This Selected Issues paper assesses Indonesia’s trade integration relative to underlying country characteristics. The paper analyzes Indonesia’s vulnerabilities, especially compared with the eve of the crisis in 1997. Various indicators suggest that the underlying fundamentals are significantly stronger. The paper examines key features of the financial safety net (FSN) in view of international standards and concludes that the current system is capable of timely addressing bank problems. It looks at determinants of, and constraints to, credit growth in recent years.

Abstract

This Selected Issues paper assesses Indonesia’s trade integration relative to underlying country characteristics. The paper analyzes Indonesia’s vulnerabilities, especially compared with the eve of the crisis in 1997. Various indicators suggest that the underlying fundamentals are significantly stronger. The paper examines key features of the financial safety net (FSN) in view of international standards and concludes that the current system is capable of timely addressing bank problems. It looks at determinants of, and constraints to, credit growth in recent years.

III. Indonesia, 1997 vs. 2007: How Far Has Crisis Vulnerability Been Reduced?9

A. Introduction

29. Of all the emerging market economies hit by the 1997-98 financial crisis, Indonesia was most severely affected, and took longest to recover. Output declined by 13 percent in 1998, while the rupiah had lost more than 80 percent of its value by June 1998. Unemployment, inflation and poverty soared. This happened despite Indonesia’s economy appearing (at least superficially, based on headline macroeconomic indicators such as the current account deficit) in better shape prior to the crisis than some of the other affected countries. However, perhaps more than in any other country, the experience of the crisis triggered deep-rooted institutional reform.

30. A key question for Indonesia, given the devastating effect of the 1997-98 crisis, is whether the changes to its economy and institutions in the intervening years have substantially reduced its crisis vulnerability. The tenth anniversary of the onset of the crisis provides a good occasion for such an assessment.

31. Financial crises can be caused by a multitude of factors,and the literature on their genesis is exhaustive (see Allen and others, 2002 and Dornbusch, 2001). However, the causes can be grouped into four broad but inter-related groups:

  • Macroeconomic policies, including unsustainable or mutually incompatible fiscal, monetary and exchange rate policies;

  • Weaknesses in the financial and corporate sectors, including excessive risk-taking by firms and balance sheet mismatches;

  • Contagion as a crisis in another country leads to a collapse in investor sentiment;

  • Weak, inefficient or corrupt institutions that can create poor policies, exacerbate structural weaknesses, and undermine investor confidence, as well as delaying the response to a crisis once it occurs.

32. The following sections discuss these factors in turn,briefly commenting on their role in the 1997-98 Asian crisis. Indonesia’s progress in overcoming these vulnerabilities is then assessed for each set of factors, both in absolute terms and relative to the other countries affected by the crisis to a greater or lesser extent (Korea, Malaysia, the Philippines, and Thailand). Overall, Indonesia has made significant progress since 1997, reflecting sound macroeconomic policies, institutional reform and, in recent years, a benign external environment.

33. Each “generation” of financial crises tends to arise from hitherto unknown or underappreciated vulnerabilities. The final section therefore considers some new potential sources of vulnerability as cited by some analysts. In as far as these vulnerabilities can be quantified, risks here too seem relatively lower in the case of Indonesia relative to many other countries.

B. Macroeconomic Policies

34. Macroeconomic policies have a central role in both the first and second generation currency crisis literature (Krugman, 1979; Flood and Garber, 1984; Obstfeld, 1994). The role of macroeconomic policies lies in the tension between domestic and external policy objectives. In particular, if a country pursues external stability by pegging the exchange rate, while at the same time attempting to achieve domestic policy objectives via expansionary monetary, credit and/or fiscal policies, the currency will become overvalued in real terms and vulnerable to speculative attack.

35. In fact, weak macroeconomic policies are not seen as having played a central role in any of the Asian crisis countries (except perhaps in as far as policies such as maintaining a fixed exchange rate encouraged excessive private sector borrowing and poor risk management in the financial sector). Fiscal policy was conservative: Indonesia ran a surplus of 0-1 percent of GDP in the run-up to the crisis (Table III.1) While external current account deficits were quite high in Thailand and Malaysia, Indonesia’s current account deficit was modest, at just over 2 percent of GDP, while the real exchange rate appreciated only mildly in the run-up to the crisis (Figure III.1) and did not appear over-valued in Indonesia (Goldstein, 1998).

Table III.1.

Selected Countries: Macroeconomic Indicators, 1992-2006

article image
Source: IMF, World Economic Outlook.

36.However, loose credit conditions and a fixed exchange rate encouraged risky lending (often in foreign currency) in a number of affected countries,which created vulnerabilities in the corporate and financial sectors (Krugman, 1999). Money and credit growth was particularly rapid in Indonesia, with broad money growing at almost 25 percent per year during 1992-96. The M2/GDP ratio, an indicator of financial intermediation, expanded by around 2-3 percentage points per year, strongly suggestive of a credit boom. Rapid credit creation interacted with structural weaknesses in the financial sector to create new vulnerabilities (discussed below), particularly with respect to currency and maturity mismatches on financial and corporate sector balance sheets. Indonesia also placed itself open to speculative pressure (as did the other Asian crisis countries) by pegging its exchange rate.10

37. Current macroeconomic policies in Indonesia are sound. Fiscal deficits are modest and debt is declining, the external current account is in surplus, the exchange rate is more flexible (and the real exchange rate is estimated to be modestly undervalued; see IMF, 2007). While credit expansion has picked up recently, the M2/GDP ratio has declined to 40 percent (a 14 percentage point decline since 1998). A steady build-up of reserves and lower external borrowing has reduced the ratio of short-term debt to reserves to less than half of the pre-crisis level, well below the 100 percent level implied by the Greenspan-Guidotti rule. However, inflation remains higher and more volatile than in other regional economies, which could ultimately put pressure on the exchange rate.

Figure III.1.
Figure III.1.

Selected Countries: Real Effective Exchange Rates

(January 1990 = 100)

Citation: IMF Staff Country Reports 2007, 273; 10.5089/9781451818413.002.A003

Sources: IMF, APDCORE database; and Fund staff estimates.

C. Financial and Corporate Sector Weaknesses

38. Rapid credit creation in the first half of the 1990s exacerbated structural weaknesses, particularly with respect to private sector balance sheets. Much of the credit expansion was undertaken via overseas borrowing, mostly short term and denominated in foreign currencies. Moreover, the extent of the rapid build-up in external borrowing only started to emerge mid-1997, contributing to the slide in confidence. Currency mismatch was encouraged by the exchange rate peg. By end-1996, Indonesia’s ratio of short-term external debt to international reserves was more than 170 percent (Table III.2). Total borrowing by the nonbank private sector from international banks stood at almost two times reserves in mid-1997 (Table III.3). A significant portion of the foreign borrowing went into sectors, such as real estate, which tended to earn little or no foreign exchange. Domestic banks were also lending in foreign currency, which totaled more than 30 percent of total credit outstanding by the end of 1997 (of which one-tenth went to the property sector; Table III.4). Commercial banks were also borrowing significantly in foreign currency (including via foreign currency deposits), with total foreign currency liabilities representing more than 30 percent of total bank liabilities by end-1997.11 Hence Indonesian banks were doubly squeezed once the crisis hit: devaluation significantly increased the value of their liabilities, while on the asset side corporate borrowers defaulted on their foreign currency loans.

Table III.2.

Short term External Debt

(In percent of reserves)

article image
Sources: WB, World Development Indicators; IMF, International Financial Statistics; and Fund staff calculations
Table III.3.

Non-bank Private Sector Cross-border Borrowing

(From BIS-reporting banks, percent of reserves)

article image
Source: BIS Quarterly Review; and IMF, International Financial Statistics
Table III.4.

Commercial Bank Credit Outstanding, End of Year

article image
Source: Bank Indonesia.

39. Lax prudential rules encouraged risky lending, leading to deteriorating asset portfolios. Prior to the crisis, the ratio of non-performing loans (NPLs) to total loans was almost 9 percent in Indonesia (Table III.5).12 High pre-crisis NPLs demonstrated how connected lending and poor credit-risk management had jeopardized the health of the banking system. In addition, they made currencies more vulnerable to speculative attack by increasing the costs of an interest rate defense of the exchange rate anchor.

Table III.5.

Non Performing Loans Ratio

(In percentage points)

article image
Sources:1996: BIS, 1997; quoted in Goldstein, 19981997-98: Ramos 1998, for Goldman Sachs; quoted in Goldstein, 19982005: Financial Soundness Indicators Consolidated Collection ExerciseExcept Thailand: April 2007 Global Financial Stability ReportMalaysia: IMF Staff Estimates

40. underlying weaknesses in the financial and corporate sectors helped to create these financial fragilities. In Indonesia more than half of the corporate sector was controlled by just 10 families prior to the crisis (IMF, 2006). This concentrated ownership structure meant that risks were not well diversified, and also contributed to corporate governance problems (Turner, 2007). Moreover, many large corporate groups owned banks that lent intra-group significantly in excess of regulatory limits. Post-crisis, around half of lending by banks that failed and ended up under the control of the restructuring agency, IBRA, was found to be intra-group (Pangestu and Habir, 2002). Financial regulation and supervision was perceived to be improving in the run-up to the crisis.13 However, post-crisis it was revealed that violations of prudential regulation (for instance, with respect to intra-group lending) were widespread, as was financial misreporting more generally. Partly, this reflected the concentrated and opaque ownership structure of the banking and corporate sectors, with a small number of players—well-connected politically—able to manipulate the system. Politically-connected banks were seen as “too important to fail,” generating moral hazard problems. Regulatory and supervisory capacity at Bank Indonesia (BI) and elsewhere was inadequate (Pangestu and Habir, 2002).

41. There have been significant improvements in financial sector regulation and supervision since the crisis. Regulatory changes in the banking sector tightened rules on, inter alia, loan classification and provisioning, related-party lending, capital adequacy and foreign exchange rate risk (IMF, 2004 provides a detailed account of the reforms). There has also been an improvement in regulatory capacity at BI, and supervision is more regular and thorough. In addition, the restructuring of insolvent banks post-crisis removed banks from related corporate groups, helping to solve the related lending problem (IMF, 2006).

42. However, weaknesses remain. In the banking sector, the post-crisis restructuring has bequeathed a large share of the banking system (more than one third by assets) to state-owned banks, which generally have weaker financial performance and loan quality, and are more exposed to the potential for government interference in operational matters (Nasution, 2007). In the corporate sector, disclosure of cross-ownership can still be insufficient, and financial statements are still not fully consistent with international norms, while ownership remains concentrated (IMF, 2006).

43. Improvements at the policy and institutional level have been reflected in a post-crisis recovery in financial and corporate sector indicators. In the banking sector, NPL ratios have fallen significantly from their crisis peak, and are comparable to other countries in the region. The corporate sector is less leveraged, with debt-equity ratios having also fallen back to below the levels in other regional economies (Table III.6), and estimated corporate default probabilities are low and below the average for emerging market economies (EMEs) in the region (although this is also the case when a similar calculation is made using 1996 data; Table III.7). However, corporate performance differs by type of company. In particular, domestic firms with a concentrated ownership structure remain more vulnerable, as they tend to be both less profitable and more highly leveraged.14 Banks’ capital-asset ratios have increased significantly (Table III.8).15 However, ratings agencies have been slow to upgrade their assessment of banks’ underlying financial strength (Table III.9).16 Ratings for Indonesian banks have improved more than in some countries, but less than in others (particularly Korean institutions).

Table III.6.

Corporate Debt as Percentage of Equity 1/

article image
Source: IMF, Corporate Vulnerability Utility database.

Weighted by capitalization

Table III.7.

BSM Probability of Default

article image
Source: Fund staff estimates.
Table III.8.

Bank Regulatory Capital

(In percent of risk weighted assets)

article image
Sources: Turner (2007); and IMF, Global Financial Stability Report April 2007.

1995 CARs can be regarded as overestimates, since inadequate provision made for impaired loar

2006 CARs: September (Indonesia and Korea); November (Malaysia and Thailand).

Data for 2005.

Table III.9.

Selected Countries: Fitch Individual Ratings on Banking Sector

article image
Source: Turner (2007) and Fitch Research.

as of December.

Most recent ratings, generally January - May 2007.

44. There also appears to be less exposure to devaluation risk via international borrowing in foreign currency. International bank lending (from BIS reporting banks) to Indonesian entities (public and private sector, including banks) remains below pre-crisis levels (Table III.10). Moreover, the proportion of such lending undertaken via local affiliates in local currency—as opposed to cross-border or locally in foreign currency—has quadrupled to around one-third, implying significantly lower currency mismatch for local borrowers. However, this proportion has increased even more markedly in Korea, Malaysia, and Thailand. International borrowing by the nonbank private sector has declined from almost two times reserves in mid-1997, to just over one-half more recently, although the level of borrowing, as a proportion of reserves, remains the highest of the Asian crisis countries.

Table III.10.

Lending to Selected Asian Countries by BIS- Reporting Banks 1/

article image
Source: Bank for International Settlements Quarterly Review (December 1997 and March 2007).

Total: international claims (cross-border loans plus local loans in foreign currency) plus local loans (made through local affiliates in local currency). Local: local currency loans made by local affiliates.

D. Contagion

45. Contagion is seen as a critical component of the Asian financial crisis (Baig and Goldfajn, 1998; Berg, 1999; Goldstein, 1998). For Indonesia, where macroeconomic fundamentals were largely benign and structural weaknesses did not show up until the crisis had broken out, contagion likely played a large role. Contagion (from country A to country B can occur via three channels (Goldstein, 1998): (i) direct real effects via bilateral trade and investment flows between countries A and B, (ii) speculative pressure on country B’s currency as the devaluation in country A undermines B’s external competitiveness, and (iii) a “wake up call” to investors as the perceived risks to investors in country B increase in response to the crisis in country A.

46. Direct trade flows within East Asia were fairly limited prior to the crisis (Goldstein, 1998). Investment flows, though harder to quantify, were likely equally insignificant. Competitive devaluation pressure could have played a role for some countries, although perhaps less so in Indonesia’s case, due to less direct export competition in third-country markets. The “wake up call” view seems more plausible: market perceptions of risk across the region were excessively sanguine in the run-up to the crisis (reflected in stable or improving sovereign credit ratings and declining spreads). The onset of the crisis led to a rapid reassessment of risks and speculative pressure on the rupiah quickly mounted.

47. Indonesia may have become more vulnerable to contagion from other regional economies,principally via the investor sentiment channel. The potential for contagion via other channels has also increased, but remains small.

  • Bilateral trade flows remain relatively low but have increased (Table III.11)

  • Competitive devaluation remains an unlikely potential source of contagion, but here again risks have increased somewhat. Indonesia’s export composition continues to differ from those of its principal regional competitors. According to the export similarity indices presented in Table III.12 (which vary between 0 and 100 and increase in export similarity), Indonesia’s export pattern remains the most specialized. However, it has become a little more similar to those of regional counterparts over time.

  • Regional financial markets are more integrated. To illustrate this increased integration, the Table III.13 shows the comovement of daily stock market returns in Indonesia and other regional economies during 1995 (pre-crisis), 1997-98 (the crisis period) and 2005-06 (to show recent behavior).17 Greater comovement in stock returns is taken as evidence of greater market integration, creating heightened contagion risks.

Table III.11.

Indonesian Exports

(In percent of total exports)

article image
Source: IMF, Direction of Trade Statistics.
Table III.12.

Export Similarity, Indonesia and Selected Economies, 1996 and 2005

article image
Source: Comtrade, via WITS, and IMF Staff Calculations.Notes:18
Table III.13.

Estimated Comovement of Daily Stock Returns, Indonesia and

Selected Countries.

article image
Sources: Bloomberg, Datastream, and Fund staff calculations.

:signifies that corrected correlation coefficient is significantly different from the 1995 measure at the 1 percent significance level (only with respect to individual countries).

Average is unweighted mean for all six countries.

  • article image
    According to this measure, comovement generally declined during the crisis period, possibly reflecting the rather different patterns of crisis and recovery in the different countries.

  • article image
    The degree of comovement in 2005-06 is generally higher than during either the pre-crisis or crisis periods (for Indonesia and most of the other countries). This would tend to suggest that Indonesia’s vulnerability to contagion has increased.

E. Institutions

48. In many of the Asian crisis economies there were very close links between governments, banks and large corporations in their very successful growth phase in the run-up to the crisis (Wyplosz, 2007). In Indonesia, banks were often owned by corporate groups with strong political connections (Hill and Shiraishi, 2007). The resulting problems of connected lending, poor credit quality and weak supervision undermined the banking sector and contributed to its collapse in 1997-98 (see section C). In addition, links between particular banks and political groups meant that political instability could trigger bank runs which were only halted when the banks’ assets were taken over by IBRA (Pangestu and Habir, 2002). Generally, these problems, which were well-known prior to the crisis, were reflected by high perceived levels of corruption (Table III.14).19

Table III.14.

Corruption Perception Index Rankings

(Percentile)

article image
Source: Transparency International.

49.Since the crisis, the institutional environment in Indonesia has improved significantly (Hill and Shiraishi, 2007). Political reforms to decentralize power have complimented improvements to financial regulation and supervision outlined in Section C, helping to reduce vulnerabilities stemming from close corporate-political ties. This broad institutional transformation has also been associated with decreased perceptions of risk and instability, which have fallen furthest from their crisis peak among affected countries (Table III,15). Although concerns remain with respect to perceived levels of corruption, there has been progress in this area too.20

Table III.15.

Selected Countries: Risk Ratings

(Rating from 0 to 100)

article image
Source:International Country Risk Guide.

F. New Sources of Vulnerability

50. While Indonesia generally appears less vulnerable now with respect to the crisis risks relevant in 1997, hitherto underappreciated sources of vulnerability tend to emerge from each new crisis. These are difficult to identify ex ante; however, some analysts have attempted to identify new potential sources of risk for the economies hit by the 1997-98 crisis.

51. One potential source of risk is the alleged pursuit of policies that do not permit the exchange rate to appreciate, with the goal of boosting exports and economic growth (Roubini, 2007). According to this view, undervalued exchange rates can increase vulnerabilities by encouraging asset price bubbles and a more rapid growth in the money supply (via only partial sterilization of the reserves build-up and below-equilibrium interest rates). Asian economies are therefore vulnerable to a growth slowdown outside Asia that could trigger an export-led slump and rapid reversal of asset prices, leading to a collapse in confidence and, potentially, second-round effects on the exchange rate, inflation and output.

52. Recent developments in Indonesia are not consistent with this model. Indonesia is less dependent on export-led growth than many other emerging market countries and the reserve build-up has been more modest. In addition, the rupiah has shown considerable flexibility in nominal terms. It has also appreciated by more than 50 percent in real effective terms in the last six years, though much of the real appreciation reflects above-average inflation (Figure III.2).21

Figure III.2.
Figure III.2.

Selected Countries: Real Effective Exchange Rate

(June 2001 = 100)

Citation: IMF Staff Country Reports 2007, 273; 10.5089/9781451818413.002.A003

Sources: IMF, APDCORE database; and Fund staff estimates.

53. As regards asset prices,over the course of the current bull market (dated from the start of the pick-up in the Hong Kong stock index in October 2002) the U.S. dollar value of Indonesian stocks has increased by around 400 percent, ahead of other regional stock markets (Figure III.3). However, Indonesia’s markets were in large part recovering from the prolonged historically low stock values following the crisis and catching up with the recoveries that occurred earlier in the other crisis countries. Good corporate performance and improving macroeconomic indicators supported the recovery. Nevertheless, in Indonesia price/equity ratios, while below historical (pre-crisis) levels, are now above those in the other Asian crisis countries, and still on an upwards trend.22 Hence, while the surge in stock prices does not necessarily imply the existence of a bubble, it does create the possibility of a reversal.

Figure III.3.
Figure III.3.

Figure III.3. Selected Economies: Stock Market Indices

(October 2002 = 100)

Citation: IMF Staff Country Reports 2007, 273; 10.5089/9781451818413.002.A003

Source: Bloomberg.

54. The recent pick-up in the volume of portfolio inflows has been of concern to the authorities. In particular, the short-term nature of the inflows creates a worry that, as in 1997-98, they could be quickly reversed, leading to a collapse in asset prices and a loss of reserves. However, inflows have largely been into liquid government and central bank certificates, rather than illiquid loans to the corporate or financial sector. Hence, a rapid reversal of the inflows could potentially affect asset prices, but is unlikely to have the kind of real effects, via corporate and banking sector balance sheets, that occurred during the crisis. In addition, the capital and financial account surplus has averaged less than 1 percent of GDP in recent years as some banks and corporates have been repaying loans, while the current account has remained in surplus, with the substantial overall surplus leading to reserves accumulation. This suggests that even a quite substantial portfolio outflow could be accommodated without necessitating a drawdown in reserves.

55. Another concern is that despite the significant reserves accumulation, the level of reserves is still insufficient to fully deter or provide cover against speculative attacks or sudden stops.23 For instance, Wyplosz (2007) argues that, while the Asian crisis countries have done much to reduce vulnerabilities, including by significantly building reserves, the level of reserves required to defend the currency in all circumstances is prohibitively large, particularly given the volume of global capital flows and an increase in investors’ risk appetite (that makes the demand for emerging market assets more elastic with respect to expected yield differentials). Therefore, traditional benchmarks of reserves adequacy may paint too complacent a picture.24

56. To illustrate this point, the level of reserves can be compared with the kind of capital account reversals associated with sudden stops and financial crises. Table III.16 shows reserves holdings (as a percentage of GDP) in the Asian crisis countries (and three other emerging market comparators) in 1996 and 2006, set against the capital account reversals that occurred in 1997-98.25 Clearly, the level of reserves in Indonesia in 1996 was insufficient to cover the capital account reversal experienced in 1997 and 1998. Moreover, the level of reserves in 2006 remains at around half the level required to cover a reversal of similar magnitude (as a share of GDP).

Table III.16.

Reserves and Capital Account Reversals

(In percent of GDP)

article image
Source: IMF, International Financial Statistics; and Jeanne and Ranciere (2006).Capital Account Reversals: 1997-98 (Asia), 2002-03 (Brazil), 2001-02 (Turkey) and 2000-01 (South Africa)Capital account reversal defined as the change in the capital account as a percentageof GDP (y/y)Combined reversal defined as 1st year plus the cumulative 1st and 2nd year (total 2nd yearvs. baseline)

57. Reserves can also be compared against the full spectrum of capital account reversals (‘sudden stops’) experienced globally.Figure III.4 compares reserves holdings (as a percentage of GDP, end-2006) in the Asian crisis countries against the percentiles of the capital account reversals experienced in a wide sample of countries (up to 150, depending on the year) over 1976-03, taken from Jeanne and Ranciere’s (2006) dataset. The extent of additional reserves accumulation required to cover the most extreme events would clearly be large for Indonesia (as for the other countries shown, with the exception of Malaysia). For instance, Indonesia’s reserves at end-2006 would be insufficient to cover around 4 percent of the 1-year capital account swings experienced globally between 1976 and 2003, and around 12 percent of the 2-year swings.26 This suggests that some additional reserves accumulation might be desirable. However, the cost of holding additional reserves (including sterilization costs and potential negative valuation effects) need to be balanced against the benefits. Other risk mitigation policies may be less costly, though only effective in the medium term: these could include policies to promote financial market liquidity (to minimize the effect of sudden capital outflows on asset prices and the exchange rate) and a heightened emphasis on financial sector surveillance (to minimize the risk of a sudden stop caused by a collapse in investor confidence).

Figure III.4.
Figure III.4.

Reserves Holdings and Capital Account Reversals

(In percent of GDP)

Citation: IMF Staff Country Reports 2007, 273; 10.5089/9781451818413.002.A003

Source: IMF, International Financial Statistics ; and Jeanne and Ranciere, 2006.

G. Conclusions

58.Indonesia has come a long way in overcoming macroeconomic vulnerabilities since 1997. Monetary and fiscal policies remain cautious, helping to reduce external and public debt levels and build up foreign exchange reserves. The exchange rate is more flexible, increasing the economy’s resilience and reducing the risk of a speculative attack. In the financial sector, the recapitalization and reorganization of the banking sector, greatly improved regulation and supervision, and less exposure to foreign exchange risk help to minimize risks. Broader institutional reforms help to reduce the risk of a repeat of the close political-corporate links that helped to create financial sector vulnerabilities in the 1990s.

59.Nevertheless, vulnerabilities have not been eliminated. Regional capital markets are more integrated, presenting opportunities for risk diversification but also increasing the potential for financial contagion. Recent large capital inflows are presenting challenges to policymakers. There are risks from a sharp tightening in global financial conditions, as for all countries, although there is no evidence that Indonesia faces a disproportionate risk (e.g. because of significant exchange rate misalignment or an asset price bubble). The build-up of reserve assets in recent years helps to insulate Indonesia against the risk of a sudden stop, although it should be recognized that these shocks can potentially be extremely large, even compared with Indonesia’s increasing stock of reserves.

References

  • Allen, Mark, Christoph Rosenberg, Christian Keller, Brad Setser, and Nouriel Roubini, 2002, “A Balance Sheet Approach to Financial Crisis.” IMF Working Paper WP/02/210. Washington, DC: International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • Baig, Taimur and Ilan Goldfajn, 1998. “Financial Market Contagion in the Asian Crisis.” IMF Working Paper WP/98/155. Washington, DC: International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • Bank for International Settlements (BIS), 1998, 68th Annual Report. Basle: Bank for International Settlements.

  • Berg, Andrew, 1999, “The Asia Crisis: Causes, Policy Responses, and Outcomes.” IMF Working Paper WP/99/138. Washington, DC: International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • Dornbusch, Rudiger, 2001, “A Primer on Emerging Market Crises,” NBER Working Paper 8326. Cambridge, Massachusetts: National Bureau of Economic Research.

    • Search Google Scholar
    • Export Citation
  • Finger, J.M. and M. E. Kreinin, 1979, “A Measure of ‘Export Similarity’ and Its Possible Uses,” The Economic Journal89, pp. 905-912.

    • Search Google Scholar
    • Export Citation
  • Forbes, Kristin and Roberto Rigobon, 2002, “No Contagion, Only Interdependence: Measuring Stock Market Comovements.” The Journal of Finance 57 (5) pp. 2223-61.

    • Search Google Scholar
    • Export Citation
  • Goldstein, Morris, 1998, “The Asian Financial Crisis: Causes, Cures and Systemic Implications,” Policy Analyses in International Economics 55, June. Washington, DC: Institute for International Economics.

    • Search Google Scholar
    • Export Citation
  • Hill, Hal and Takashi Shiraishi, 2007. “Indonesia After the Asian Crisis.” Asian Economic Policy Review 2, pp.123-141.

  • International Monetary Fund, 2007. “Indonesia: Staff Report.” Washington, DC: International Monetary Fund.

  • International Monetary Fund, 2006. “Corporate Sector in Indonesia: Financial Performance and Underlying Vulnerabilities.” In International Monetary Fund, 2006, “Indonesia: Selected Issues.” (Washington, DC).

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, 2004. “Assessing Indonesia&rsqus Banking Sector Reforms.” In International Monetary Fund, 2004, “Indonesia: Selected Issues.” (Washington, DC)

    • Search Google Scholar
    • Export Citation
  • Jeanne, Olivier and Romain Ranciere, 2006. “The Optimal Level of International Reserves for Emerging Market Countries: Formulas and Applications.” IMF Working Paper WP/06/229. (Washington, DC: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Krugman, Paul, 1999, “Balance Sheets, The Transfer Problem and Financial Crises,” (mimeo, MIT, January 1999, http://web.mit.edu/krugman/www/FLOOD.pdf).

    • Search Google Scholar
    • Export Citation
  • Nasution, Anwar, 2007. “Comment on ‘Are Banking Systems in East Asia Stronger?’” Asian Economic Policy Review 2. pp. 96-97.

  • Pangestu, Mari and Manggi Habir (2002), “The Boom, Bust and Restructuring of Indonesian Banks,” IMF Working Paper 02/66. (Washington, DC: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Reinhart, Carmen and Kenneth Rogoff, 2004. “The Modern History of Exchange Rate Arrangements: A Reinterpretation.” Quarterly Journal of Economics 119 (1), pp. 1-48.

    • Search Google Scholar
    • Export Citation
  • Roubini, Nouriel, 2007. “Asia is Learning the Wrong Lessons from its 1997-98 Financial Crisis: The Rising Risks of a New and Different Type of Financial Crisis in Asia.” New York: New York University, Mimeo. First Draft, May 2007.

    • Search Google Scholar
    • Export Citation
  • Turner, Philip, 2007. “Are Banking Systems in East Asia Stronger?” Asian Economic Policy Review 2, pp. 75-95.

  • Wyplosz, Charles, 2007. “Is East Asia Safe from Financial Crises?” HEI Working Paper No. 2/2007. Geneva: Graduate Institute of International Studies.

    • Search Google Scholar
    • Export Citation
9

Prepared by Christopher Crowe (ext. 35303), with thanks to Agnes Isnawangsih, Armando Morales, Romaine Ranciere and Wiwit Widyastuti for help with various data requests.

10

According to the Reinhart and Rogoff (2004) de facto classification, Indonesia’s and Korea’s exchange rate regimes at end-1996 can be classified as crawling pegs, Malaysia’s as a crawling band (with a tolerance zone of less than 2 percent around the central peg), while the Philippines and Thailand followed fixed pegs.

11

Source: CEIC.

12

Substantial financial misreporting means that the pre-crisis NPL measures should be interpreted cautiously; they are almost certainly underestimates (so that the real improvement in credit quality is higher than that implied by the figures presented here).

13

For instance, the Economist Intelligence Unit’s assessment of the Indonesian banking sector in the first quarter of 1997 noted that “Although vulnerable on a number of counts—its poor asset quality, its overexposure to the property sector and its growing reliance on external funds—the sector is reasonably closely supervised.”

14

Domestic firms with more than 40 percent of equity held by a single family or group have a debt/equity ratio of 144 percent and a return on assets of 0.3 percent, compared with an average of 109 percent and 1.4 percent (respectively) in the sample as a whole (the top 100 nonfinancial firms listed on the Jakarta Stock Exchange, data for 2005; Source: Worldscope database).

15

However, Indonesia’s high Capital Asset Ratios are partly the result of the illiquidity of the recapitalization bonds on the banks’ balance sheets, which encourage holdings of additional liquid assets, primarily SBIs (Nasution, 2007).

16

Fitch’s Individual Ratings represent Fitch’s assessment of the probability of a bank requiring financial support. Ratings vary from A through E (where E represents the greatest risk). The number of banks in each sample is shown in parentheses. Turner (2007) presents the Fitch ratings for end-1998 and end-2005, in addition to ratings by Moody’s which show a very similar picture.

17

The comovement measure shown here is the unconditional correlation coefficient (Forbes and Rigobon, 2002) which attempts to control for the impact of heteroskedasticity (in periods of greater volatility, the correlation coefficient will tend to increase even if the underlying correlation structure does not change). The assumptions necessary for the unconditional correlation coefficient presented here to be unbiased are unlikely to be met; however, it is likely to be a less biased measure than the conditional correlation coefficient. The original data series are the daily stock market return (using the main stock market index for each country, in U.S. dollar terms). Contemporaneous correlations are obtained from the variance-covariance matrix of the errors from bivariate VARs run with 5 lags. Dummies for market closures in either market and returns on the New York market (current and lagged to 5 days) are included as controls (the latter controls for the influence of common shocks that influence both markets, which is distinct from spillovers from one market to the other). If p* is the conditional correlation coefficient between the shock to stock returns in country A (assumed the source of contagion) and country B (assumed the victim of contagion, in this case Indonesia) then the unconditional correlation coefficient is given by: ρ=ρ*1+(σA02σA121)(1ρ*2);where(σA02σA121)denotes the change in the variance of the stock return in country A between the (low-variance) pre-crisis period 0 and the high-variance crisis period 1. In the calculations presented here, for comovement with respect to Indonesia, country A is the country denoted in the first row of the table and country B is Indonesia; period 0 is 1995, while period 1 is either 1997-98 or 2005-06.

18

Following Finger and Kreinin, 1979, the Index of Export Similarity is defined as: S(ab,c)=100.ΣiMin.[Xi(ac),Xi(bc)]

where a and b are the two countries whose export patterns are being compared, c is the third market (in this case, the world), and Xi(ac) etc. is the share of good i in country a’s total exports to market c. Goods are defined using the 4-digit SITC code. The average is a simple unweighted mean.

19

For instance, according to the 1996 Corruptions Perceptions Index (CPI; Transparency International’s survey-based measure of perceived corruption in a cross-section of countries), Indonesia, the Philippines and Thailand were perceived to be more corrupt than average (Table III.14). In 1995 Indonesia topped the global survey.

20

The two sets of corruption indicators presented here give contrasting pictures for the recent period, suggesting that some caution is needed in interpreting the indicators. The difference reflects in part that the former shows relative global rankings, while the latter gives absolute scores. Hence, Indonesia’s score has improved recently in absolute terms, but is less impressive set against a downwards trend in perceived corruption globally. Moreover, the ICRG scores are more focused on perceived concerns for international investors, whereas the CPI has a broader civil society focus.

21

The nominal effective exchange rate has also been on an upwards trend since September 2005, appreciating by around 8 percent (by April 2007).

22

The weighted average price-equity (PE) ratio for April 2007 was around 18.2, compared with an average of 21.4 for the pre-crisis period (January 1992-July 1997; Data from CEIC). A fuller assessment of asset price sustainability would compare PE ratios with forecasts of future earnings, which is beyond the scope of the current chapter.

23

A sudden stop (defined as a substantial reversal of the capital account), from the basic Balance of Payments identity, must be covered by a combination of a movement toward surplus in the current account and the rundown of reserves. To the extent that reserves depletion cannot cover the capital account reversal, then the current account must move into surplus, generally requiring a significant real exchange rate depreciation as well as a contraction in domestic demand (this is the ‘transfer problem’ highlighted by Krugman, 1999). Passthrough to inflation means that the nominal depreciation required for a given real depreciation is significant, while lower domestic demand implies a substantial recession which impacts on firm solvency. These heightened currency and solvency risks exacerbate the balance sheet problems highlighted in Section C.

24

Jeanne and Ranciere (2006) calibrate an optimal reserves holding model and find that the optimal level of reserves is close to that implied by the Greenspan-Guidotti rule. However, their model differs from Wyplosz’s in assuming exogenous crises (with reserves having a mitigating role on the crisis’s effect on consumption), whereas Wyplosz’s model is essentially a bank run model with endogenous crises potentially occuring when reserves are below a threshold. This threshold can be extremely high.

25

The crisis years differ for the non-Asian comparators (see notes accompanying the table).

26

The sample of countries includes some very small and open economies which typically experience much greater capital account swings; hence, a relatively large and closed economy such as Indonesia is probably less likely to experience a shock of a given high magnitude than the probability implied by the full distribution of capital account reversals.

Indonesia: Selected Issues
Author: International Monetary Fund
  • View in gallery

    Selected Countries: Real Effective Exchange Rates

    (January 1990 = 100)

  • View in gallery

    Selected Countries: Real Effective Exchange Rate

    (June 2001 = 100)

  • View in gallery

    Figure III.3. Selected Economies: Stock Market Indices

    (October 2002 = 100)

  • View in gallery

    Reserves Holdings and Capital Account Reversals

    (In percent of GDP)