Mark ở vũ trụ khácVIMark học AI
Mark ở vũ trụ khác · Another Universe Big Events Revisited #9 · Finance · Economics

The Asian Financial Crisis

Thailand, Indonesia, Korea, Malaysia and the whole region · July 1997 to 1998

What happenedone currency floated, then the region fell like dominoes
Scalethe rupiah lost 81% and the baht 56% of their value in about six months
Role-playyou are Thailand's central bank governor, mid-1997
Main sourcesIMF (programme releases, 2003 IEO independent evaluation), Bank of Thailand

Source: IMF, International Capital Markets Report 1998, chapter II (rupiah down 81%, baht down 56% by January 1998, against early July 1997).

Recap in 1 minute

Essence

Borrow short, build longMany firms and banks borrowed dollars short term to fund projects and land that took years.
Reserves looked full, were emptyThailand defended its exchange rate with forward contracts it did not disclose.
2 July 1997: the first dominoThe baht floated. Within five months the IMF had packages for Thailand, Indonesia and Korea.
The cure was contested tooThe IMF's own independent evaluation (2003): the initial fiscal tightening in Indonesia and Korea was not warranted.
The lessons stayedReal reserves, matched debt maturities, clear deposit guarantees, a regional swap network (Chiang Mai 2000).

Source: IMF, Boughton (2012), Tearing Down Walls, chapter 11 (undisclosed forward swaps, the float on 2 July 1997); IMF press releases 97/37, 97/50, 97/55; IEO (2003), chapter 4; ASEAN+3 joint statement, Chiang Mai, 6 May 2000.

Part 1

The question

Economies praised as "miracles" in the early 1990s: why did they fall within months into their deepest recessions in decades? Not just because of a few speculators. Because too many people borrowed short term, with other people's money, to do long-term things. When the lenders asked for it back at once, nobody had enough to pay.

Source: 1998 GDP from IMF Working Paper 99/138 (Berg, 1999); the "sudden reversal of capital flows" mechanism from the IEO (2003). The answer is Mark's view.

early 1990sMoney pours inNear-fixed exchange rates, cheap dollar loans, property and stock investment.
→
Jul 1997 to Jan 1998Money leaves at onceThe baht floats, lenders stop rolling over loans, regional currencies lose 55% to 81%.
→
1998 to 2000Recession, new rulesGDP falls, a government falls, then reserves, banks and a regional fund are rebuilt.

Source: depreciation from IMF, International Capital Markets Report 1998, chapter II.

Part 2

Timeline

From an exchange-rate decision in Bangkok to three large IMF packages in five months.

Timeline of seven milestones: 2 July 1997 Thailand floats the baht; 20 August 1997 IMF approves Thailand package; 5 November 1997 IMF approves Indonesia package; 4 December 1997 IMF approves Korea package; 21 May 1998 Suharto resigns; 1 September 1998 Malaysia imposes capital controls; 6 May 2000 Chiang Mai Initiative. The first four are only five months apart. 5 months: from one country to three IMF packages 2 Jul 1997Thailand floatsthe baht 20 Aug 1997IMF approvesThailand 5 Nov 1997IMF approvesIndonesia 4 Dec 1997IMF approvesKorea 21 May 1998Suhartoresigns 1 Sep 1998Malaysiacapital controls 6 May 2000Chiang MaiInitiative spacing between milestones is not to time scale
The first four milestones fit into five months. After that the crisis moved from currencies to politics (Indonesia) and to a policy debate (Malaysia chose a different path from the IMF).Source: IMF, Boughton (2012), chapter 11 (2 July 1997); IMF press releases 97/37 (20 Aug 1997), 97/50 (5 Nov 1997), 97/55 (4 Dec 1997); Suharto's resignation per Britannica and IPS, 21 May 1998 (secondary source); Malaysian capital controls per Jomo K.S., Oxford GEG Working Paper 2004/08 (secondary source); ASEAN+3 joint statement, Chiang Mai, 6 May 2000.
Depreciation against the US dollar by January 1998: the Indonesian rupiah lost 81 percent from 1 July 1997; the Thai baht lost 56 percent from early July 1997; the Korean won lost 55 percent from 1 October 1997 to its low in late December 1997. loss against the USD, to the low around Dec 1997 to Jan 1998 rupiah−81% baht−56% won−55% rupiah, baht: against early July 1997. won: against 1 October 1997.
One dollar of debt became two, then five. Anyone who owed dollars but earned rupiah saw their debt balloon several times over in a few months, without borrowing a cent more.Source: IMF, International Capital Markets Report 1998, chapter II. "Several times over" is derived from the depreciation: an 81% loss means a dollar costs about 5 times as much.
Real GDP growth in 1998, IMF estimates from 1999: Indonesia minus 13.7 percent, Thailand minus 9.4, Malaysia minus 6.7, Korea minus 5.8. real GDP in 1998, % (IMF estimates from 1999) −13.7−9.4−6.7−5.8 IndonesiaThailandMalaysiaKorea
Indonesia fell furthest. Later datasets were revised several times (one table shows Thailand at −10.2% and Malaysia at −7.4%), so the article uses a single dataset to compare fairly.Source: IMF Working Paper 99/138, Berg (1999), "The Asia Crisis: Causes, Policy Responses, and Outcomes". Other dataset: FAO table citing IMF data (secondary source).
1996 · A large deficitThailand's current account deficit was about $14.7 billion, roughly 8% of GDP: the country spent more abroad than it earned, filling the gap with borrowed money.
Early 1997 · Defending the pegThe Bank of Thailand sold dollars through forward swaps to hold up the baht. Published reserves still looked intact; forward commitments reached about $23 billion and were not disclosed.
27 Jun 1997 · 16 finance companies suspended42 more on 5 August 1997, 58 in total.
2 Jul 1997 · The baht floatsIt lost nearly 17% against the dollar in offshore trading that day.
20 Aug 1997 · IMF package for ThailandA stand-by credit of SDR 2.9 billion (about $3.9 billion), drawn over 34 months.
Nov 1997 · IndonesiaThe IMF approved SDR 7.3 billion (about $10.1 billion) on 5 November. That month 16 banks were closed.
4 Dec 1997 · KoreaThe IMF approved SDR 15.5 billion, 19 times Korea's quota. With other countries and institutions, the package was about $57 to 58 billion.
Jan 1998 · Currency lowThe rupiah had lost 81% and the baht 56% since early July 1997.
21 May 1998 · Suharto resignsAfter 32 years in power, amid riots and economic crisis. Vice President Habibie took over.
1 Sep 1998 · Malaysia goes its own wayCapital controls, a peg of 3.80 ringgit to the dollar, no IMF loan. The peg was dropped in 2005.
6 May 2000 · Chiang Mai InitiativeASEAN with China, Japan and Korea set up a currency swap network to help each other in a liquidity squeeze.

Source: CRS (1998), "The 1997–98 Asian Financial Crisis" (1996 deficit; Korea package $58.2 billion); PIIE, chapter 5 (about $57 billion; secondary source); IMF, Boughton (2012), chapter 11 (undisclosed swaps, −17% on 2 July); forward commitments of about $23 billion per studies of the Thai crisis (secondary source, figures vary); Bank of Thailand, "Lessons learnt from the Asian Financial Crisis" (58 finance companies); IMF press releases 97/37, 97/50, 97/55; IMF Issues Brief 00/05 (16 banks, November 1997); IMF, International Capital Markets Report 1998; Britannica, IPS (Suharto; secondary source); Jomo K.S. (2004) and Malaysian press (capital controls; secondary source); ASEAN+3 joint statement, 6 May 2000.

How Vietnam was affected

No currency crash like the countries with open capital markets, but a clear slowdown: GDP growth of 8.15% (1997), 5.76% (1998), 4.77% (1999); exports grew 26.6% in 1997 but only 1.9% in 1998; registered FDI fell from $4.6 billion (1997) to $1.5 billion (1999).

Where sources disagree

The 5.76% for 1998 is the General Statistics Office figure as quoted by the press. An IMF table from 1999 shows only 3.5% for 1998 (an early estimate). Mark uses the official domestic figure and notes the gap.

Source: General Statistics Office figures as quoted by VnEconomy and the Government News portal (secondary source); IMF Staff Country Report 99/55 (3.5%); FDI per a Vietnamese paper at the IMF FDI conference, Hanoi 2002 (secondary source).

Post 1

Timeline

2 Jul 1997rupiah −81%3 IMF packagesVietnam
Post 2

Key decisions

play the governor16 banksIMF debatespeculator myth
Post 3

What the world learned

real reservesChiang Maimatched maturitiesMark's score
Part 3

Five key decisions

Seen through the eyes of the time: who decided, what they knew, what they chose, and what happened. No judging with hindsight.

1 · Holding the peg with forward contracts

forward swaps · peg defenceearly 1997 to 2 Jul 1997
Who decidedBank of Thailand
What they knewThe near-fixed rate had held for over a decade. Floating would hurt firms with dollar debts. Speculators were selling baht.
What they choseSell dollars forward to support the baht, without disclosing the commitments.
What happenedBy 2 July 1997 net reserves were almost gone, forcing a float. Markets, and the IMF, could not see the real figure.
Illustrative diagram, not to scale: published reserves still looked full; but about 23 billion dollars had been promised in forward sales; what was still usable on 2 July 1997 was only about 2.8 billion dollars according to the Nukul Commission. published reserves still look full in reality promised in forward sales: about $23 billion still usable: about $2.8 billion illustrative diagram, not to scale
Real reserves = reserves minus what you have promised. The biggest mistake was not only trying to hold the peg, but holding it with commitments nobody could see, so when it had to stop there was no cushion left.Source: IMF, Boughton (2012), chapter 11 (swaps not disclosed to markets or the IMF); about $23 billion per studies of the Thai crisis (secondary source, figures vary); about $2.8 billion per the Nukul Commission report (1998), read via an online copy (secondary source).

2 · Floating the baht

the float2 Jul 1997
Who decidedBank of Thailand and Ministry of Finance
What they knewUsable reserves were nearly exhausted. Finance companies had been suspended since 27 June.
What they choseDrop the basket peg and move to a managed float.
What happenedThe baht lost nearly 17% that day and 56% by January 1998. Investors began scrutinising countries that looked like Thailand.

3 · Accepting IMF packages with tightening conditions

IMF programmesAug 1997 to Dec 1997
Who decidedThe governments of Thailand, Indonesia and Korea with the IMF
What they knewDollars were needed urgently. The IMF's familiar crisis recipe was tighter budgets, higher interest rates and cleaning up weak banks.
What they choseThailand (20 Aug), Indonesia (5 Nov) and Korea (4 Dec) took stand-by credits, drawn in tranches against conditions.
What happenedRecessions were deeper than forecast. Budget targets were eased once the output collapse was clear. Still the biggest point of debate (see Fact-check).

4 · Closing 16 banks in Indonesia

bank closures · deposit guaranteeNov 1997
Who decidedThe Indonesian government, within its IMF programme
What they knewMany banks were weak. Keeping "zombie" banks alive costs money. But there was no blanket deposit guarantee yet.
What they choseClose 16 banks; small depositors were guaranteed up to 20 million rupiah, large depositors were not.
What happenedPeople withdrew money from other banks too. In January 1998 a blanket guarantee on deposits had to be given.

Source: IMF Issues Brief 00/05 (16 banks, lesson on clear deposit guarantees); Yale Program on Financial Stability (20 million rupiah limit, spreading runs, blanket guarantee in January 1998; secondary source). IMF, "Views and Commentaries", 26 Jan 1999, rejects the claim that the IMF advised closure after "a few days of quick study".

5 · Malaysia imposes capital controls, no IMF loan

capital controls1 Sep 1998
Who decidedThe Malaysian government (Prime Minister Mahathir)
What they knewThe countries with IMF loans were in deep recession. The ringgit was falling and capital was leaving.
What they choseRestrict capital outflows, peg at 3.80 ringgit to the dollar, ban offshore ringgit trading.
What happenedInvestors objected strongly at the time. The real effect is still debated. The peg was dropped in 2005.

Source: Jomo K.S. (2004), Oxford GEG Working Paper 2004/08; The Malaysian Insight, iMoney (secondary source).

General sources for this part: IMF, Boughton (2012), chapter 11; IMF press releases 97/37, 97/50, 97/55; IMF press conference, 20 Jan 1999; CNN Money, 2 Jul 1997 (secondary source); Bank of Thailand (suspensions of 27 Jun 1997).

Part 4

What the world learned

What changed after 1998: how reserves are kept, how banks are rescued, how neighbours back each other up, and how the IMF grades itself.

Before and after. Before: thin and opaque reserves; short-term foreign-currency debt; bank closures without deposit guarantees; only the IMF to borrow from. After: thicker reserves disclosed clearly; short-term debt tracked against reserves; clear deposit guarantees; an added regional swap network, Chiang Mai. BEFORE 1997AFTER 1998 thin reserves, hidden commitments short-term dollar debt, long projects close banks, deposits partly covered urgent dollars: only the IMF real reserves = reserves minus promises track short-term debt against reserves deposit guarantees stated up front a regional net: Chiang Mai 2000
Four breaks, four patches. Each line on the right is a lesson written down in an official document: the Bank of Thailand's own "Tom Yum Kung" lessons, the IMF's lesson on deposit guarantees, the ASEAN+3 Chiang Mai Initiative.Source: Bank of Thailand, "Lessons learnt from the Asian Financial Crisis" (short-term debt 65% of total foreign debt, reserves only 70.4% of short-term debt at end-1997); IMF Issues Brief 00/05; ASEAN+3 joint statement, 6 May 2000.
Real reserves, clearly disclosedThailand's biggest lesson: reserves you can see are not reserves you can use. Subtract the commitments.
Short-term debt against reservesAt end-1997 Thailand's reserves covered only 70.4% of its short-term debt. The ratio became a watched indicator.
Say the deposit guarantee firstIndonesia 1997: closing banks before depositors feel safe sets off runs across the system.
A regional netThe Chiang Mai Initiative (2000), multilateralised into a single agreement in 2010.

Source: Bank of Thailand, "Lessons learnt from the Asian Financial Crisis"; IMF Issues Brief 00/05; ASEAN+3 joint statement, 6 May 2000; ASEAN's ASEAN+3 finance cooperation page (multilateralisation in 2010).

The IMF reassesses itself: three times, three tones
Three IMF look-backs. January 1999: staff self-assessment, policy design broadly appropriate, but budget targets rested on wrong forecasts and could have been eased sooner. June 2000: advice not flawless, but corrected promptly. July 2003: Independent Evaluation Office, initial fiscal tightening in Indonesia and Korea not warranted, though not the cause of the collapse. Jan 1999"broadly appropriate"wrong forecasts, eased late Jun 2000"not flawless"but corrected promptly Jul 2003 · IEOinitial tightening"not warranted" the later and more independent the review, the clearer the admission
An admission, with limits. The IEO wrote: "In view of output developments, the initial tightening of fiscal policy in Indonesia and Korea was not warranted." But the same report says the tightening was mild, and the collapse came from balance-sheet effects (foreign-currency debt ballooning as currencies fell) that the programme design did not factor in.Source: IMF press conference, 20 Jan 1999, and Lane et al. (1999), IMF Occasional Paper 178; IMF Issues Brief 00/05 (June 2000); IEO (2003), The IMF and Recent Capital Account Crises, chapter 4.
Part 5

Fact-check: what's true and what's myth

14 things often said about the 1997 crisis, checked against IMF documents, the Bank of Thailand and clearly cited sources. Tap each row for the reasoning.

Why speculators won in Thailand: four conditions already in place. A current account deficit of about 8 percent of GDP in 1996. Short-term debt at 65 percent of foreign debt. Reserves locked in forward contracts of about 23 billion dollars. 58 finance companies suspended. Speculators only pushed dominoes that were already standing close together. speculators can only push dominoes that already stand close together deficit ~8% GDPin 1996 short-term debt 65%of foreign debt ~$23bn forwardsreserves locked 58 finance cos.suspended baht sellers: real, but the last push, not the reason the dominoes stood so close
"Speculators did it" is half the story. There were baht sellers, and the Bank of Thailand spent heavily fighting them. But in Korea and Indonesia the biggest blow was lenders refusing to roll over short-term loans, what the IEO calls a sudden reversal of capital flows.Source: CRS (1998) and UNCTAD, Akyüz (deficit about 8% of GDP); Bank of Thailand (65%, 58 companies); forward commitments per studies of the Thai crisis (secondary source); IEO (2003).
SOLIDThailand floated the baht on 2 July 1997, starting the chain of crises

The IMF's official history (Boughton 2012, chapter 11): the baht lost nearly 17% against the dollar in offshore trading that day. CNN Money the same day (secondary source): a move to a managed float.

SOLIDSpeculators attacked the baht

The Bank of Thailand sold dollars forward to fight selling pressure on the baht (Boughton 2012). Speculation was real; the question is why it won (next row).

FALSE"The crisis was caused by foreign speculators"

False if meant as the main cause. A current account deficit of about 8% of GDP (CRS, UNCTAD), short-term debt at 65% of foreign debt and larger than reserves (Bank of Thailand), reserves locked in forward contracts, 58 weak finance companies. In Korea and Indonesia, lenders stopped rolling over loans (IEO 2003). The best-known blaming of speculators came from Malaysian Prime Minister Mahathir in 1997 (Jomo 2004, secondary source).

SOLIDThe rupiah lost more than 80% of its value

IMF, International Capital Markets Report 1998: by January 1998 the rupiah had lost 81% against 1 July 1997. The baht lost 56%, the won 55% (from 1 October 1997).

SOLIDThe IMF package for Korea was unprecedented relative to quota

Press release 97/55: SDR 15.5 billion, 1,939% of Korea's quota. The total package with other parties was about $57 to 58 billion (CRS $58.2 billion; PIIE about $57 billion, secondary source).

SOLIDIndonesia closed 16 banks without fully guaranteeing deposits, setting off mass withdrawals

IMF Issues Brief 00/05 cites the closure of 16 banks in November 1997 as a lesson on clear deposit guarantees. Yale YPFS (secondary source): coverage only up to 20 million rupiah, runs spread across the system, blanket guarantee from January 1998. The IMF (26 Jan 1999) rejected the claim that it advised closure "after a few days of quick study", saying it agonised because more banks were known to be weak.

SOLIDThe IMF itself concluded that initial fiscal tightening in Indonesia and Korea was not needed

IEO (2003), chapter 4: "In view of output developments, the initial tightening of fiscal policy in Indonesia and Korea was not warranted." This is the IMF's own independent evaluation office.

SOLIDSuharto resigned on 21 May 1998 after 32 years in power

Britannica, IPS 21 May 1998 (secondary source): amid riots and economic crisis; Vice President Habibie took over.

CAUTION"IMF austerity caused the collapse"

The IEO says the initial tightening was not warranted, but "mild" and "not the cause of the output collapse"; the cause was balance-sheet effects the programmes did not factor in. Many economists criticise more strongly. A fair conclusion: the medicine was partly the wrong dose, but the illness came first.

CAUTION"Malaysia refused the IMF, imposed capital controls and so recovered better"

Malaysia imposed controls on 1 September 1998, after the region's worst panic had passed. Studies still disagree; there is no clean comparison. Investors objected strongly at the time (Jomo 2004; Malaysian press, secondary source).

CAUTION"On the day of the float, Thailand's net reserves were only about $2.8 billion"

The Nukul Commission's figure, but Mark could only read it via an online copy. Other secondary sources count differently (28 of 30 billion used, or 32 of 33 billion). The direction is certain: usable reserves were nearly exhausted.

FALSE"The IMF never admitted mistakes in Asia"

IMF press conference January 1999 (wrong forecasts, budgets could have been eased sooner), Issues Brief June 2000 ("not flawless"), IEO July 2003 ("not warranted"). The IMF's admissions have limits, but they exist.

FALSE"IMF packages are free money"

They are stand-by credits: drawn in tranches, each tied to conditions and reviews, then repaid. For example Thailand could draw SDR 1.2 billion at once and Korea SDR 4.1 billion, the rest on a schedule (press releases 97/37, 97/55).

FALSE"Vietnam was not affected"

GDP growth of 8.15% (1997), 5.76% (1998), 4.77% (1999); exports grew only 1.9% in 1998; registered FDI fell sharply (General Statistics Office via the press and a paper at the 2002 IMF conference, secondary source). Vietnam was hit less because its capital market was not yet open, but it was not untouched.

Source: as noted in each row; full list in Part 9.

Part 6

Assessment: how the world responded

Scores are Mark's assessment based on the sources in Part 9, not data from a study.

Early detection
2/5
Transparency
2/5
Coordination
3/5
Learning lessons
4/5
Done right
  • Thailand's IMF package came in under two months, under emergency procedures
  • Budgets were eased once recessions proved deeper than forecast
  • Public self-assessment, including an independent one (2003)
  • The region built its own net: Chiang Mai 2000
Done wrong
  • Thailand's forward commitments were not disclosed
  • Initial tightening not warranted (IEO)
  • 16 banks closed before deposits were fully guaranteed
  • The ballooning of foreign-currency debt was not factored in

Source: IMF press releases 97/37, 97/50, 97/55 (emergency procedures); IMF press conference, 20 Jan 1999; IEO (2003); IMF Issues Brief 00/05; Boughton (2012), chapter 11; ASEAN+3 joint statement, 6 May 2000.

The illness came from the economies themselves: borrowing short, building long, hiding the real numbers. The medicine arrived fast but partly at the wrong dose. What deserves most credit is that both sides rewrote their own lessons.

More: why the "speculators did it" myth is easy to believe

A villain from outside is a neat, comfortable story: nobody at home has to own a mistake. The truth is messier: many people at home borrowed dollars short term because they were cheap, believing the exchange rate would stand still forever.

The messy version is more useful at work: most breakdowns don't come from one blow, but from short-term commitments piling up under a long-term project.

The "More" section is Mark's commentary, based on sourced facts in Parts 2, 3 and 5.

Part 7

Work lesson: don't borrow short term with other people's money to do long-term work

At work, "money" can be people borrowed from another team, a budget granted quarter by quarter, a manager's goodwill, or a supplier you don't control.

Two ways to fund a 12-month project. Mismatched: resources borrowed three months at a time, each expiry needs a renewal, and if all are recalled at once the project stops halfway. Matched: resources committed for the full 12 months, plus a buffer. 12-month project mismatched borrowed 3 monthsrenewed?renewed?recalled every expiry is a moment when someone else decides for you matched committed for all 12 monthsbuffer illustrative diagram, no data
Each habit patches one break from 1997. Borrowing short to build long, borrowing in something you can't produce, reserves that ignore commitments, and holding an "exchange rate" too long.Source: the breaks come from Parts 2, 3 and 5; the habits are Mark's suggestions.
Match maturitiesA 12-month job needs its main resources committed for 12 months. Write down the expiry date of each one.
Borrowed in which "currency"?Does your team control the resource? Borrowed people and another team's tools are "dollar debt": if they change their mind, you pay many times over.
Real buffer = buffer minus promisesThe team's free time minus the things already promised but not yet on the calendar. Make that number public.
Write the float threshold in advanceDefending a deadline or a promise? Decide now: at what cost you stop defending it and tell the truth.
Check your own team: 5 questions

1. List the three main resources of your biggest project. Until what date is each committed? Does any expire before the project ends?

2. If every borrowed person and temporary budget were pulled in the same week, what percentage of the project could still run?

3. Is there a commitment the team has promised but not put on the board (like an undisclosed forward contract)?

4. What "exchange rate" are you defending: a deadline, a price, a promise to a client? What does one more month of defending cost?

5. If you must "close" a workstream, have the people involved been told clearly what is still guaranteed (the 16-bank lesson)?

In view of output developments, the initial tightening of fiscal policy in Indonesia and Korea was not warranted.IMF Independent Evaluation Office (IEO), 2003, chapter 4.
Part 8

Further reading

IEO (2003), The IMF and Recent Capital Account Crises: Indonesia, Korea, BrazilThe IMF's independent evaluation office grades the IMF. Read chapter 4 on fiscal policy and bank closures.
Lane et al. (1999), IMF Occasional Paper 178IMF staff's first self-assessment of the programmes in Indonesia, Korea and Thailand.
Boughton (2012), Tearing Down Walls, chapter 11 "Asian Flu"The IMF's official history of 1990–1999, week by week through the summer of 1997.
Bank of Thailand, "Lessons learnt from the Asian Financial Crisis"Lessons written by the institution at the centre of it.
Big Events Revisited #10: Hurricane Katrina (2005)After a drill, fix what the drill taught you.
Part 9

Sources

IMF and government documents first. Secondary sources are labelled. The working environment could not reach imf.org or bot.or.th, so these documents were checked through excerpts of those very pages in search results, not in full text.

  1. IMF, Press Release 97/37, 20 Aug 1997 (Thailand, SDR 2.9 billion). imf.org · Letter of Intent, 14 Aug 1997 imf.org
  2. IMF, Press Release 97/50, 5 Nov 1997 (Indonesia, SDR 7.338 billion). imf.org
  3. IMF, Press Release 97/55, 4 Dec 1997 (Korea, SDR 15.5 billion, 1,939% of quota). imf.org
  4. IMF Independent Evaluation Office (IEO), The IMF and Recent Capital Account Crises: Indonesia, Korea, Brazil, 28 Jul 2003, chapter 4. imf.org
  5. IMF, Issues Brief 00/05, "Recovery from the Asian Crisis and the Role of the IMF", June 2000. imf.org
  6. IMF, press conference, 20 Jan 1999, and Lane et al., Occasional Paper 178 (1999). imf.org · imf.org
  7. IMF, "Views and Commentaries", 26 Jan 1999. imf.org
  8. IMF, Boughton (2012), Tearing Down Walls: The International Monetary Fund 1990–1999, chapter 11. imf.org
  9. IMF, International Capital Markets Report 1998, chapter II. imf.org
  10. IMF, Working Paper 99/138, Berg (1999); Staff Country Report 99/55 (Vietnam). elibrary.imf.org
  11. Bank of Thailand, "Lessons learnt from the Asian Financial Crisis". bot.or.th
  12. US Congressional Research Service (CRS), "The 1997–98 Asian Financial Crisis" (1998). fas.org
  13. UNCTAD, Akyüz, "Causes and sources of the Asian financial crisis". unctad.org
  14. ASEAN+3, Finance Ministers' joint statement, Chiang Mai, 6 May 2000. adb.org
  15. Secondary sources: Nukul Commission report (1998) via an online copy; studies of the Thai crisis (forward commitments of about $23 billion); CNN Money, 2 Jul 1997; Yale Program on Financial Stability (Indonesian deposit guarantee); Boediono via detik.com and Kontan (2018); Britannica, EBSCO, IPS (Suharto); Jomo K.S., Oxford GEG Working Paper 2004/08, The Malaysian Insight, iMoney (Malaysia); PIIE, chapter 5 (Korea package); VnEconomy, Government News portal (General Statistics Office figures for Vietnam); paper at the IMF FDI conference, Hanoi 2002; G24 (2024), "Beyond the Chiang Mai Initiative"; FAO (GDP table citing IMF data).