This article was downloaded by: [Tufts University]On: 14 November 2014, At: 13:55Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number:1072954 Registered office: Mortimer House, 37-41 Mortimer Street,London W1T 3JH, UK

Journal of the Asia PacificEconomyPublication details, including instructions forauthors and subscription information:http://www.tandfonline.com/loi/rjap20

Asia's financial crisis inhistorical perspectiveJames W. Dean aa Simon Fraser Universityb Western Washington UniversityPublished online: 02 May 2007.

To cite this article: James W. Dean (1998) Asia's financial crisis in historicalperspective, Journal of the Asia Pacific Economy, 3:3, 267-283, DOI:10.1080/13547869808724653

To link to this article: http://dx.doi.org/10.1080/13547869808724653

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of allthe information (the “Content”) contained in the publications on ourplatform. However, Taylor & Francis, our agents, and our licensorsmake no representations or warranties whatsoever as to the accuracy,completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views ofthe authors, and are not the views of or endorsed by Taylor & Francis.The accuracy of the Content should not be relied upon and should beindependently verified with primary sources of information. Taylor andFrancis shall not be liable for any losses, actions, claims, proceedings,demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, inrelation to or arising out of the use of the Content.

This article may be used for research, teaching, and private studypurposes. Any substantial or systematic reproduction, redistribution,reselling, loan, sub-licensing, systematic supply, or distribution in any

form to anyone is expressly forbidden. Terms & Conditions of accessand use can be found at http://www.tandfonline.com/page/terms-and-conditions

Dow

nloa

ded

by [

Tuf

ts U

nive

rsity

] at

13:

55 1

4 N

ovem

ber

2014

ASIA'S FINANCIAL CRISIS IN HISTORICAL

PERSPECTIVE

James W. Dean

Abstract

This article places the debate between prescribing illiquidity and prescribinginsolvency in historical perspective, by comparing Asia's current debt crisiswith Latin America's during the 1980s. It is then suggested that none of theAsian economies is in any meaningful sense 'insolvent,' and that those whooppose the IMF-led 'bailout' fall into three equally misguided camps; - fiscalconservatives, left wing moralists, and right wing academics. This is not tosuggest that the IMF's response to the Asian crisis has been flawless: it has not.Finally, the article argues that South Korea's currency crisis spread irration-ally from the others', and was compounded by a 'reverse free rider' phenom-enon among foreign lenders. A short appendix looks at South Korea's debthistory between 1960 and 1989.

Keywords Asian crisis, banking crisis, illiquidity, insolvency, IMF, South Korea.

1. I N T R O D U C T I O N

The 1980s saw an 'international debt crisis' that began in Mexico and quicklyspread throughout Latin America and to parts of Asia and Eastern Europe.The crisis was triggered in mid-1982 and took seven years to resolve. Centralto the resolution process was a distinction between illiquidity and insolvency,first emphasized by Cline (1983, 1984). If sovereign borrowers were simplyilliquid, it was argued, policy should center on 'liquidity relief: reschedul-ing and new lending. If they were insolvent, it should center on 'debt relief:permanent reduction of debt obligations. Ultimately, liquidity relief provedinsufficient, and the crisis was resolved with the so-called 'Brady Plan' thatinvolved substantial debt reduction by commercial bank creditors.

The period 1994-95 saw a second Mexican crisis that was correctlydeemed to be a liquidity crisis, and rather adeptly resolved via a massive andunprecedented infusion of funds from the US Treasury and the IMF. Thatcrisis was more or less confined to Mexico, albeit with some tremors trans-mitted to Argentina. In short, the crisis proved both transitory and local.

Mid-1997 saw a currency crisis in Thailand that spread almost overnight

Journal of the Asia Pacific Economy 3(3) 1998:267-283 1354-7860 © Routledge 1998

267

Dow

nloa

ded

by [

Tuf

ts U

nive

rsity

] at

13:

55 1

4 N

ovem

ber

2014

JAMES W. DEAN

to Malaysia, Indonesia, and the Philippines. Within weeks the IMF, joinedby other multilateral and bilateral lenders, had committed $24 billion toThailand and $32 billion to Indonesia. But it was not until November of1997 that the crisis hit Asia's third largest economy, South Korea. When itdid, the impact was far more dramatic than expected. The won fell repeat-edly by 10 per cent per day, the limit of its trading band; several major firmswere declared insolvent, and by mid-December smaller firms were failing atthe rate of fifty per day. Short-term interest rates soared to 30 per cent andabove. The IMF, World Bank, Asian Development Bank and governmentsof eleven countries committed $57 billion to a rescue package, dwarfing thepackages already committed to Thailand and Indonesia, not to mention therecord-setting 1995 Mexican bailout.

As is its practice, the IMF attached strenuous conditions to its Korean1

loan package, including the closure of firms and financial institutionsdeemed to be insolvent, and sharp hikes in interest rates. This latter partof the package is highly controversial (see, for example, Sachs 1997).Indeed, as of mid-January 1998, the relevant Asian economies are continu-ing to deteriorate, and deteriorate substantially: despite (or, some wouldsay, because of) the IMF programs. While substantially higher interest ratesshould attract and retain capital inflows from abroad, they have palpablyfailed to do so. But higher interest rates have increased debt service obli-gations on outstanding stocks of internal debt. In some cases, such increaseshave been sufficient to render firms and financial institutions that werehitherto simply illiquid, over the brink into insolvency, especially sincedebt-equity ratios - especially in Korea - are extraordinarily high by inter-national standards. This line of argument suggests that the macroeconomicconditions being imposed by the IMF are dead wrong. It may also suggestthat other harsh (but necessary) aspects of the program, such as closuresof insolvent banks and firms, should be subordinated to its 'bailout' dimen-sion,2'3 at least for the moment. Indeed the chief economist of the WorldBank, Joseph Stiglitz, is remarkably critical of his sister institution's policiesin Asia: 'You don't want to push these economies into severe recession. Oneought to focus on the things that caused the crisis, not on things that makeit more difficult to deal with.'4

A second line of argument suggests the reverse: that bailouts should besubordinated to conditionality. Since it was misguided internal allocationof credit that got Asian countries into difficulty, no bank or firm in a bor-rowing country should be shielded from bankruptcy by IMF or other officialbailouts, even temporarily. Nor should foreign lenders be spared the con-sequences of their past folly. To do otherwise would be to generate evenmore 'moral hazard' than was generated (it is claimed) by the Mexicanrelief package of 1995. Paul Krugman, while stopping short of endorsingthis view in its extreme, has given it implicit support with an analysis of theunderregulated and overcomforted Asian banks' 'Panglossian' overlending

268

Dow

nloa

ded

by [

Tuf

ts U

nive

rsity

] at

13:

55 1

4 N

ovem

ber

2014

ASIA'S FINANCIAL CRISIS

(Krugman 1998). According to Krugman, the rot began in Japan during the1980s, and progressed to the 'little tigers' in the 1990s.

The two sides of this debate ascribe the failure of the IMF's Asian pro-grams to stem capital flight and currency collapse to virtually oppositecauses: one side to excessively harsh (macroeconomic) conditionality thathas led to unnecessary bankruptcies, and the other side to excessively laxconditionality that has kept ill-conceived enterprises alive for too long.

This article places the debate between prescribing illiquidity and pre-scribing insolvency in historical perspective, by comparing Asia's currentdebt crisis with Latin America's during the 1980s. It is then suggested thatnone of the Asian economies is in any meaningful sense 'insolvent,' andthat those who oppose the IMF-led 'bailout' fall into three equally mis-guided camps: fiscal conservatives, left-wing moralists, and right-wing aca-demics. This is not to suggest that the IMF's response to the Asian crisis hasbeen flawless: it has not. IMF conditionality has wrongly emphasized macro-economic frugality, prescribing fiscal, monetary, and growth targets thatmay have been appropriate for Latin America in the 1980s but ignore thereality of an already-frugal Asia in the 1990s. Finally, the article argues thatwhereas currency crises and consequent illiquidity in Thailand, Indonesia,and Malaysia were arguably of their own making, South Korea's currencycrisis spread irrationally from the others, and was compounded by a 'reversefree-rider' phenomenon among foreign lenders. A short appendix looks atKorea's debt history between 1960 and 1989.

2. ILLIQUIDITY AND INSOLVENCY IN THE 1980s5

Undoubtedly, many of the severely indebted developing countries (SIMICs)that encountered payments difficulties in the 1980s had misallocated muchcf their borrowed capital. However, the source of the crisis that began inmid-1982 was not insolvency, but rather illiquidity. The trigger for a liquid-ity squeeze was the sharp rise in international interest rates that resultedfrom a sustained effort by the US Federal Reserve and other developed-country central banks to squeeze inflation and inflationary expectations outcf their economies. Between late 1979 and early 1982, the London InterbankOffer Rate (LIBOR) rose from about 10 per cent to over 20 per cent. Sincemost syndicated bank loans to the SIMICs were priced on floating LIBOR,debt service obligations literally doubled within a two-year period.

The liquidity squeeze on borrowing countries' balance of payments canbe analyzed with the help of the following formula:

(1) NPR= (iD+K-L)/X

where NPR denotes the 'net payments ratio,' ithe rate of interest on foreigndebt, D the stock of foreign debt obligations, K annual capital flight, L

269

Dow

nloa

ded

by [

Tuf

ts U

nive

rsity

] at

13:

55 1

4 N

ovem

ber

2014

JAMES W. DEAN

annual new lending from abroad, and X annual export revenues. In theearly 1980s, NPR for most SIMICs increased sharply because i (LIBOR) vir-tually doubled. To make matters worse, the central-bank-induced recessionin developed countries caused commodity prices to plummet and importvolumes to decline; in combination, this meant that developing-countryexport revenues, X, also declined sharply.

But that was far from the end of the story. When Mexico became the firstmajor borrower to interrupt debt service payments in August 1982, foreignlenders almost immediately interrupted their flows of new funds: that is, Ldropped dramatically. Moreover, international lenders also truncated theirflows of funds to all developing-country borrowers that they believed,rightly or wrongly, to pose similar risks as Mexico. Thus it was that a local,Mexican, balance-of-payments crisis was transformed almost overnight intoa global, developing-country crisis.

In fact the inflows of new lending, L, had been so large in the late 1970sand early 1980s that, for most SIMICs, NPR was negative. The net flow offunds was from lenders to borrowers, rather than the reverse. In effect,between late 1979 and early 1982, bank lenders from the developed coun-tries had been financing developing countries' interest payments on previ-ously contracted debt. Beyond that, they had, in many cases, been financingsubstantial capital flight, K,

By September 1982, the NPRs of perhaps a dozen borrowing countries,mostly in Latin America, Asia and Eastern Europe, had turned negative.Creditor banks quickly recognized this as an unsustainable liquidity crisis:debt service payments, iD, simply could not be maintained in the short runwithout a resumption of new lending, L, or equivalently at this level ofabstraction, rescheduling of payments whereby interest or, more com-monly, principal payments were either postponed or amortized over alonger period of time. Accordingly, with reinforcement from the IMF andthe so-called Paris Club of bilateral country lenders, the large internationalbanks mounted a strategy of 'concerted liquidity relief,' under which L wasresumed and rescheduling agreements were implemented.

The term 'concerted' referred to the fact that voluntary new lending andrescheduling by individual banks was blocked by a free-rider barrier. Freshfunds or principal postponement from any individual bank could not befully recaptured, as increases in debt service payments could not be dedi-cated to particular creditors. Rather, liquidity relief to sovereign borrowerswas likely to be spread more or less pro rata amongst all creditors. Thisproblem was reinforced by the so-called 'sharing' clauses written into syn-dicated loan contracts. The incentives facing individual lenders weresubject to a free-rider barrier: it was in their interests to hold back from liq-uidity relief in the hope of free riding on liquidity relief by others. As aresult, no liquidity relief was likely unless all creditors agreed to contributein concert, on a pro rata basis.

270

Dow

nloa

ded

by [

Tuf

ts U

nive

rsity

] at

13:

55 1

4 N

ovem

ber

2014

ASIA'S FINANCIAL CRISIS

The rationale for liquidity relief rested on the assumption that paymentsproblems were temporary rather than permanent or, in Cline's terminol-ogy, the assumption that sovereign borrowers were illiquid rather thaninsolvent. In effect, this was an assumption that expected default rateswould fall rather than rise if liquidity was infused. If d denotes the expecteddefault rate6 before new lending, and d* the rate after new lending, thisreasoning suggested that new lending should proceed as long as:

(2) d*L< (d-d*)D

The left side of equation (2) represents creditors' expected losses from newlending, and the right side their expected gains from reduced default rateson outstanding debt claims (Krugman 1988, 1989).

The rationale embodied in equation (2) was valid under three assump-tions:

(i) All creditors contributed their pro rata shares of new lending: that is,new lending was organized on a concerted basis that was involuntaryfrom each individual lender's point of view but voluntary on a collec-tive basis.

(ii) Each contributing lender was already exposed to the borrowingcountry: that is, Dj > 0 for each lender, j .

(iii) Expected default rates would decline rather than rise: that is, d* < drather than d* > d.

By 1985, it was apparent that none of these assumptions held. The informalcartel of large creditor banks was beginning to break down: it could nolonger enforce involuntary 'new money calls' on the plethora of smallerbanks that had been part of the original syndicates. Moreover, manyregional US banks, as well as European banks with smaller original debtexposures, had by then either eliminated their exposures entirely viasecondary market sales, or raised sufficient new capital that defensivelending to forestall default was no longer a priority. Finally, and most funda-mentally, it was now evident that default rates were rising, not falling. Thestrategy of liquidity relief seemed not to have been sufficient. Nevertheless,at the IMF/World Bank meetings in October 1985, liquidity relief was givenofficial status with the announcement of the 'Baker Plan,' under which theUS Treasury undertook to breathe new life into concerted new lending bycoordinating multilateral, bilateral and commercial bank contributions.

At the same time, academic contributions, in particular by Kenen (1983),Sachs (1986), and Krugman (1988, 1989), began to argue for a new strat-egy of debt relief. The rationale for this strategy rested on the assumption thatexpected debt payments could be maximized by supplementing temporarynew lending flows with permanent write-offs of old debt stocks. This ration-ale was captured in the so-called debt Laffer curve (DLC) (Krugman 1989),

271

Dow

nloa

ded

by [

Tuf

ts U

nive

rsity

] at

13:

55 1

4 N

ovem

ber

2014

JAMES W. DEAN

which depicts die relationship between a country's stock of debt obli-gations, D, on the horizontal axis, and the expected present value of actualpayments on those obligations, V, on the vertical axis. Theoretically, acountry might accumulate so much debt that V begins to decline as it takeson further obligations.7 Such a curve is depicted in Figure 1. As debt accu-mulates to say Dj, beyond the maximum point A at Do, the expected valueof creditors claims, V, declines from A to B. Conversely, a debt write-off fromDl to Do would make creditors better off by increasing the expected valueof their claims from B to A. However, as shown by Froot (1989), the optimaldebt write-off would not be from Z)j to Do but rather from Z)j to D2, sincesimultaneous liquidity relief could shift the DLC upward to DLC'. Thisanalysis set the stage for the official Brady Plan that orchestrated simul-taneous debt and liquidity relief throughout the first half of the 1990s.8

In the late 1980s, creditor banks were slow to accept this rationale fordebt forgiveness, partly because they were well aware of a free-rider barrieranalogous to that facing liquidity relief. Mutually beneficial debt reliefcould only occur if coordinated across all banks. The IMF, World Bank, andodier multilateral financial institutions were also reluctant to accept theprinciple of debt relief, in their case because of reluctance to abandon theirstatus as senior and sacrosanct claimants. However in 1989, after theannouncement of the US Treasury's Brady Plan, both the multilaterals andthe banks were induced to sign on, the former by a release from responsi-bility for debt relief in return for a disproportionate role in liquidity reliefsand the latter via 'menus' of choices between debt and liquidity relief thatpromised to enhance, or at least preserve, their net worth. By 1995, nearly$70 billion (or one-third) of SIMIC commercial bank debt had been writtenoff. The Brady Plan is widely credited as the trigger that released theexplosion of voluntary private lending to developing countries in the 1990s,suggesting that the sovereign debt crisis of the 1980s had in some sense pro-gressed beyond illiquidity to insolvency.

If, in any sense, the sovereign debt crisis of die 1980s was ultimately oneof insolvency, it was because it became a fiscal crisis. Indeed, as Wiesner(1985) pointed out, between 1979 and 1982, three of the four largest

v-

0Do D2 D, D

Figure 1 Debt Relief and the Debt Laffer Curve

272

Dow

nloa

ded

by [

Tuf

ts U

nive

rsity

] at

13:

55 1

4 N

ovem

ber

2014

ASIA'S FINANCIAL CRISIS

debtors - Argentina, Brazil, and Mexico - had more than doubled theirfiscal deficits, from about 6 to 15 per cent of GDP. Further fiscal pressureensued as many countries assumed sovereign responsibility for what hadpreviously been private-sector debt obligations. For example, as of 1985,private debt accounted for (an unweighted average of) only 17 per cent oftie long-term external debt of seven major Latin American and Asiand ebtors: Argentina, Brazil, Mexico, Venezuela, Indonesia, South Korea, andthe Philippines.9 Governments' ability to service external debt was subjectto the following budget constraint:

(3) G+aNFT= AM+AD + AD*+ T

where Gis government spending net of NFT, NFT is net financial transfersto foreign creditors, a the fraction of NFT for which the government hasassumed responsibility, AM is money creation to finance fiscal deficits, ADand AD* are increases in domestic and foreign debt, and Tis tax revenue.The difficulties associated with meeting this budget constraint came to becalled the 'internal transfer problem.'

Countries that adjusted successfully to the need to finance foreign debtcut of fiscal revenue were those that reduced government spending, G.From 1981-84, G as a fraction of GDP fell by about one-fifth in Indonesiaand South Korea, by nearly one-third in Mexico and the Philippines, andby nearly one-half in Venezuela. But a dilemma that complicated fiscaladjustment in the context of substantial state enterprise was that suchspending cuts typically focused on government investment - investmentthat was not readily replaced by domestic or foreign private investment. Forexample, in Venezuela, investment in the state oil firms collapsed.

Very little fiscal adjustment was accomplished by raising taxes. Instead,on the revenue side debtor governments either incurred more domesticand foreign debt, or indulged in inflationary finance by printing moremoney. A vicious circle was set up by successive currency devaluations aimedat increasing export surpluses; these devaluations had the unfortunate sideeffect of increasing domestic-currency debt obligations, leading to the needfor further inflationary finance (Dornbusch 1988). The essence of theinternal transfer argument was that the difficulties and dilemmas facingfiscal adjustment seriously delayed the return of debtor countries to credit-worthiness in the mid-1980s, despite economic recovery in the developedworld that might otherwise have propelled export-led growth.

The Brady Plan's debt reduction helped to break this vicious fiscal andcurrency cycle. But the direct fiscal impact of debt reduction should not beexaggerated. Even complete elimination of long-term debt to foreign bankswould typically have reduced claims on government revenue by only 6 to 8per cent, and the Brady reduction was only of the order of one-third of suchdebt. The more important impact of Brady debt relief was on countries'

273

Dow

nloa

ded

by [

Tuf

ts U

nive

rsity

] at

13:

55 1

4 N

ovem

ber

2014

JAMES W. DEAN

willingness to pay rather than their fiscal ability to pay. Cline (1995: 250-2)terms this 'catalytic debt relief,' emphasizing the interaction of economic,psychological, and political factors. Hence, it can be argued that sovereigndebtors were insolvent rather than illiquid in terms of practical politicalreality, even though they were undoubtedly solvent in terms of their strictlyeconomic ability to pay.

3. ILLIQUIDITY OR INSOLVENCY IN THE 1990s?

The first major post-Brady international payments problem was Mexico's,in 1994-95. In retrospect, it was clearly a crisis of illiquidity rather thaninsolvency. Mexico's payments problems were righted with the help of anunprecedented $40 billion line of credit from the US Treasury and the IMFthat was only partially drawn upon, and was repaid, ahead of time and atmarket rates of interest, within two years. Will the verdict on the presentAsian crisis be so straightforward?

To answer this question it is instructive to compare broad features of thecurrent crisis with those of the 1980s. The 1980s crisis was triggered by ahike in international interest rates that virtually doubled debt service pay-ments. This was compounded by a decline in developing-country exportrevenues due to recession in the major developed economies. No suchexternal events triggered the Asian crisis of 1997. It is true that the pastyears' rise in the US dollar against the Japanese yen and the German markhurt exports of Asian countries whose currencies were pegged to the dollar.But the impact of this event on current accounts was nowhere near as dra-matic as the sharp hike in international interest rates from 1980-82.

The two crises were, however, similar in the way they spread beyond theirorigins. The 1980s crisis spread rapidly from one country (Mexico) to mostof Latin America and beyond because new lending to other heavilyindebted developing countries dried up almost overnight. Similarly, thecrisis of 1997 spread rapidly from Thailand to half a dozen other Asiancountries because of contagious cessation of capital inflows.

The 1980s crisis was gready magnified in many but not all affected coun-tries by massive domestic capital flight, both before and after the onset of thecrisis itself. This was not a problem in Asia during the 1980s (except in thePhilippines). Nor does it appear to have been a problem leading up to the1997 crisis, although the crisis itself has been marked by substantial shiftsof domestic funds into Asian-based branches of foreign banks, includingshifts into foreign-currency deposits.

The 1980s crisis was curtailed initially via a strategy of concerted liquidityrelief, involving coordinated new lending, as well as payments reschedulingby commercial banks and by bilateral and multilateral lenders. The banksin particular perceived this strategy to be in their short-term interest sinceit would 'defend' the long-term prospects of receiving repayment of their

274

Dow

nloa

ded

by [

Tuf

ts U

nive

rsity

] at

13:

55 1

4 N

ovem

ber

2014

ASIA'S FINANCIAL CRISIS

outstanding debt claims. In due course, this strategy became both less wiseand less workable, as debt claims accumulated and arrears mounted: inshort, as illiquidity slid toward insolvency. The rationale for supplementing1 iquidity relief with debt reliefbecame correspondingly more compelling.

Thus far in the Asian crisis, liquidity relief from private lenders has provedmore problematic than it was in the early stages of the crisis of the 1980s.Borrowers are no longer sovereign states as they were in the 1980s; they arefrom the private sector. They are also diverse. Thus lenders cannot so readilynegotiate with a single sovereign borrower. Moreover, capital inflows duringthe 1990s have been dominated by direct investment and portfolio flows,rather than by commercial bank lending. Because portfolio debt and equityi:> widely held, rather than concentrated as unsecuritized loans in the handsof a few large banks, incentives to defend outstanding claims are both widelydisbursed and less imperative from the standpoint of individual claimants(Dean 1995). And finally, when portfolio capital is short term, as much of iti:>, the incentives for what might be called 'reverse free-rider' behavior arehigh: if an individual investor sees other investors pulling out, that alone isan incentive for him to pull out, irrespective of the merits of the investmentitself. As George Soros put it recently, 'We are dealing with a self-reinforc-ing process' (Soros 1997).10 The free-rider barrier to reinvesting short-termportfolio capital is very difficult to overcome. Perceived shortfalls becomeactual shortfalls, and perceptions of liquidity crises become self-fulfilling.

Probing beneath the liquidity squeeze, debtor countries both in the1980s and the 1990s indulged in macroeconomic mismanagement. Becausesovereign states were the primary obligators for foreign debt in the 1980s,raacroeconomic mismanagement took the form of inflationary finance andaccumulation of domestic debt under the pressure of a fiscal 'internal trans-fer problem.' Even microeconomic mismanagement was often a conse-quence of government policy, since state and parastatal enterprisesborrowed for their own accounts, and bore responsibility for misallocatedcredit and squandered funds.

Macroeconomic mismanagement in the 1990s assumed different forms.Excessive fiscal deficits and inflationary finance were by and large phenom-ena of the past. Rather, mismanagement in many developing economies(although not in South Korea!) took the form of prolonged commitment toovervalued, fixed exchange rate regimes. This in turn generated excessivecurrent account deficits, financed by unsustainable capital account sur-pluses. When the latter put pressure on domestic money supplies and causedinflationary pressure, borrowing countries often attempted to sterilize theireffects. This in turn tended to raise domestic interest rates and attract furthercapital, as well as incur quasi-fiscal costs (Dean 1996). As inflation proceededat rates above their trading partners', borrowing countries' real exchangerates became increasing overvalued. A corollary was that real interest ratesturned negative, at least for those willing to borrow in foreign currency, on

275

Dow

nloa

ded

by [

Tuf

ts U

nive

rsity

] at

13:

55 1

4 N

ovem

ber

2014

JAMES W. DEAN

the assumption that nominal exchange rates would not be devalued. In short,both Mexico in 1994 and a range of Asian countries in 1997 - notably Thai-land, Indonesia, and Malaysia — were poised for exchange rate crises.

Microeconomic mismanagement in the 1990s took the form of seriouscredit misallocation by domestic banks and other private financial insti-tutions. This misallocation occurred either because of benign neglect bygovernment (inadequate supervision and regulation), as in Thailand, acombination of neglect and self-interested involvement by government (asin Indonesia), or direct but misguided involvement by both governmentand industry (as in Korea). In short, Asia/Pacific countries in 1997 werevulnerable to banking and financial crises. But contagious exchange rate criseswere the triggers.

Differences from the 1980s help to explain why the liquidity crisis thistime round has been more severe and prolonged. But they hide a paradox.The most fundamental difference between the Asian economies now andLatin America's then is that Asian economies are stronger on the supplyside. They have high savings rates, a strong work ethic, a highly educatedlabor force, and high productivity built on non-resource-based activity.Although it is true that they have overinvested and misallocated capital, thisdoes not spell 'insolvency' in the long run provided they submit themselvesto medium-term structural adjustment. Finally, the demand side of Asianeconomies - except for excessive current account deficits in Thailand,Malaysia, and Indonesia - has been well managed, with low inflation andsmall fiscal deficits.

In short, Asia today differs from Latin America fifteen years ago in thatit is simultaneously more prone to illiquidity and less to insolvency. Not thatLatin America was ever 'insolvent' in a strict sense; but for practical pur-poses it became sufficiently insolvent fiscally that debt relief was warranted.It is inconceivable that Asia will ever reach that point: inconceivable, thatis, unless we the creditors so mismanage our liquidity relief that we precip-itate insolvency unnecessarily. The potential for such mismanagement wasevident during the final weeks of 1997, as large banks met in emergency ses-sions to negotiate debt relief for Korea.

4. LIQUIDITY MANAGEMENT BY SOUTH KOREA'SCREDITORS

Despite a shift toward portfolio finance in the 1990s, large banks do holdsubstantial claims on Asia, and on Korea in particular. It is between themthat a coalition must emerge if liquidity relief from private lenders is to beforthcoming. Japanese banks, for example, hold almost $24 billion in claimson South Korea. But in practice, with Japanese banks short of capital andridden with non-performing domestic loans, liquidity relief has been hardto muster. Indeed, during the last few weeks of 1997, commitments by the

276

Dow

nloa

ded

by [

Tuf

ts U

nive

rsity

] at

13:

55 1

4 N

ovem

ber

2014

ASIA'S FINANCIAL CRISIS

Japanese banks simply to rollover existing short-term loans were accepted byKorea and other borrowers with sighs of relief. Negotiation of new or longer-term commitments, or of rescheduling, was painful. As a result, Korea hadto rely on unprecedentedly large and quick disbursements from the IMF.

However, on December 25, in an apparent effort to induce belated par-ticipation by commercial banks, the IMF, the US, Japan, and nine othercountries announced plans to accelerate disbursement of $10 billion in newloans to Korea. Within a few days this strategy was promising to pay divi-dends. On December 29, executives from thirteen large US, Canadian,Japanese, and European commercial and investment banks met at the NewYork Federal Reserve to discuss rollovers, rescheduling, and new loans. Theimmediate objective was to inject sufficient short-term liquidity to enableSouth Korea to meet about $15 billion in loan repayments over the comingmonth. This objective was in fact met within days of the December 29meeting, when the major commercial bank creditors, led by Citibank,agreed to rollover their claims for another thirty days.

The next objective was to restructure South Korea's debt on a muchlonger-term basis, and preferably attract new loans as well. By the first weekof January 1998, the banks' initial reluctance had rather ironically meta-morphosed into moderate competition between commercial and invest-ment banks for pieces of the bailout pie. WithJ. P. Morgan in the forefrontand Goldman, Sachs and Salomon Smith Barney close behind, proposalsbegan to emerge for securitizing existing commercial bank debt as longer-term, tradable bonds, and for raising new money by issuing new bonds. Notincidentally, these proposals promised to earn their investment bank advo-cates tens of millions in fees, as well as relieve the commercial banks of theirexposures. The banks also called for Korean government guarantees onsuch bonds, thereby virtually guaranteeing that they would profit from theirill-considered past lending. Despite their obvious potential for moralhazard, such guarantees may be necessary (and are apparently favored bythe IMF), since the bulk of Korea's borrowers were banks. Failures of theseKorean banks would likely be multiplied into failures of chaebols (conglom-erates) as well as other, smaller firms in the real sector, on a magnitude tooterrible to contemplate. In short, the strategy as of early 1998 was to convertshort-term commercial bank loans to Korean banks into long-term bondsguaranteed by the Korean government.

For its part, the Korean government was not about to provide guaranteeswithout hard bargaining. Accordingly, by early January of 1998, the presi-dent-elect, Kim Dae Jung ('D.J.') had dispatched a negotiating team to NewYork and Washington. The Koreans stressed that the debt-restructuringpackage must include new syndicated commercial bank loans to the govern-ment, in order to replenish foreign exchange reserves. In a somewhatbizarre twist, it emerged that in late December the South Korean govern-ment had retained E. Gerald Corrigan, a former president of the New York

277

Dow

nloa

ded

by [

Tuf

ts U

nive

rsity

] at

13:

55 1

4 N

ovem

ber

2014

JAMES W. DEAN

Federal Reserve and presently a partner at Goldman, Sachs. Mr Corriganwas retained as an 'independent advisor': so independent, in fact, that hewas bound by the terms of his contract not to communicate his dealings toGoldman, Sachs. ' "If you open up someone's eyes to the possibility ofrestructuring their debt, there's no reason for them to think they can't doit themselves," one American banker said yesterday, speaking on conditionof anonymity.'11

5. CONCLUSION

In summary, the crisis in Asia was precipitated by currency attacks on over-valued exchange rates, but fundamentally rooted in massive misallocationof both domestic savings and borrowed foreign funds by domestic bankingsystems. It is important to distinguish conceptually between exchange ratecrises and credit misallocation, because the two have become intertwined asbanking and financial crises. Herein also lies the importance of distinguish-ing between illiquidity and insolvency. Exchange rate collapses led to severeliquidity squeezes on both the financial and the real sectors. These wereexacerbated by the free-rider barrier to resumption of portfolio capitalinflows from abroad, and the reluctance until several months into the crisisof major foreign banks to mobilize rollovers, rescheduling, and newlending. The fact that it took several months to negotiate private re-lendinghas confronted Asian economies with the clear and present danger that alargely nominal exchange rate and financial crisis will deepen into a creditcrisis with severe real consequences. In short, Asian economies, or at leastAsian firms, are in danger of being pushed from illiquidity into unwar-ranted insolvency.

Because illiquidity has been so extensive and life-threatening, officiallenders - the IMF, World Bank, and bilateral lenders - have had to play anunprecedentedly large role. This role has come under attack from threequarters: from fiscal conservatives (for example, elements in the US Con-gress) who fear the cost, from left-wing moralists who inveigh against bailingout big banks, and from right-wing academics who fear moral hazard. Allthree are misguided. Fiscal conservatives fail to understand that the IMF'sso-called 'bailout' is nothing of the sort. The IMF package is a commitmentto disburse a series of loans on an as-needed basis at market rates of inter-est, loans that will almost certainly be repaid and repaid on time: unless,paradoxically, the opponents of such loans prevail. Left-wing moralists havea point, but at this delicate juncture it must be tempered by pragmatism.Moreover, the South Korean negotiating team is tough and well-advised,and can be counted upon not to concede more to the banks than is inKorea's own self-interest. Right-wing academics also have a point, but onceagain, pragmatism must prevail, at least for the short term. The challengenow is to forestall further real damage, while attaching strong enough

278

Dow

nloa

ded

by [

Tuf

ts U

nive

rsity

] at

13:

55 1

4 N

ovem

ber

2014

ASIA'S FINANCIAL CRISIS

conditions to the loans to ensure that past mistakes are both rectified andrendered unlikely to recur.

The challenge for official lenders is to distinguish between banks andfirms that are genuinely insolvent, and those that are simply temporarilyshort of funds. This challenge is compounded by the fact that both banksand firms, especially in South Korea, are seriously overgeared, withdebt-equity ratios far in excess of international norms. The immediate chal-lenge for the IMF in particular is to ensure that liquidity relief is allocatedexclusively toward banks and firms with the potential for long-run solvency,while maintaining pressure to restructure, consolidate, or close the rest. Afurther challenge is to make official funding conditional on rupturing cozyrelationships between government and industry, and as well on radical andpermanent reforms of bank regulation and supervision. In this task the IMFfaces further dilemmas given the political resistance such reforms willencounter in Asia, and given also that withholding funds profoundly endan-gers the international financial system.

To some extent these dilemmas will be eased if foreign direct investorsare willing to inject equity into failing enterprises. But such investment, par-ticularly in South Korea, must be preceded by rapid liberalization of restric-tions on foreign ownership. For example, despite recent easing ofrestrictions on inward foreign direct investment, South Korea as of late 1997still prohibited foreign direct investment by means of mergers and hostileacquisitions, although this prohibition was dropped in early 1998. Needlessto say, domestic restructuring and foreign control will encounter strongpopular and political opposition.12

South Korea as a special case

Although all four of the failing Asian economies are tarred by the samebrush of overinvestment and capital misallocation, it is important to empha-size that in several respects South Korea is less culpable. Although it over-borrowed and overlent on a massive scale, South Korea is not, asPresident-elect Kim inappropriately remarked early in the crisis, 'bank-:rupt.' Nor was it illiquid on its own account. It was illiquid simply becauseof a financial panic that spread from other Asian countries and was com-pounded by a 'reverse free-rider* phenomenon. South Korea's externaldebt of about $160 billion is not high, either relative to the country's GDP,or relative to a decade ago.13 However, the fraction of external debt thatwas short term, coming due within a year or less, was dangerously high -about two-thirds. Until early 1998, when lenders belatedly agreed to rolloverdebt until March, roughly $15 billion was maturing each month.

What occurred during the last two months of 1997 can be termed areverse free-rider' phenomenon. Prompted by the problems in other Asian

countries, speculators, short-term lenders and portfolio investors began to

279

Dow

nloa

ded

by [

Tuf

ts U

nive

rsity

] at

13:

55 1

4 N

ovem

ber

2014

JAMES W. DEAN

bet against the Korean won, exhausting much of Korea's foreign exchangereserves in the process. Short-term lenders then noted that at least $100billion in external debt was due to mature within one year, and that the IMFpackage of $57 billion, together with official reserves of less than $10billion, fell far short of closing the gap. It was quite rational for individualshort-term lenders not to renew, despite the fact that collectively it was verymuch in their interests to do so, since that would close the liquidity gap andend the immediate crisis. As in the crisis of the 1980s, concerted liquidityrelief from private lenders was crucial to preventing widespread insolvency.The hole in the dike had to be plugged before the whole dike collapsed.14

For the past decade or more, South Korea's macroeconomic managementhas been fundamentally sound, with low inflation, low fiscal deficits, anda relatively low current account deficit (2-3 per cent of GDP): in fact, atrade account surplus by late 1997. Moreover, unlike in Thailand,Malaysia, and Indonesia, the real exchange rate was not overvalued. Beforethe won collapsed, Korea's real effective exchange rate was some 5 percent lower than in the early 1990s, and the real rate against the US dollarwas 25 per cent lower! The IMF must be cautioned against demandingunrealistically low inflation targets for 1998, since that would imply a'brutal monetary squeeze' (Sachs 1997) in the context of the won's recentdepreciation by more than 50 per cent. This caution is especially acute inthe context of the extraordinarily high debt-equity ratios prevalentamong Korean firms.

Despite the spurious nature of South Korea's present financial panic, andthe sound nature of the country's macroeconomics, there is no denyingserious microeconomic problems stemming from massive, bank- and govern-ment-abetted, misallocation of capital, and a correspondingly irrationalindustrial structure, with overdiversified chaebols at the core. This is thelegacy of an underregulated and overcomforted banking system, a systemthat indulged in 'Panglossian' lending practices, that was reinforced by easymoney from abroad, and that must, and now will, be reformed.15 But thosewho bemoan such practices in the past are quite wrong to argue that theIMF and the world's major banks should now pull the plug on South Korea.This is not the time to agonize about moral hazard.

Simon Fraser University and Western Washington University

Received: May 1998; Revised: July 1998

ACKNOWLEDGMENTS

An earlier version of this paper was presented at the American EconomicAssociation annual meeting, Chicago, Illinois, January 3-5,1998. A shorterand less formal version was published as Dean (1998).

280

Dow

nloa

ded

by [

Tuf

ts U

nive

rsity

] at

13:

55 1

4 N

ovem

ber

2014

ASIA'S FINANCIAL CRISIS

NOTES

1 For convenience, I use the titles 'Korea' and 'South Korea' interchangeably.2 Indeed the IMF has already admitted to excess harshness in at least one case. In

the fall of 1997, the IMF forced closure of twelve financial institutions inIndonesia. This led to a run on banks that immediately worsened the crisis. Inearly January 1998, an internal IMF document admitting error became public.

3 I use the word 'bailout' advisedly. The $24 billion, $32 billion, and $57 billionpackages put together for Thailand, Indonesia, and Korea, respectively, arecommitments to be drawn upon only as needed to meet sovereign obligations,notably to supplement foreign exchange reserves depleted in defending plum-meting exchange rates (although as the crisis deepens it seems increasinglylikely that the full commitments will be drawn upon). They are also market-rateloans, not grants, and are thus bailouts of the debtor countries only if and whenthey are not repaid (and historically the IMF has almost always been repaid).However, they are indirect 'bailouts' of foreign banks and other private creditors.

4 As quoted in the International Herald Tribune, January 16, 1998, p. 11.5 For extensive reviews of the 1980s international debt crisis, see Bowe and Dean

(1990, 1997b), Cline (1995), and Dean (1992).6 Expected default rate in this context means that fraction of contractual debt

obligations, D, which is not expected to be paid.7 Such a decline in the DLC was called into question after empirical work by

Claessens (1988) and Dean and Xu (1991, 1993) found that for all but half adozen severely indebted countries, the curve was more or less flat but notdownward sloping at their current levels of outstanding debt. Nevertheless, thefact that the DLC pointed to very low marginal costs to creditors of providingdebt forgiveness proved highly influential in formulating the strategy for debtrelief that culminated in the Brady Plan of 1989. Moreover, analyses of the'catalytic' effects of debt reduction on V, operating via 'willingness to pay' ratherthan 'ability to pay' (see Cline 1995: 250-2; and the last paragraph of Section 2of this paper) to some extent rehabilitated the prognosis of the DLC after thefact.

8 For explanations and analysis of the origins, workings, and effectiveness of theBrady Plan see Dean (1992) and Bowe and Dean (1997b).

9 In the case of Korea, however, private-sector obligations in 1985 were 44 per centof long-term external debt, substantially higher than the unweighted average of17 per cent for the seven.

10 In the same Financial Times editorial, Soros calls for the formation of an 'Inter-national Credit Insurance Corporation' to supervise and regulate the allocationof international capital. While not an entirely new suggestion (see Cline'sadvocacy of an 'International Bondholders Insurance Corporation' (1995:482-93)), Soros's proposal may well fall on receptive ears, given its timeliness inconjunction with his power and prestige.

11 As quoted in the New York Times, January 5, 1998, p. 1.12 Debt restructuring can be explicitly linked with foreign direct investment via

debt-equity swaps; this mechanism flourished and helped ameliorate theimmediate problems of at least a dozen troubled sovereign debtors during themid-1980s (Bowe and Dean 1993, 1997a, 1997b).

13 See Appendix below.14 I am indebted to my Western Washington University colleague, Dennis R.

Murphy, for the metaphor of the dike.15 Frequent banking and financial crises in countries that have recently liberalized

their internal and external controls on interest rates and capital flows have led

281

Dow

nloa

ded

by [

Tuf

ts U

nive

rsity

] at

13:

55 1

4 N

ovem

ber

2014

JAMES W. DEAN

some to question the appropriate sequencing of financial liberalization: forexample, whether completely unrestricted capital inflows and outflows ought tobe introduced before proper bank supervision and regulation is in place (see,for example, Dean 1997).

16 This Appendix draws on World Bank (1990-97).

APPENDIX: H I S T O R I C A L PERSPECTIVE ON KOREANFOREIGN DEBT 1 6

South Korea demands special attention because it is the third largest economy inAsia, and was the eleventh largest in the world until the won's recent sharp, 50 percent depreciation against the US dollar. The crisis in Korea was perhaps the leastexpected among Asian economies: for example the World Economic Forum's lastreport, in June 1997, rated South Korea the fifth-best place in the world to invest.

By contrast between 1960 and the mid-1980s, few would have rated Korea high ona global scale of creditworthiness. In fact, over the past thirty-five years, Korea wasunique among the world's heavily indebted countries (both developed and develop-ing) in reaching a current-account-to-exports ratio exceeding 30 per cent: the ratioreached a phenomenal 50 per cent in 1968. In 1982, at the outset of the last globaldebt crisis, South Korea's debt was the fourth largest among heavily developing coun-tries, after Brazil, Mexico, and Argentina, and just ahead of Venezuela. Korea's totalexternal debt was then $37.3 billion, $24.9 billion of it long term, with $11.3 billiondue to commercial banks. Debt peaked in 1985 at $47.1 billion.

However, throughout the 1980s Korea engaged in an exemplary program ofexport promotion and debt repayment, realizing a current account surplus of $14.2billion by 1988. Unlike most Latin American debtors, Korea's export growth wasbased on manufactures rather than commodities; hence it failed to benefit from thecommodity price boom of the early 1980s, but was not hit by the downturn of com-modity prices in the latter part of the decade. Indeed, Korea, as an oil-importer, instriking contrast to Venezuela and Mexico, both heavily dependent on oil exports,was blessed rather than cursed by the oil price collapse of 1986. In short, Korea,unlike most other major debtors, benefited from both good policy and good luck.By 1989, total external debt had been reduced to $37.8 billion. Although sometwenty countries sought and received debt reduction between 1989 and 1995 underthe Brady Plan, Korea was not among them.

REFERENCESt

Bowe, Michael and Dean, James W. (1990) 'Voluntary debt relief and the Philip-pines,' ASEAN Economic Bulletin 7 (July): 11-38.

Bowe, Michael and Dean, James W. (1993) 'Debt-equity swaps: investment incentiveeffects and secondary market prices,' Oxford Economic Papers 45 (January): 30-47.

Bowe, Michael and Dean, James W. (1997a) 'Debt-equity swaps and the enforce-ment of sovereign loan contracts,' Journal ofInternational Development 9 (January):59-83.

Bowe, Michael and Dean, James W. (1997b) Has the Market Solved the Sovereign-DebtDebt Crisis'? Princeton Studies in International Finance, No. 83, Princeton, NJ:Princeton University, August, 66 pp.

Claessens, Stijn (1988) 'The debt Laffer curve: some estimates,' Working Paper,World Bank.

Cline, William R. (1983) International Debt and theStability of the World Economy, Policy

282

Dow

nloa

ded

by [

Tuf

ts U

nive

rsity

] at

13:

55 1

4 N

ovem

ber

2014

ASIA'S FINANCIAL CRISIS

Analyses in International Economics 4. Washington, DC: Institute for Inter-national Economics.

Cline, William R. (1984) International Debt: Systemic Risk and Policy Response, Washing-ton, DC: Institute for International Economics.

Cline, William R. (1995) International Debt Reexamined, Washington, DC: Institute forInternational Economics.

Dean, James W. (1992) 'The debt confessional,' World Competition 15 (March):99-113.

Dean, James W. (1995) 'Why are we lending to Latin America once again?' KaiserChair Inaugural Address, Western Washington University. Reprinted in Bank ofValletta Review, Spring.

Dean, James W. (1996) 'Recent capital flows to Asia Pacific countries,' Journal of theAsia Pacific Economy 1 (3): 287-317.

Dean, James W. (1997) 'Can financial liberalization be pre-mature?' Paperpresented at the second annual conference of the Latin American and CaribbeanEconomists' Association, Bogota, Colombia, October.

Dean, James W. (1998) 'Why Fiscal Conservatives, Left Wing Moralists and RightWing Academics Are All Wrong About Asia,' Challenge, Vol. 41, No. 2,March/April.

Dean, James W. and Aihua Xu (1991, 1993) 'Debt relief for Eastern Europe: whyPoland is a special case,' Rivista di Politico Economica 81 (June): 1991, 189-230.Reprinted in Mario Baldassarri and Robert Mundell (eds) Building the New Europe,Vol. 2: Eastern Europe's Transition to a Market Economy, New York: St Martin's;London: Macmillan, 1993.

Dornbusch, Rudiger (1988) 'Our LCD debts,' in Martin Feldstein (ed.) The UnitedStates in the World Economy, Chicago: University of Chicago Press.

Froot, Kenneth A. (1989) 'Buybacks, exit bonds, and the optimality of debt andliquidity relief,' International Economic Review 30 (February): 49-70.

'.'ienen, Peter (1983) 'A Bailout for the Banks,' New York Times, March 6."<rugman, Paul (1988) 'Financing vs. Forgiving a Debt Overhang,' Journal of

Development Economics, 29, Nov., 253-68.I-Crugman, Paul (1989) 'Market-Based Debt-Reduction Schemes,' in Jacob Frenkel et

al, eds., Analytic Issues in Debt, IMF, Washington, D.C., 258-78.jirugman, Paul (1998) 'What happened to Asia?' Available at website h t tp : / /

web.mit.edu/krugman/www/DISINTER.html.Sachs, Jeffrey (1986) 'Managing the LDC Debt Crisis,' Brookings Papers on Economic

Activity, No. 2, 397-440.•Sachs, Jeffrey (1997) 'IMF is a power unto itself,' Financial Times, December 11.Soros, George (1997) 'Avoiding a breakdown,' Financial Times, December 31.Wiesner, Eduardo (1985) 'Latin American Debt: Lessons and Pending Issues,'

American Economic Review 75, no. 2 (May): 191-95.World Bank (1990-97) World Debt Tables, Washington, DC, various issues.

283

Dow

nloa

ded

by [

Tuf

ts U

nive

rsity

] at

13:

55 1

4 N

ovem

ber

2014