Information about Asia and the Pacific Asia y el Pacífico
Journal Issue
Share
Article

Indonesia: Selected Issues

Author(s):
International Monetary Fund
Published Date:
October 2011
Share
  • ShareShare
Information about Asia and the Pacific Asia y el Pacífico
Show Summary Details

IV. Cross-Border Financial Linkages and Spillovers to Indonesia1

Over the last decade, Indonesia’s financial linkages to the rest of the world have become stronger and more diversified. This phenomenon theoretically benefits Indonesia through risk diversification of financing sources, but also increases its exposure to systemic risks. In this regard, Indonesia was greatly impacted by the global crisis in late 2008, but has improved its buffers and fundamentals since then. This paper focuses on the evidence of financial spillovers in a more recent period. The recent market turmoil in 2010 appears to have been systemic but had a short-lived effect on Indonesia. Policy announcements in the United States, including quantitative easing, do not appear to have significant effects on the correlations between Indonesian and U.S. asset prices. Euro area policy announcements, however, appear to have tightened the link between core euro area and Indonesian sovereign risk. Such evidence shows that despite the relative stability in Indonesia’s financial markets since early 2010, potential spillovers from global events could still be significant.

A. Cross-Border Financial Linkages

1. Indonesia’s international investment position has changed significantly over the last decade, both in level and composition. After the crisis in the late 1990s, Indonesian banks and corporates as well as the public sector have deleveraged over time, resulting in declining external liabilities to the rest of the world. Indeed, Indonesia’s net foreign liabilities (NFLs) moved from 70 percent of GDP in 2001 to 40 percent of GDP in 2009. It was achieved largely through reduction in liabilities, as foreign assets accumulation had not kept up with economic expansion over the years. Reduction in other investment liabilities—mainly foreign loans—was responsible for most of the NFL dynamic.2

Other ASEAN-4 countries experienced similar (and even larger) reductions in their NFLs. Compared to Indonesia, Malaysia and Thailand reduced their NFL position mainly through foreign asset accumulation in reserves assets, FDI, and portfolio assets while the Philippines’s experience was similar to Indonesia’s. Indeed, Indonesia’s reduction in external loans was the highest among the four countries while portfolio and FDI liabilities picked up by about 10 percent of GDP each. ASEAN-4 countries’ composition of foreign liabilities have become more balanced between the three types of liabilities, unlike in the period right after the Asian crisis where foreign loans represented more than 50 percent of total foreign liabilities.

Figure IV.1.ASEAN-4: Foreign Assets

(In percent of GDP)

Source: IMF, Balance of Payments Statistics.

Figure IV.2.ASEAN-4: Foreign Liabilities

(In percent of GDP)

Source: IMF, Balance of Payments Statistics.

Figure IV.3.ASEAN-4: Net Foreign Assets

(In percent of GDP)

Source: IMF, Balance of Payments Statistics.

Figure IV.4.Indonesia: Foreign Liabilties by Type

(In percent of GDP)

Source: IMF, Balance of Payments Statistics.

Figure IV.5.ASEAN-4: Composition of Foreign Liabilities, 2001

(In percent of total)

Source: IMF, Balance of Payments Statistics.

Figure IV.6.ASEAN-4: Composition of Foreign Liabilities, 2009

(In percent of total)

Source: IMF, Balance of Payments Statistics.

Figure IV.7.Cross-Border Bank Claims Between Selected Advanced Countries and Asia, 1999 and 2009 1/

Sources: Bank for International Settlements; and IMF staff calculations.

1/ The Figures show foreign bank claims among a subset of advanced economies and Asian countries in 1999 and 2009. The origin of the arrows indicates the country of origin of the banks holding the claims, while the arrows’ thickness is proportional to the size of the claims scaled by the recipient’s GDP.

3. The geographical landscape of Indonesia’s financial linkages to the rest of the world has also changed over time. Similar to other ASEAN-4 economies, foreign loans from the 1990s, mainly funded by Japanese banks, have been unwound. In addition, Indonesia’s source of bank and portfolio funding has become more diversified in terms of country sources.3

4. Within a smaller bank lending pie, European banks have increased their share at the expense of Japanese banks. Claims by banks headquartered in Europe increased from almost half of total claims to more than 60 percent. German and Italian banks in particular experienced the largest increase. This trend of European banks becoming more important in bank funding has also been observed in other ASEAN-4 countries.

Figure IV.8.Foreign Bank Claims on Indonesia, 2001

Sources: Bank for International Settlements.

Figure IV.9.Foreign Bank Claims on Indonesia, 2009

Sources: Bank for International Settlements.

5. While bank lending has become less important, portfolio financing has picked up. Both portfolio debt and equity claims on Indonesia have exhibited significant growth from 2001 to 2009, albeit from small bases.4 With strong capital inflows towards rupiah debt instruments in 2010–11, portfolio debt claims on public sector instruments alone are already about 6 percent of GDP. Only Malaysia seemed to have experienced a similar increase in portfolio debt claims, while all ASEAN-4 countries saw strong growth in portfolio equity claims.

Figure IV.10.Portofolio Debt Claims

(In percent of GDP)

Sources: IMF, IMF, Coordinated Portfolio Investment Survey; and IMF, World Economic Outlook.

Figure IV.11.Portofolio Equity Claims

(In percent of GDP)

Sources: IMF, IMF, Coordinated Portfolio Investment Survey; and IMF, World Economic Outlook.

6. Similar to bank funding, the composition of portfolio claims on Indonesia has also become more diversified. On the portfolio debt side, the United States and Singapore stayed as top investors, together accounting for 55 percent of total investment. However, while the bulk of the claims remained within G-7 countries, there has been a significant increase in investment from small European financial centers such as Luxembourg, the Netherlands, and Ireland. This trend was also seen in other ASEAN-4 countries. This may suggest a larger pool of advanced countries’ investors (and funds) diversifying their portfolio into emerging markets, including Indonesia and its neighbors. Portfolio debt holdings within emerging Asian countries, however, have not increased. Similar trends were observed for portfolio equity claims.

Figure IV.12.Portfolio Debt Claims on Indonesia, 2001

Sources: IMF, Coordinated Portfolio Investment Survey.

Figure IV.13.Portfolio Debt Claims on Indonesia, 2009

Sources: IMF, Coordinated Portfolio Investment Survey.

Figure IV.14.Portfolio Equity Claims on Indonesia, 2001

Sources: IMF, Coordinated Portfolio Investment Survey.

Figure IV.15.Portfolio Equity Claims on Indonesia, 2009

Sources: IMF, Coordinated Portfolio Investment Survey.

B. More Diversified = Less Vulnerable?

7. There are two sides to the diversification of the sources of funding. On the one hand, having more sources of funding helps diversify risks from source countries. It could also help reduce concentration risks which arise when recipient countries have unusually large concentration of exposures to only a few sources. If a large shock hit a main funding source, the recipient country could experience a more severe cross-border deleveraging. On the other hand, if the new sources of funding are also highly connected to other “core” source countries, the benefit from risk diversification may not be high and the recipient country may become more vulnerable to systemic risks.

8. Insight from network theory indeed suggests that financial linkages are a double-edged sword. As explained in IMF (2011a), financial interconnectedness has the potential of making the network (of source and recipient countries linked by financial claims) more robust (through risk diversification) but it may raise the network’s fragility to systemic breakdowns. If a localized shock hit a single source country, the resulting cross-border deleveraging can be manageable for the recipient country. However, if a highly interconnected source country is hit by a large shock, the shock can be propagated widely via its linkages to the rest of the network. In practice, financial shock transmission can be seen through comovement of financial and risk indicators as well as evidence of shift in exposures to different markets.

C. Recent Evidence of Financial Spillovers to Indonesia

9. Empirical evidence from the Lehman crisis period suggests that spillovers from the external environment were important. In a spreads model incorporating external financial, as well as domestic macroeconomic, financial and political variables, Goyal and Ruiz-Arranz (2009) show that external factors, including global risk aversion and international liquidity, accounted for over 50 percent of the increase in Indonesia’s EMBI spreads during 2009. The model fit the increase in Indonesia’s spreads well.

10. Indonesia’s external financial conditions have improved markedly since 2009. Improved fundamentals, including political stability, a stronger external position and robust growth prospects, have helped lower market perceptions of Indonesian risks. As of August 2011, despite a renewal of global risk aversion and a subdued growth outlook in the advanced economies EMBI and CDS spreads remained below their 2010 peaks during the Greek turmoil around April to June 2010. Simple charts do show evidence of spillovers from the euro area debt crisis. Figure IV.16 shows spikes in Indonesia’s EMBI and CDS spreads during the 2010 turmoil, similar to spikes in other large EMs. More recently, a shift towards global risk aversion in July-August 2011 has also translated into increases in spreads across EMs and falls in equity prices.

Figure IV.16.Indonesia and Selected Countries: Recent Financial Indicators

Sources: Bloomberg L.P.; Haver Analytics; Datastream; and staff calculations.

11. Simple correlation plots show evidence of spillovers from the European periphery distress in 2010. Asset price comovement is one indicator of financial spillovers. In its simplest form (Figure IV.17), Indonesia’s equity prices appear less correlated with the U.S. equity price indices compared to Turkey or Brazil. Spillovers from the euro area debt crisis were observed through correlation with Greece-Ireland-Portugal (GIP) CDS spreads and more recently, German CDS spreads. The period of high correlation appear to last for about three months in 2010.

Figure IV.17.Selected Emerging Markets: Simple Correlation

Sources: Bloomberg L.P.; Haver Analytics; Datastream; and staff calculations.

12. Results from a dynamic conditional correlation model suggest that spillovers from the euro area problems in 2010 were short-lived. To correct a potential bias in the presence of time-varying volatility, a dynamic conditional correlation GARCH model is used to infer correlations between Indonesian and other market indicators. Results from the model show a more stable (with movements being short spikes) relationship between Indonesian, German and GIP CDS spreads (Figure IV.18).5 The same holds for other EMs in the sample. The results also show that on average, Indonesia and other EMs are more linked to Germany (representing the core euro area) than the European periphery.

Figure IV.18.Implied Correlation

Source: IMF staff estimates.

13. Indonesia’s equity prices also appear to be less correlated with global risk aversion, compared to Brazil or Turkey. Both simple correlation and implied correlation from the model suggest that Indonesia exhibited a smaller degree of correlation with VIX, an index widely used to capture global risk aversion.

Figure IV.19.Selected Emerging Markets: Simple Correlation with VIX

(60-day rolling correlation of daily percentage changes)

Sources: Bloomberg L.P.; and IMF staff calculations.

Figure IV.20.Implied Correlation Between Jakarta Stock Indices and VIX

Source: IMF staff estimates.

14. A model employing regime switching techniques confirms the presence of strong spillovers post Lehman and a modest one associated with euro area debt crisis in 2010.6 The model basically identifies when the market conditions signal a regime change from tranquil periods to medium or high volatility states. As expected, the results using Indonesian CDS spreads and equity prices show a long period of high volatility from late 2008 to mid 2009. The spike in volatility associated with the 2010 Greek turmoil was rather short-lived and the markets returned to a tranquil state quickly thereafter.

Figure IV.21.Indonesian Sovereign CDS: State of Volatility 1/

Sources: Bloomberg L.P.; and staff estimates.

1/ Probability (from 0 to 1) of being in the low, medium or high state of volatility on the y axis.

Figure IV.22.Jakarta Stock Index: State of Volatility 1/

Sources: Bloomberg L.P.; and staff estimates.

1/ Probability (from 0 to 1) of being in the low, medium or high state of volatility on the y axis.

15. Spillovers that originated from core countries’ policies are also investigated. The impact of G-3 policy on others is difficult to measure in general. However, looking at the responses of key financial market prices to changes in asset prices in the G-3 countries around the policy announcement days is one way of capturing market perception of the impact of the announced policy. Following Bayoumi and Bui (2011), event studies are used to analyze the impact of U.S. fiscal and monetary policies on Indonesian asset prices.7 In particular, yield (external bond yield) and equity price spillovers are analyzed on days of significant U.S. policy announcements and checked to see whether such announcements change the typical bilateral relationship between the U.S. and Indonesian yields and equity prices.

16. Results show some impact of U.S. policy announcements on Indonesian equity prices but no significant impact on bond yields. Controlling for other global and domestic conditions, Indonesian equity prices are significantly and positively linked to U.S. equity prices. The relationship, however, has become less strong post 2007. The U.S. fiscal stimulus package in 2008 appears to have increased the link but the 2009 and 2010 package shows the opposite effect and no effect respectively. The results are consistent with Bayoumi and Bui (2011) which does not find systematic impact on the correlations between the U.S. and foreign markets. Contrary to Bayoumi and Bui (2011) findings that bond yields in emerging markets and advanced countries in the G-20 appear to have strong correlations with U.S. treasury yields, Indonesia’s link is not statistically significant. In addition, quantitative easing (both QE1 and QE2 announcements) appears to have had no significant impact on equity price and yield correlations between Indonesia and the United States.

17. Recent Euro Area policy announcements appear to have statistically significant impacts on Indonesia’s sovereign risk. Controlling for other global conditions, Indonesian CDS spreads are significantly and positively linked to German CDS spreads over the period 2003 to mid-2011. Similar to other Asian countries, Indonesia seems to have delinked from Germany starting from the beginning of the distress in the euro area periphery. Announcements of IMF-EU programs in Greece, Ireland and Portugal, however, appear to bring back the link between the two CDS spreads, suggesting that Indonesia as well as other emerging Asian economies remain exposed to euro area debt crisis developments. The results are consistent with the recent rise of Indonesian CDS spreads in August and September 2011.

Figure IV.23.Estimated Relationship between Asian Countries’ Sovereign CDS spreads and German CDS spreads 1/

Source: IMF staff estimates.

1/ The estimated equation is based on daily change in 5-year CDS spreads in bps from 2003 to April 2011.

18. These results confirm the findings that an intensification of the euro area debt crisis could have major global consequences and thus large spillovers to Indonesia.8 Small real and financial linkages (including direct banking and portfolio linkages) between Indonesia and the European periphery as well as strong growth prospects and improved domestic fundamentals have kept spillovers short-lived and contained within the financial markets so far. If the core euro area is affected, spillovers could already be larger due to a larger historical correlation with the core. Deleveraging by core euro area countries through the banking and portfolio linkages could be significant, with further deleveraging by others possible if global risk appetite disappears.

D. Conclusion

19. Indonesia has become more financially interconnected over the last decade. Indonesia’s sources of funding have broadened both in a geographical sense and in the type of funding. This phenomenon in theory should add the benefit of risk diversification, but it could also increase Indonesia’s exposure to systemic shocks.

20. There is evidence that spillovers from the recent market turmoil in 2010 have been short-lived. Distress in the euro area periphery in 2010 through mid 2011 did affect Indonesian financial indicators; however the effects were not lasting. This evidence is shown through simple correlation, implied correlation and empirical identification of high volatility periods. Indeed, Indonesia’s growth prospects and improved fundamentals as well as search for yield in emerging markets have led to subdued yields and risk indicators up to August 2011.

21. Further distress in the core euro area could have significant spillovers to Indonesia. Evidence from the event studies show that the comovement of sovereign risk measure has become tighter between the core euro area and Indonesia during important policy announcements related to the euro area debt crisis. This suggests that spillovers from the euro area debt crisis could be large as markets become more linked during a more distressed period. Such spillovers can already be identified with the recent increases in Indonesian risk indicators in August and September 2011.

References

    BayoumiT. and T.Bui2011Unforeseen Events Wait Lurking: Estimating Policy Spillovers from U.S. to Foreign Asset Prices,IMF Working Paper 11/183 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    FrankN. and H.Hesse2009Financial Spillovers to Emerging Markets during the Global Financial Crisis,IMF Working Paper 09/104 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    González-HermosilloB. and H.Hesse2009Global Market Conditions and Systemic Risk,IMF Working Paper 09/230 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    GoyalR. and M.Ruiz-Arranz2009Estimating Indonesia’s Sovereign Spreads,” in Indonesia: Selected Issues IMF Country Report No. 09/231 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    International Monetary Fund2011aMapping Cross-Border Financial Linkages: A Supporting Case for Global Financial Safety Nets,IMF Policy Paper. Available via the Internet: http://www.imf.org/external/np/pp/eng/2011/060111.pdf.

    • Search Google Scholar
    • Export Citation

    International Monetary Fund2011bEuro Area Policies: Spillover Report for the 2011 Article IV Consultation and Selected Issues IMF Country Report No. 11/185 (Washington).

    • Search Google Scholar
    • Export Citation
1

Prepared by Mali Chivakul with contributions from Heiko Hesse and Trung Bui.

2

Owing to data limitations, the data on international investment position for Indonesia is only up to 2009. During 2010–11, Indonesia also had a significant increase in its net foreign assets.

3

This chapter does not consider foreign direct investment (FDI), an investment class generally viewed as relatively stable and driven by longer-term considerations. One caveat is that the increased use of special purpose vehicles and other financial conduits by direct investors may suggest that not all FDI may be as stable as normally held.

4

CPIS data are only available through end-2009.

5

The model setup can be found in Frank and Hesse (2009). The estimates were done through April 2011.

6

The model setup can be found in Gonzalez-Hermosillo and Hesse (2009). Data used for estimation start in 2003 for equity prices and 2004 for CDS.

7

In Bayoumi and Bui (2011), Indonesia is included in the estimates, but grouped with India and Russia. Here Indonesia is analyzed separately. The impact of euro area policy announcements are studied separately but follow similar specifications.

Other Resources Citing This Publication