|BANK INDONESIA AND THE RECENT CRISIS
Oleh : J. Soedradjad Djiwandono
The Indonesian crisis is now more than two years old, yet there has not been a clear signal that the economy is truly heading toward recovery, let alone sustainable, balanced and broad-based development. Following the much better than expected general election of 7 June 1999, there were encouraging developments in the economy and in the social and political spheres as macro indicators improved. However, this did not last long enough to support a more sustained path toward recovery and growth. Expectations arising from both external developments and domestic events produced a recurrence of economic strains, resulting in further pressures on the rupiah and on equity prices. Of the three worst hit crisis countries in Asia, Indonesia’s recovery is by far the slowest. Thailand and especially Korea seem to have remained on a steady path of recovery. The latest developments confirm that Indonesia is the ‘worst case’ among the Asian crisis economies. More careful analysis is needed to explain the peculiar character of the Indonesian crisis.
This paper is a personal assessment of the Indonesian crisis from the vantage point of my position as Governor of Bank Indonesia during a critical part of the crisis period. I would like to offer my recollections of how the crisis developed from an external shock in the currency market, rapidly spreading wider and deeper in a complicated process of contagion to become a fully-fledged crisis, through the impact of shocks, policy responses and market reactions. The period of the crisis from its beginning in mid July 1997 to the time of my untimely departure from office in mid February 1998 may be the most confusing part of Indonesia’s history of economic policy management, and deserves to be documented for future scrutiny.
THE CRISIS AND ITS IMPLICATIONS
There have been many conferences and writings on the Asian crisis, discussing its nature and causes, its implications, and the lessons to be learned from it. One can be sure that such studies will continue in the years to come. This is because, despite the attention it has received, the crisis has persisted for longer than expected, and its effects have been more devastating than most people would be prepared to accept. The Asian crisis has been followed by the Russian and Brazilian crises. Recent developments seem to suggest that the crisis is over, or close to over, and most people agree that it is generally behind us. For example, at the Interim Committee Meeting of the International Monetary Fund (IMF) in the northern spring of 1999 it was reported that, of the three hardest hit countries in Asia, Korea was experiencing a recovery, the Thai crisis was bottoming out, and Indonesia was following suit. Brazil was considered to be out of the crisis, while Russia was close to having a new IMF-supported program. The ‘Survey of Recent Developments’ in the August 1999 issue of BIES (Pardede 1999) presented a similar view of Indonesia’s progress.
But more recent developments in Indonesia, and to a certain degree also in Thailand, have raised new concerns (Krugman 1999; Arnold 1999). The still fragile recovery in some crisis countries has been challenged by new problems arising from the expectation and ultimate implementation of an interest rate hike in the US, and from the new fear of a yuan devaluation. In Indonesia, the good news about a much better than expected general election was tainted by the Bank Bali scandal and the atrocities in East Timor, which immediately put pressure on the rupiah and on equity prices.1
Views on the causes of the crisis are many. However, over time some consensus seems to be developing. Differences might be narrowed down to two perspectives, which could be distinguished much as we differentiate between the Classical and the Keynesian economists’ views on the workings of the market. With respect to the Asian crisis, we could distinguish between those who see its origin as domestic and those who view it as originating externally. The first group argues that the crisis arose from practices of crony capitalism and weak financial structures, together with inept macro policies; the second group sees it as triggered by a shift of sentiment in the financial market that caused a contagious process of financial panic. The first view, put by people such as Professor Krugman of MIT, could be considered as a structural explanation of the crisis (Krugman 1998). The second view argues that the crisis was essentially a case of financial panic in the Keynesian tradition, as succinctly explained by Professor Kindleberger in his seminal work two decades ago (Kindleberger 1996). Professor Jeffrey Sachs has been the major proponent of this view (Radelet and Sachs 1998).
As one who had first-hand experience of the crucial part of the Indonesian crisis, I would argue that it was not determined by one single variable, whether external or domestic in origin. It is my view that the Asian crisis, and in particular the Indonesian crisis, arose from a combination of the workings of ‘contagion’ forces from outside the national economy on the one hand and weak domestic economic and financial structures on the other. The contagion factor emanated from a sudden change in market sentiment in the region that led to a shock in its currency markets. This resulted in panic selling of local currencies, and of other assets denominated in local currencies, for dollars. The rapid downgrading of the region’s sovereign credit ratings by international rating agencies further fuelled the shift in market sentiment, triggering panic selling of foreign-owned local assets. In the media, the term ‘Asian miracle’ disappeared suddenly, to be replaced by ‘Asian crisis’ or ‘Asian meltdown’. But the most telling sign was the Institute of International Finance’s publication on capital flows for Thailand, Malaysia, Indonesia, the Philippines and South Korea. The estimates showed a reversal in flows of capital of $105 billion in these five countries in a single year, from inflows of $93 billion in 1996 to outflows of $12 billion in 1997. For Indonesia alone, the reversed capital flow was estimated at $22 billion, from inflows of $10 billion to outflows of $12 billion. This was indeed a financial panic in the Keynesian tradition, as explained by Kindleberger (1996).
But the external shock itself need not have caused a crisis of the magnitude that these countries suffered, if only their domestic economic, social and political structures had been robust. Confronted with a contagious external shock the Indonesian economy, with its embedded inefficiencies and weak financial system, could not withstand its impact.2 The domino effect of the weakening rupiah adversely affected the financial sector, and then the real sector of the national economy. Thus, a combination of severe external shock triggered by changes in market sentiment, and financial-cum-real sector structural weaknesses, caused a contagious process that severely damaged the entire economy. The Indonesian crisis developed as a sequence of events which began with an external shock—part of a contagious financial panic in the region—that hit Jakarta’s financial market. The shock, the policy response and the market reaction exposed the fragility of the national banking system, leading to a banking crisis. In turn, the banking crisis seeped through the payment system, revealing weaknesses in the real sector of the national economy—which was basically a system of ‘ersatz capitalism’ in Yoshihara’s characterisation (Yoshihara 1988), with embedded inefficiencies and widespread corruption—and led to an economic crisis.3 The economic crisis exposed institutional weaknesses in Indonesia’s social and political system. In a complex and intertwined set of relationships, with feedback effects, the financial crisis turned rapidly into a multi-faceted crisis, crippling not just the national economy, but society and politics as well.
Many argue that the Asian crisis stands out as one of the major crises since the depression of the 1930s. Its impact has been so devastating that even pessimists acknowledge it has been worse than they expected. Indonesia, Thailand and South Korea have suffered more severely than other countries. And it is becoming apparent that, of the three, Indonesia has suffered the most in terms of the decline in economic growth, the depreciation of the currency, social dislocation and other problems. The virulent nature of the Indonesian crisis has been well documented. Its chief features may be summarised as follows.
POLICY RESPONSES AND MARKET REACTIONS
Faced with the currency shock in early July 1997, the government took a decision to widen Bank Indonesia’s intervention bands on 11 July. It was almost a routine exercise, since Bank Indonesia had done this a number of times before. This decision was lauded by many, including representatives of donor countries at the CGI (Consultative Group on Indonesia) meeting in Tokyo several days later. Some papers reported Bank Indonesia’s step as a ‘preemptive’ measure.
Market reaction was not as expected, however; in fact it was the reverse of previous experience. In the past, whenever Bank Indonesia. had widened the bands, the rupiah had appreciated, and the dollar spot exchange rate had closely followed Bank Indonesia’s buying rate. But this time the rupiah depreciated after the bands were widened. The dollar spot rate not only broke through the mid rate, it also broke through the selling rate or upper band. What happened in July 1997 was different from previous periods of pressure in the currency market. A contagion effect was in progress.7 Since investors were convinced, partly by rating agencies and financial analysts, of the presence of similar structural weaknesses within the economies in the region, their ‘herd instinct’ dictated that they should move their investments out of Asia.
When the spot rate broke BI’s selling rate the central bank intervened in the market, first by selling dollars forward, and later in the spot market. Market intervention from the third week of July until the day before the rupiah was floated in the spot market amounted to $1.5 billion.8 When these efforts did not stabilise the rupiah, and given that it was now the only currency in the region that was not freely floating, the government decided to free float the rupiah on 14 August 1997.
This was supported by monetary tightening through the raising of interest rates and sterilisation and by fiscal tightening. Bank Indonesia had already reduced bank liquidity in late July by ceasing to allow banks to rediscount their own papers (SBPUs). But a more stringent step was taken in mid August, when the central bank raised the rates for some BI certificates (SBIs) to at least double their previous level. For example, the one-week SBI rate was raised from 10.5% to 20%, and the three-month rate from 11.5% to 28%. On 29 August Bank Indonesia also issued a new rule limiting the forward sale of dollars to non-residents to $5 million, to reduce currency speculation.
On the fiscal side, the Minister of Finance cut government spending by rescheduling some projects and limiting routine expenditures on non-priority items. He also instructed a number of state enterprises and foundations to transfer their bank deposits of Rp 3.5 trillion to SBI holdings.9
But following the shock, the policy response and the market reaction, a new phenomenon developed. In reaction to the monetary and fiscal tightening, the weak banking sector began to suffer distress. Bank runs emerged as early as the second half of August 1997, when the process of ‘flight to safety’ began. And the interbank money market became compartmentalised, with ‘suspect’ banks having to pay much higher interest rates to borrow from other banks. Interbank rates increased dramatically, from an average of 22% to more than 80%. There was even an occasion when the overnight interbank rate reached 200% per annum. By the end of August 1997, more than 50 banks had failed to comply with the minimum reserve requirement of 5%.
Realising that the problem had spread to the banking sector, the government launched a broad economic policy initiative on 3 September 1997. The policy package encompassed not just monetary and fiscal measures, but also steps in the real sector, in the form of further trade liberalisation steps and the rescheduling of government projects with high import content. The government promised to continue adhering to an open foreign exchange policy. With respect to the banking sector, it announced steps that included a promise to keep helping solvent banks facing liquidity problems, a more active effort to encourage bank mergers, and a policy of closing insolvent banks, with a pledge to take care of small deposit holders’ interests with state banks.10 Despite all these efforts, not much progress was made in stabilising the currency or containing the banking problem. In the process, more banks became unable to comply with the minimum reserve requirement, and some even began to have a negative balance with Bank Indonesia. The economic team in the government was convinced that Indonesia was facing a confidence problem. A concerted effort had to be made to restore confidence. This led to the idea of seeking International Monetary Fund assistance to boost market confidence. The preliminary discussions on inviting an IMF team were held when Dr Stanley Fischer, First Deputy Managing Director of the Fund, visited Jakarta at the invitation of the Minister of Finance, on his way to the 1997 Annual Meetings of the World Bank and the IMF in Hong Kong. This happened some time in mid September 1997.
During the meeting with Dr Fischer, I proposed a precautionary arrangement with the Fund, instead of a fully-fledged standby arrangement. This idea was pursued further in Washington, but it was pushed aside during the follow-up discussions between the IMF team headed by the Fund’s Director for the Asia–Pacific, Hubert Neiss, and Minister Mar’ie Muhammad and myself, during the annual meetings in Hong Kong. Because of the rapid deterioration in Indonesia’s financial position, the discussion shifted rapidly to a standby arrangement. My original preference for a precautionary rather than a standby arrangement was based on a conviction that our economic circumstances were not so bad, at least in terms of the foreign exchange reserves that BI held at that time. I would still argue that this is true. Thailand and Korea were in a much worse position in terms of their reserve holdings at the time the Fund was invited in. Our problem then was to restore market confidence, and this objective could be well served by the presence of the IMF under a precautionary arrangement. But a more important reason was that I was afraid I would not be able to persuade the President to agree to the stringent conditionality of a standby arrangement. A precautionary arrangement would bear much less stringent conditions, even though it did not automatically include funds. Funding was not our major problem—at least that was how I looked at the circumstances then.11 The most crucial issue was the lack of market confidence in macroeconomic management, which would have been restored through IMF support of government policy. In the past, other countries like India and South Korea had been successful in addressing their problems through IMF precautionary arrangements.
The IMF-supported program for Indonesia summarised in the Memorandum on Economic and Financial Policies (MEFP) was submitted to the Fund via a letter of intent (LOI) on 31 October 1997. The program comprised a package of policies for economic reform in both the real and the financial sector, supported by prudent monetary and fiscal policy. The monetary and fiscal measures consisted of standard programs of macroeconomic management to stabilise the exchange rate and other monetary variables.
The core of the program was a policy package to deal with insolvent and weak banks and strengthen financial infrastructures, and to overcome structural rigidities in the real sector of the national economy. Thus, the framework was put in place for a comprehensive policy to restore confidence and arrest the decline of the rupiah. In essence, the program was built around three areas: 12
However, the domestic reaction to the closing of banks was the reverse of what was expected. It was ironic that a step designed to restore confidence in the banking sector instead resulted in a collapse of confidence, plunging the sector into chaos. It continued to suffer from a process of ‘flight to safety’, with bank runs becoming common. Many banks lost their deposit base, and the interbank money market became seriously compartmentalised. In addition, from January 1998 many banks found that trade and other financial lines from their bank business abroad were terminated. Letters of credit issued by many Indonesian banks were not accepted overseas. This plunged the banking sector from a state of distress into a state of crisis, with market confidence almost completely lost.13
After some ‘flip-flop’ implementation of the IMF-supported program, the confidence problem shifted from being just an economic problem to being one of national leadership. The negative reaction to the IMF-supported reform program became more pronounced when the loss of confidence caused by the bank closures was further complicated by a government announcement to reverse a decision to postpone several large government projects. At about the same time, monopoly practices and other inconsistencies reappeared in the implementation of the program for restructuring the real sector. In this way market confidence in the government’s commitment to the economic restructuring program evaporated. As a result, not only was the rupiah’s slide difficult to stop, but the economic crisis was spiralling into a ‘total crisis’.
SOME NOTES ON THE IMF-SUPPORTED PROGRAM
With the crisis now more than two years old, policy responses to the problems and market reactions, including feedback responses, can be better understood with the benefit of hindsight. Some comparisons can also be drawn with the experience of other countries in Asia. It should be noted here that, even though we tend to treat problems in the crisis countries as similar, they are not identical, since each crisis has its own peculiarities. Indonesia, despite its relatively better conditions and better early policy response, has ultimately become the worst case among the countries experiencing crisis, and the slowest in its path towards recovery. Among Asia’s crisis countries, Thailand, Indonesia and Korea all sought IMF support, while other countries in Asia did not. Malaysia faced the crisis on its own, including resorting to capital controls, even though before being sacked Minister of Finance Anwar Ibrahim was known to subscribe to stringent policies in the IMF tradition. Meanwhile, the Philippines had been on precautionary arrangements with the IMF for some time.
Thailand signed its first letter of intent on 2 July, Indonesia on 31 October, and Korea on 1 December, all in 1997. Each of these countries was granted standby support under the Emergency Financial Mechanism (EFM), introduced a year earlier to hasten the decision making processes of the IMF’s Board and the disbursement of loans. Korea was the first country to receive a standby loan based on the newly modified EFM with its faster disbursement mechanism. All these programs have two types of funding sources, namely the ‘first line of defence’, comprising IMF, World Bank and Asian Development Bank (ADB) loans, and the ‘second line of defence’, consisting of bilateral loans from countries attached to these programs.
Before seeking IMF support, all three countries had been dealing with strong pressure on their respective currencies by resorting to their own means of foreign exchange management. They began their defence against the currency onslaught by way of intervention in the foreign exchange market, in line with their adherence to a pegged system of foreign exchange management—a rigid pegged system in the case of Thailand and Korea, and a managed float in the case of Indonesia. After losing substantial reserves in market intervention, particularly in the first two countries, one by one the three abandoned their pegged systems, and moved to a floating foreign exchange system. But when market confidence was shaken, each sought IMF support in an effort to restore market confidence.
In general, an IMF-supported program contains adjustment policies that a country promises to adopt to deal with imbalances from a shock or crisis. Since the Asian crisis is multifaceted, reflecting a variety of problems and weaknesses, a sustainable program to address the problems of the crisis effectively must also comprise several aspects. There has been much criticism of the IMF’s handling of the Asian crisis. Its general thrust is that this crisis is not typical of the problems the IMF was set up to deal with. On the whole, the Asian crisis countries have not suffered from budget deficits, hyperinflation or even chronic balance of payments deficits. Yet the IMF therapy, at least the one usually emphasised, is a tight money policy with sky-high interest rates. There is some truth in this criticism, even if it is not entirely valid. Among the many factors causing or contributing to the crisis in Asia, the important common elements in the crisis-affected countries are a weak banking or financial system and unsustainable corporate short-term debt in foreign currencies. No stabilisation and recovery program will be effective in solving the crisis unless it seriously addresses these two problems.
There are slight variations in each country, but generally the IMF-supported programs adopted to face the crisis comprised financial reform and economic restructuring, complemented by prudent monetary and fiscal policy. The classic IMF approach is, of course, monetary and fiscal policy to address monetary instabilities, like inflation and balance of payments problems, and fiscal imbalances. But its capacity to deal with problems of economic structure, such as monopolies and oligopolies and the practices of crony capitalism, granted that these are real issues in the countries concerned, has raised controversy. Measures to address problems of bank restructuring and corporate debt have also drawn criticism. It is important to acknowledge that in all these cases the Fund has worked together with its sister institutions, the World Bank and the ADB, to mobilise a pool of expertise to deal with these complex problems. Since the fundamental problem facing these countries is one of market confidence, the presence of the multilateral agencies is crucial. They must show their support of these countries’ adjustment programs through their expertise, but also through the provision of significant funds that can be speedily drawn. Hence, the Emergency Financial Mechanism and the huge magnitude of the loans.14
Thailand’s standby loan from the Fund was $4 billion, which constituted 505% of its quota. Indonesia’s original loan was $10 billion, or 490% of its quota.15 In July 1998 an additional $1.5 billion loan was approved, making the total more than 500% of Indonesia’s quota. Korea’s loan of $21 billion amounted to 1,939% of its quota (Lane et al. 1999). In addition, each country also acquired loans from the World Bank and the ADB, which could be drawn relatively quickly. Likewise, each received some lines of credit from bilateral countries, which were linked to the IMF standby facility. In general, these loans could only be used when sources from the multilateral institutions had been exhausted, hence their status as a second line of defence.
The practice of maximising the apparent magnitude of loans in the context of an IMF standby arrangement, by including loans from different countries in the package as part of a second line of defence, has not been very meaningful in confidence building.16 The market is not easily impressed with these numbers, and most market players know how difficult it is for recipient countries to cash in the second line of defence facilities.17 On the other hand, since the media always used these higher numbers, in the case of Indonesia the well known phrase ‘the $43 billion IMF-bail out’ may have given the wrong impression to an uninformed public, at least about the size of the national debt.
In a way, since the original purpose of acquiring IMF support was to restore market confidence, the availability of lines of credit from bilateral donors may have provided creditors and investors of the recipient countries with some sense of security in their respective claims. These credit lines would certainly cushion the reserves held by the countries accepting IMF support. If creditors and investors were convinced, then this would reduce the demand for foreign exchange. It would also reduce the pressure for creditor countries to stop providing trade financing and money lines to the Indonesian banking community. And this could happen without the recipient countries drawing on the facilities provided in the scheme. However, if the program were not acceptable to creditors and investors, as was the Indonesian case, the pressure to withdraw existing loans would remain, and the demand for foreign exchange in the market would keep rising. Indonesia’s experience with the gimmick of describing the program as a huge ‘IMF bail-out’ was that it did not seem to contribute to the objective of restoring market confidence in the banking sector and in macroeconomic management.18
I should also mention that there was never a clear statement of the amount of ‘first line of defence’ assistance available to Indonesia when the IMF agreed to provide a loan in support of a standby arrangement based on the 31 October 1997 LOI. It was announced that the amount of the IMF loan was $10.5 billion. If this amount is added to the World Bank and ADB loans of $7.5 billion, the amount readily available becomes $18 billion. But the IMF announcement stated that the amount of available funds was $23 billion. The media have always given the total amount of the package as $43 billion, including $20 billion of loans coming from bilateral donors or the ‘second line of defence’.
If the funds readily available from the three multilateral institutions or first line of defence were $18 billion, then what was the $23 billion? In fact, the additional $5 billion was Indonesia’s own reserves. This amount came from the method of calculating reserves that Bank Indonesia used, which was different from the way the IMF calculated them. The Indonesian national reserves held by Bank Indonesia to January 1998 were calculated using the concept of ‘official reserves’, while after January 1998 Indonesia adopted the IMF’s concept of ‘gross reserves’. Indonesia’s official reserves were equal to gross reserves minus some illiquid reserves (like monetary gold) and those that were highly liquid and therefore very volatile. At the time of the negotiations with the IMF team in October 1997 the difference between the official reserves and the gross reserves was $5 billion. This was the amount that was added to the IMF, World Bank and ADB loans to form the first line of defence of $23 billion. Citing a large amount of funds as being available may be consistent with the aim of raising market confidence in the face of increasing demand for foreign exchange in the market. But this method of calculating the amount was certainly very confusing.
A note is in order here on the Indonesian government’s use of the concept of ‘official reserves’. This has been part of a long-time practice of erring on the conservative side in determining the international reserves that Bank Indonesia is entrusted by the government to hold. In other words, it was in the tradition of prudent reserve management. International reserve holdings are important not just economically in financing the current account deficit and the deficit on the capital account, but also psychologically, because exchange rate stability hinges on both the size and the relative stability of international reserves growth. Certainly, public confidence in a managed float would be disturbed by volatility in the international reserves held by the central bank. Based on this line of reasoning, the basic tenet of reserves management in Indonesia has been accumulation in a stable growth pattern. To achieve this objective, official reserves are managed in such a way as to exclude the volatile components of reserves held by the central bank. The value of gold is also excluded, since it has historically been the case that liquidating gold is, psychologically, close to impossible.
In any case, Bank Indonesia’s management of international reserves holdings had proven constructive in supporting the managed float that was working so well until the crisis hit. Indonesia had been very prudent in its reserve management, so that when the IMF arrived there was a ‘surplus’ of $5 billion of foreign exchange arising out of the change from the concept of official reserves to the IMF method of reserve calculation to which Bank Indonesia shifted officially in January 1998. The total amount of reserves available and the conservative method of reporting, plus prudent use of reserves in market intervention in the early period of the crisis, put Indonesia in a less precarious position than Thailand or Korea at the time the decision to seek a standby arrangement was made. Why, then, has Indonesia’s recovery been so much slower than that of Thailand and Korea? The problems of the banking sector in these three countries seem to be similar. The difference lies in the relative size of their respective external debt and short-term corporate external debt. In terms of the ratio of total debt to GDP, of total debt to exports or of short-term external debt to international reserves, Indonesia has been the most vulnerable. Indonesia also embarked on the resolution of corporate debt much later and through a much slower process than either Korea or Thailand.
Even though the condition of the banking industry was similar in the three countries, the restructuring programs were not identical, in particular with respect to bank closures. Cole and Slade (1999) point to substantial differences between the closing of 16 banks in Indonesia and the suspension of 58 finance companies in Thailand and 14 merchant banks in Korea. The 16 banks in Indonesia were part of the national payments system and their closure, which led to public concern that other bank closures would follow, had a substantial impact on it. In addition, the bank closures were permanent and settlement was immediate, while the suspension of finance companies and merchant banks in Thailand and Korea was not permanent and the settlement was carried out without immediately affecting the payments system.
Some studies show that the characteristics of the prevailing regime, as measured by various indicators, play an important role in the duration and depth of a recession (Hussain and Wihlborg 1999; Pomelearno 1998; Claessens et al. 1999). This research suggests that the long and deep recession that Indonesia experienced was to be expected, because of its high index of corruption, its high concentration of asset ownership, its low rule-of-law score and its weak creditor-oriented insolvency procedures.
Furthermore, the problems of Indonesia’s social and political infrastructure seem to have played a more serious role in the crisis and its resolution than was the case in Thailand and Korea. These factors have all contributed to the more precarious nature of Indonesia’s financial and economic recovery.
SOME CONTROVERSIAL ISSUES
I will not touch on issues during the latter part of the crisis, nor will I discuss the policies and the challenges for economic recovery. What I would like to highlight here are some issues relating to the crisis and its resolution that still arouse controversy: the decision to float the rupiah; the issue of bank closures; the dilemma of bank rescue using Bank Indonesia liquidity credits; and the currency board question.
The Dilemma of the Bank Closures
The closure of 16 insolvent banks as the first instalment of the economic stabilisation program on 1 November 1997 aroused widespread criticism both domestically and internationally. It may even be correct to say that criticism of this action (irrespective of whether it is valid or not) was stronger than that faulting the IMF for giving incorrect advice to Indonesia, leading it to raise interest rates so high that the economy collapsed.19
It is certainly ironical that an action originally aimed at restoring confidence in banking ended up causing that confidence to be almost totally lost. Various arguments have been launched against the bank closures of November 1997, including the following.
The government announced in early September 1997 that in the event of bank closures it would guarantee small deposit holders. In a well planned national operation to close more than 400 bank offices all over Indonesia, Bank Indonesia officials, assisted by members of the private banks association (Perbanas), performed a fine job in repaying more than 800,000 small deposit holders of Rp 20 million or less. It was certainly not the case, as the IMF’s Dr Boorman alleged in a press conference (see note 19), that the problems arose from lack of public information from Bank Indonesia on the deposit guarantee scheme.
The fact was that bank runs happened because large deposit holders withdrew their funds after losing confidence in the banking system. As deposit insurance would not normally cover large depositors, it is difficult to argue that the outcome would have been any different had Indonesia had a deposit insurance scheme before the bank closures.
On the argument that there were other weak banks besides the 16 that were closed, this is indeed true. However, to argue that more banks should be closed is definitely not an answer to the loss of confidence following the closures. In other words, it is absurd to say that the bank closures would have been more successful—i.e. that there would not have been bank runs—if a larger number of banks had been closed.21 Part of the public anxiety that led to the panic withdrawal of bank deposits lay in the fact that the environment was characterised by lack of transparency, coupled with weak rules on disclosure and weak governance in both the regulator and the private sector. There was thus no established yardstick by which the public could evaluate bank status, and public confidence was very fragile. Meanwhile, in the absence of reliable official information on banking conditions, some publications about banking in Indonesia kept reporting on the fragile condition of national banks, adding to public anxiety.22 In this climate, a process of ‘flight to safety’ occurred, and the banking industry experienced a real crisis.
Claims that other banks were worse than some being closed, or that there was political motivation behind the bank closures, were unfounded. They emerged as a protest from some bank owners who in normal circumstances would not dare to resort to such absurd arguments. Were the bank closures of November 1997 a mistake? I think a more careful analysis still needs to be conducted. However, I would like to underline here that I subscribe fully to the need to close insolvent banks as a part of bank restructuring. In fact, I had already submitted a report to the President recommending closure of a number of banks at the end of 1996, almost one year before the actual decision to do it. It is true that the number of banks proposed for closure was less than 16. But all these banks were among the 16 insolvent banks that were actually closed in November 1997. 23
In general, it can be argued that closing insolvent banks is a must, but the question of when and how to execute bank closure is critical. The Indonesian experience teaches us that when public expectations are fragile, bank closures can produce adverse results. Bank liquidation should be done when the economy is not fragile, but in general the sooner the better. This means that the sooner the problem is identified, the sooner the authorities accept the reality of the issues involved, and the sooner a well designed resolution is prepared and properly implemented, the better. This will cost less and hence has a better chance of success. With the benefit of hindsight, it is clear that these issues were not taken into account by those involved in the decision, myself included. My focus in preparing for the bank closures was on how to deal with payment to small deposit holders, and this was conducted so well by Bank Indonesia staff that no serious problems occurred—a fact overlooked by many. As it turned out, it was an irony that the 16 banks were the only private banks protected from massive deposit withdrawals (Cole and Slade 1999: 6). The state banks did not experience large deposit withdrawals. Indeed they gained more deposits, as people moved their funds from banks perceived to be weak to stronger ones. During the panic, perception was clearly guiding people’s decisions. It did not matter whether the perception was supported by substance. If only the public had been well informed, they would have known that most state banks were not in a better position than many of the private banks.24 Faced with massive deposit withdrawals, the banking industry was in crisis indeed. Even though some banks, state banks and foreign banks in particular, were in a surplus position, they were restricting supply to the interbank money market to their primary customers. The interbank money market became compartmentalised and some banks, in particular weak banks, could not get access to liquidity. Bank Indonesia was forced to carry out a rescue operation, resorting to the existing mechanisms of providing liquidity through discount windows and liquidity credits. This became more pronounced after the President stated publicly, just before leaving on an overseas trip in late November 1997, that the government would not close any more banks in the future.
There had been much criticism of Bank Indonesia’s policy of rescuing banks in distress during the turbulent period after the loss of confidence in the banking sector. The IMF and other foreign observers criticised Bank Indonesia’s excessive liquidity expansion in support of banks. Others criticised the policy by questioning its motivation. Even if these criticisms were valid, in general they overlooked the fact that, especially after the government had stated that it would close no more banks, Bank Indonesia did not have much choice in carrying out its mutually conflicting functions as lender of last resort and guardian of the payments system on the one hand, and monetary manager on the other.
The last matter I would like to discuss here is the adoption of a blanket guarantee or full protection for depositors and creditors towards the end of January 1998. This decision arose from a suggestion by the IMF team during the discussions on the second LOI in mid January 1998. Officially the negotiations were personally conducted by President Soeharto. It was never clear why this was done. But in part the reason must be that President Soeharto had by this time become impatient with the way the crisis was developing; it may also have been because he no longer trusted the Governor of BI and the Minister of Finance.
I and the other members of the Indonesian team were never comfortable with agreeing to the government provision to guarantee in full not just bank deposits and savings but also creditors. But the final decision was that the full guarantee was for both banks’ liabilities and their assets, except those belonging to shareholders and holders of subordinated debts.25 A more serious study should be conducted to analyse the real impact of the guarantee on the development of Indonesia’s banking industry. But suffice it here to say that the guarantee was a desperate policy adopted in the midst of a panic caused by bank runs and the policy response to them, during the most turbulent period of the Indonesian crisis.26
It should also be mentioned that one of the two most critical causes of the Indonesian crisis, namely the problem of unsustainable corporate debts in foreign currency, was not touched on during the preparation of the first LOI;27 nor was the question of social safety nets. These two issues were tackled by the IMF only after strong argument from the Indonesian team, and following criticism by pundits all over the world of the IMF’s role in the Asian crisis. On corporate debt, the initial reaction of the IMF team was simply to reiterate the position against bailing out private debtors, while issues such as social safety nets have never really been the bread and butter of IMF staff.28
The CBS Controversy
The idea of fixing the rupiah to the dollar, coupled with the adoption of a currency board arrangement or system (CBS), as it is popularly known, attracted considerable public debate, even if only for a short period amidst the anxiety about the crisis and how to cope with it. It was coloured by some curious stories about how the idea had come to the attention of President Soeharto, about how he had toyed with it, and about its ultimate rejection. For one thing, it has never been clear who was responsible for inviting Professor Steve Hanke, a known proponent of the CBS and a professor of Applied Economics at Johns Hopkins University, to come to Indonesia, and introducing him to President Soeharto. In one of his visits to Jakarta Professor Hanke is reported to have registered at the Hotel Shangri La under a different name (Simon Holland) as a guest of a large Indonesian company. He was Vice Chairman of the Friedberg Mercantile Group, a finance company involved in currency transactions, including rupiah, at the end of 1997, and was appointed adviser to the Indonesian Council for Economic and Financial Resilience (AWSJ, 24 February 1998). It would be difficult to know the real reason why President Soeharto seemed desperate to adopt Mr Hanke’s advice to peg the rupiah to the dollar under a currency board arrangement. I can only speculate on the following reasoning. At the outset the President was convinced that the crisis was basically a financial crisis, i.e. drastic depreciation of the rupiah due to speculative attacks in the currency market. He believed the economic team, in particular the Minister of Finance and Bank Indonesia’s Governor, should be able to fix it within a few months.29 When he had to face the fact that, even after six months of distress and the humiliating decision to request the IMF standby arrangement, the problem was actually spreading, he believed a new alternative must be found. And by this time he desperately wanted to get the economic problem over in order to concentrate on more pressing issues related to his re-election bid.
In this environment, Mr Hanke’s plan, which promised to strengthen and stabilise the rupiah ‘by the stroke of a pen’—by issuing a law setting a new rupiah/dollar value and adopting a CBS—was just what President Soeharto dreamt of. Indeed, while the bill was being drafted by a number of Bank Indonesia officials and senior officials from the Ministry of Finance, the President had already made an approach to the leadership of the parliament to smooth the way for its enactment.30 However, the CBS plan was aborted. International pressure was mounting for President Soeharto to adhere to the IMF-supported program, which meant continuing to float the rupiah and implementing the programs for stabilisation and recovery set out in the LOI to the IMF. Phone calls from heads of government of industrial countries and visits by prime ministers and cabinet members from several countries formed a chorus urging President Soeharto to keep the IMF-supported program on course. Domestically, a memorandum by the Monetary Board showing the arguments for and against adopting a fixed exchange system with a currency board must have influenced the President’s decision to discard the plan.
If we put all the social and political considerations aside, there were still many unresolved technical issues surrounding the implementation of a pegged exchange rate and currency board. Choices such as which concept of money supply to use and what rupiah /dollar rate to apply would determine the level of reserves the government should have. There was also some question as to where Indonesia could raise additional reserves should the need arise. Certainly, if base money were chosen as the basis of money supply for the purpose of calculating the international reserves Bank Indonesia had to hold, the result would be very different from that obtained if broad money were used as the basis. If the rupiah were fixed at 50% stronger than the current rate of 10,000, i.e. at 5,000/$, the amount of reserves to be held would increase from $30 billion to $70 billion. It was my contention that since Indonesia adhered to an open capital account, the amount of reserves held must cover the broad money supply, including at least some portion of bank deposits. The available reserves of around $20 billion could only cover about one-third of what was needed.
On the other hand, the factor that worried me most was that a pegged exchange rate with a CBS required very strong discipline from political leaders, the bureaucracy and the private sector. This system is sometimes referred to as ‘an auto pilot system’. But I was not confident that Indonesia had sufficient discipline in these spheres. Nor was the legal infrastructure commensurate with what was required. I therefore believed that adopting this system was not an option for Indonesia at that time. In my discussions with Professor Hanke I told him that Indonesia faced problems of confidence in the rupiah, in the banking system and in the corporate sector’s ability to repay its debt. Successful adoption of his proposal would address only the first problem. Obviously, he did not accept my argument. The Question of My Dismissal
The last point I would like to highlight here is the issue of my premature departure from the position of Governor of Bank Indonesia. There were many reports in the newspapers that this was due to my opposition to the adoption of a CBS. Were these claims right? And what was the actual reason for my being ‘discharged with honour’ a couple of weeks before my term was over?
I am afraid we will never know the real reason for my early departure. When President Soeharto summoned me to his residence on 17 February 1998, he told me that he wanted to make a few personnel changes, as he had done with the Commander of the Armed Forces on the previous day and was to do with the Attorney General when the then cabinet was dissolved. He told me that he wanted to replace me with Dr Syahril Sabirin, whom he had appointed as a Bank Indonesia managing director, together with three others, just before Christmas 1997, and that the swearing-in would be conducted on 19 February.31 He then thanked me for my service in his cabinet over 10 years and with difficult responsibilities. In return, I thanked him for entrusting me to serve in his cabinet in a post of strategic importance. I told him I felt sorry to see that there were so many unresolved problems which would complicate his administration. And that was it.32
My dismissal made news domestically as well as internationally, though not because it was against the law. It was, however, definitely unusual in Indonesia during Soeharto’s reign of 32 years. He did indeed discharge the Minister of Trade when he merged that ministry with Industry (December 1995), and he did something similar when the Minister of Information took the job of Speaker of the parliament (June 1997). But the circumstances were different from those when I was dismissed.
In a presidential cabinet, a cabinet member can be dismissed at any time. This is true. And since 1983 it has been established that the Governor of Bank Indonesia concurrently also holds the status of a cabinet member. I think it was perfectly legal to dismiss the Governor as a member of the cabinet. It was simply uncommon. However, one might raise a question about the dismissal of the Bank Indonesia Governor before his term is over. The law actually stipulates that the President can dismiss the Governor before the end of a five-year term only in the following circumstances: at the personal wish of the Governor; on the Governor’s death; or when, for some reason, the Governor cannot serve in the office.33 Thus it is indeed confusing to interpret the actual meaning of ‘discharge with honour’, the official term used in the decree of my dismissal. I think the foreign press used more straightforward and realistic terms in their reporting of the issue.34
This has been a personal note offering an analysis of a crucial part of the Indonesian crisis, which started in mid July 1997 as an external disturbance in the currency market in Jakarta in the context of a financial panic in the region. Through a dynamic process of policy response and market reaction with feedback effects, the disturbance spread rapidly to become a banking crisis and an economic crisis, and later a social and political crisis. This is not a comprehensive account or analysis of the Indonesian crisis, but a personal note describing aspects of what happened.
The Indonesian crisis, and the part that I played in the midst of the turbulence, needs to be clearly understood so that some lessons can be drawn for future management of the national economy. It is generally true that in the dynamic development of an economic and financial process characterised by uncertainties, not much can be learned from the past, from history. But it is still the case that studying the past, even a small part of it, will teach us not to make similar mistakes. After all, as Keynes reminded economists, we must study the present in the light of the past for the purpose of the future.35 Clarifying even a small part of so complex a problem as the Indonesian crisis may contribute to more rigorous study and allow us to benefit from lessons that are to be learned from this important, if dark, period of our history.
1 Indonesian newspaper headlines in the month of August were full of the Bank Bali scandal, dubbed ‘Baligate’ by some.
2 I would argue that the external shock was a severe one. This is clearly demonstrated by the fact that in previous shocks, in January 1995 as an effect of the Mexican crisis and in July 1996 after the Jakarta riots, Bank Indonesia was able to contain the problem and prevent it from developing into a crisis (see note 7).
3 There is another way of reading the Asian crisis. Delhaise (1998) distinguishes between the financial panic, which resulted in drastic currency depreciations, and the actual crisis arising from high growth unsupported by proper development in the financial sector.
4 The World Bank produced a report on the social impact of the Indonesian crisis (World Bank 1998a) and a more general report on the social consequences of the East Asian crisis (World Bank 1998b). There have been new data on the social impacts of crisis which may differ from those cited here. On this issue I benefited greatly from reading two papers, McCawley (1999) and Feridhanusetyawan (1999).
5 From a column by Lim Say Boon, ‘The Art of the Possible’, Far Eastern Economic Review, 4 February 1999.
6 For a discussion of how the crisis unfolded, see Djiwandono (1998).
7 When the previous external shocks occurred, in response to the Mexican crisis in January 1995 and the Jakarta riots in July 1996, the rupiah weakened, but recovered within days after Bank Indonesia’s intervention of around $700 million in the first case and a little over this amount in July 1996.
8 Bank Indonesia also engaged in forward sales and purchases during this period, but not on a very substantial scale.
9 This was a controversial action. I had said publicly on many occasions that I did not subscribe to any form of shock therapy such as this step, which was known previously as a ‘Sumarlin shock’ (see Glossary). In a period of banking distress, when liquidity was becoming very scarce for many banks, this action, especially taken in combination with other liquidity squeeze measures, placed unnecessary strain on the banking sector.
10 This policy package comprised the macroeconomic policy steps that a standard IMF-supported program would include. In fact, within Bank Indonesia I called the 3 September policy ‘a self-imposed IMF program’.
11 Of course this argument could be challenged. Proponents of the ‘financial panic’ explanation of the Asian crisis had been arguing for a concept of prudent reserve management in which a country holds at least enough international reserves to cover short-term external debt. But even using this definition it could be argued that at the time the crisis started Indonesia’s reserves, using the international standard for reserve calculation, were around $29 billion, which was ‘only’ $6 billion short of the national short-term debt. Nonetheless, I agree that maintaining a prudent ratio between national reserves and short-term national debt (both state and corporate) should be an integral part of prudent macroeconomic management. It is unfortunate that for so long the ‘conventional wisdom’ about reserve holdings still involved the ‘magic number’, i.e. the equivalent of the financing of three months of imports.
12 Taken from ‘Indonesia: Memorandum of Economic and Financial Policies’ and the Letter of Intent of 31 October 1997 (mimeo). It is unfortunate that these documents and the second LOI signed by President Soeharto were not publicly available. All other letters of intent after these two are available on the IMF website, as part of the government program to increase transparency.
13 The foreign market reaction to the bank closures was favourable at first. Bank Indonesia was viewed as being serious in addressing the problem, since even banks known to be owned by politically well connected people were included in the closures. But when, with the purchase of a bank by the owner of a closed bank, it was perceived that Bank Indonesia was yielding to political pressure, the reaction of foreign markets became strongly negative. The domestic market’s perception was that the government was serious, but its reaction was completely different. Many thought that if even politically well connected banks were closed, then there must be more bank closures to follow. Thus, with the change in the foreign market perception of the bank closures, both markets reacted negatively to the closures, and loss of confidence in the banking sector became a sad reality.
14 The EFM was established in 1996, but was improved in December 1997, and again in January 1999, to hasten the decision making process and the drawing of funds by recipients when the IMF was dealing with the Brazilian crisis. In the last mid year Interim Committee Meeting of the IMF in April 1999, a new facility, the Contingent Credit Line (CCL), was introduced to deal with potential currency problems of member countries. This last facility was designed to help member countries early, to avert the onset of a crisis altogether.
15 When a country is admitted as a member of the Fund it is required to subscribe an amount of capital—its quota—25% of which is to be in reserve currencies, usually US dollars, and the rest in its own currency. The size of each member country’s quota is used as a basis for calculating its voting rights and the amount of facilities it can draw from the Fund.
16 For a slightly different explanation of this issue, see McLeod (1999): 40–1.
17 In Indonesia, confusion about the real meaning of the second line of defence did create some unnecessary public debate and embarrassment in December 1997, when the Chairman of the Indonesian Chamber of Commerce and Industry (Kadin) inadvertently made a public announcement about President Soeharto’s intention to use Singapore’s loan as part of the second line of defence to finance projects for the small and medium-scale business sector. With a little better knowledge of the issues Indonesia could have avoided this embarrassment.
18 The term ‘bail-out’ may not be proper either. ‘Bailing out’ has the connotation of a free-of-charge facility provided by the rescuer. This is the major reason for the moral hazard associated with the facility. If this had been a bail-out associated with a free rescue operation, many would have argued against the IMF decision to provide standby facilities to crisis countries. But all these facilities are loans with all sorts of associated costs.
19 This criticism was often made by reference to a parable about a traffic accident in which a paramedic is busy fixing the brakes of the car and leaving the driver dying from his wounds.
20 Transcript of a press conference given by Dr Jack Boorman, Director of the Policy Development and Review Department of the IMF, at the public announcement of its study IMF-Supported Programs in Indonesia, Thailand and Korea: A Preliminary Assessment (Lane 1999); see also Radelet and Sachs (1998).
21 Dr Morris Goldstein (1998) argued that more banks than the 16 should have been closed.
22 Some weeklies, for example, reported that the number of unsound banks was 31, instead of 16 (Warta Ekonomi, no. 26/Th. IX, 17 November 1997), and others came up with a larger number.
23 I made my last report to the President on BI’s analysis of the need to close a number of banks some time in March 1997. The President in fact agreed to the policy. However, it was to be implemented after the May 1997 general election and, of course, almost immediately after the election the crisis started.
24 It must have been an unpleasant surprise for the public when IBRA (the Indonesian Bank Restructuring Agency) announced in April 1999 that all state banks were in C category (with capital adequacy ratios below minus 25% and a very high percentage of problem loans).
25 Those questioning Bank Indonesia’s motivation in rescuing banks through the use of discount windows and liquidity credits also attacked the regulation providing this guarantee. They proposed the repeal of this measure, since it could be used as a vehicle in money politics. This argument was brought into the open after the eruption of the Bank Bali scandal.
26 Of course the blanket guarantee had been adopted by all three Asian countries accepting IMF standby loans, as well as by Malaysia (Balino et al. 1999).
27 The only mention of it was the policy for the government not to bail out corporate debtors.
28 The first public debate on the private debt issue was only about why Bank Indonesia’s Governor had no information about it. I have to reiterate here, as I did on several occasions at the end of 1997, that what BI knew about in detail then was corporate debt with loan contracts. On other corporate debt, BI data were not complete. But I initiated talks with corporate owners (conglomerates) having external exposure of $25 million, $50 million and $100 million and over, requesting them to report to Bank Indonesia. And by the time Indonesia called in the IMF, BI already had better information. It was reported in the first LOI that the total national debt was $140 billion, of which $80 billion was private sector debt, $35 billion of it short-term. These were the data submitted to the Private Debt Team chaired by Mr Radius Prawiro. If only this first LOI had been made public, it would have answered BI’s critics on this issue.
29 This view is based on the fact that, at the beginning of the crisis, after studying a report on financial development by Minister Coordinator Saleh Afiff, the President gave an instruction to the Monetary Board to prepare a daily report on the issue for about three months.
30 In one of the meetings of the Council for Economic and Financial Resilience, President Soeharto stated that he would adopt the CBS proposal. This must be the basis of Minister of Finance’s statement during a parliamentary hearing that the government would be establishing a CBS. The new central bank. Governor also made a statement on Bank Indonesia’s role in a CBS (Jawa Pos, 25/2/98).
31 On 20 December 1997, without consulting me, the President dismissed four of my seven managing directors, and replaced them with officials of his own choosing. This was unusual, to say the least.
32 Of course, that was the formal notice given to me by the President. But his decree to ‘discharge me with honour’ had already been signed on 11 February. And since secrets cannot be kept for long in Indonesia, I had actually learned about my dismissal five days earlier, on 12 February. In fact, I had to go to Denpasar on the very day I heard of the dismissal, to host the SEACEN (Southeast Asia Central Banks) Governors Meeting.
33 Article 17 of Law No 13, 1968, on Central Banking.
34 Reports published in several papers abroad on 18 February 1998 used the following terms in describing my dismissal: ‘dismissed’ (New York Times and Straits Times), ‘sacked’ (Asian Wall Street Journal), ‘dumped’ (Financial Times), and ‘fired’ (Washington Post and International Herald Tribune).
35 Adapted from Keynes’s characterisation of a ‘master economist’ in his biographical essay on Alfred Marshall, cited by Solomon (1999: vii).
Arndt, H.W., and Hal Hill (eds) (1999), Southeast Asia’s Economic Crisis: Origins,
Lessons and the Way Forward, Institute of Southeast Asian Studies, Singapore.