperDP/09/2007.

Prasartset, Suthy (1998). “The Economic Crisis in Thailand: Causes and Conse-quences”, paper submitted to a forum on “The Asian Economic Crisis and theRole of the Church: The IMF, Human Rights and the Church”, Seoul, 23-29August (Pax Romana ICMICA and Woori Theological Institute).

Reyes-Cantos, Jessica (2002). “How the Philippines Can Regulate Short-Term CapitalFlows: Looking at the Experience of Southeast Asia, Hong Kong, Taiwan,Latin America, and the Czech Republic”, in ARENA, A Reader on Short TermCapital Flows: Arresting Speculation and Volatility.

Rodrik, D. (1998). “Who Needs Capital Account Convertibility?”, Princeton Es-says in International Finance, 207: 55-65.

Sachs, J.D., Tornell, A. and Velasco, A. (1996). “Financial Crisis in Emerging Mar-kets: The Lessons from 1995”, Brookings Papers on Economic Activity, No.1.

Sangsubhan, Kanit (2005). “Exploring Ways to Enhance the Functions of theChiangmai Initiative in the Medium Term”, ASEAN Secretariat, March.

Sangsubhan, Kanit (2008). “Managing Capital Flows: The Case of Thailand”, ADBInstitute Discussion Paper No. 95.

Siamwalla, Ammar (1997). “Trying to figure out how Thailand got into such a mess”,The Nation, 12 November.

Suwanaporn, Chodechai (2007). “One year after the capital controls”, Fiscal PolicyOffice, Academic articles.

53

Taylor, Lance (1999). “Lax Public Sector, Destabilizing Private Sector: Origins ofCapital Market Crises”, in UNCTAD, International Monetary and FinancialIssues for the 1990s, Vol. X.

TDRI (Thailand Development Research Institute) (1997). “Financial Volatility andCurrency Speculation”, report submitted to the Department of Economic Af-fairs, Ministry of Foreign Affairs, Bangkok, November.

The Nation (Bangkok daily).Traisorat, T. (1998). “The Thai Financial and Economic Crisis of 1997-98: An Op-

portunity to Re-address the Fundamentals”, Warasarn Nitisat (Law Journal ofThammasat University), Vol. 28, No. 3, September.

Vichyanond, Pakorn (2001). “Complicated Difficulties behind Financial Resolu-tion: The Case of Thailand”, paper presented at the Sixth