56790 INDONESIA ECONOMIC QUARTERLY Looking forward September 2010 Preface The Indonesian Economic Quarterly reports on and synthesizes the past three months' key developments in Indonesia's economy. It places them in a longer-term and global context, and assesses the implications of these developments and other changes in policy for the outlook for Indonesia's economic and social welfare. Its coverage ranges from the macroeconomy to financial markets to indicators of human welfare and development. It is intended for a wide audience, including policy makers, business leaders, financial market participants, and the community of analysts and professionals engaged in Indonesia's evolving economy. This Indonesian Economic Quarterly was prepared and compiled by the macroeconomic analysis team at the World Bank's Jakarta office, under the guidance of Lead Economist Shubham Chaudhuri and Senior Country Economist Enrique Blanco Armas: Magda Adriani (rice prices), Andrew Blackman (trade flows, balance of payments), Andrew Carter (government revenues), Andrew Ceber (national accounts, domestic demand, medium term macroeconomic projections, executive summary), Fitria Fitrani (trade flows and ASEAN-China FTA), Faya Hayati (prices and real incomes), Ahya Ihsan (government expenditure), Diva Singh (financial markets, monetary conditions, banking sector). Banking sector input was also provided by Neni Lestari and P.S. Srinivas. Andrew Ceber, Tia Chandra, Ashley Taylor and Tim Bulman shared the editing and production. Enrique Blanco Armas provided detailed comments on earlier drafts. For more World Bank analys is of Indonesia's economy: For information about the World Bank and its activities in Indonesia, please visit www.worldbank.org/id In order to be included on an email distribution list for this Quarterly series and related publications, please contact madriani@worldbank.org. For questions and comments relating to this publication, please contact tbulman@worldbank.org. iii Table of contents Preface iii Executive Summary: Looking forward viii A . ECONOMIC AND FISCAL UPDATE 1 1. Indonesia's robust recent growth performance is expected to continue 1 a. Private demand accelerated in the first half of 2010, offsetting a slowdown in government spending .......................................................................................................................................... 1 b. Import growth continues to outpace that of exports .................................................................... 2 c. Indonesia's economic growth is likely to rise further .................................................................. 3 2. The balance of payments surplus was near historic highs in Q2, but future narrowing is expected 5 3. Indonesian financial markets strengthened with the return of strong capital inflows 5 a. Indonesian asset prices strengthened in line with renewed capital inflows .............................. 5 b. With money supply growth and inflation picking up, BI has raised the statutory reserve requirement for commercial banks ................................................................................................ 7 c. Credit growth increased significantly in Q2, boosted by working capital and investment loans ................................................................................................................................................. 8 4. Core inflation continues to recover while headline inflation accelerated sharply due to temporary food supply shocks 10 a. CPI inflation rose in mid-2010 because of weather-related supply disruptions ...................... 10 b. The inflation outlook for 2010 remains unchanged, but stronger price growth is expected in 2011 ................................................................................................................................................ 11 5. Poverty continues to trend downwards 13 6. Reallocating and improving the quality of fiscal spending is on the Government's agenda 14 a. The government proposes a reallocation of expenditures in the 2011 budget........................ 14 b. Slow disbursement rates have continued through 2010 ........................................................... 15 c. World Bank projections are for the budget deficit for 2010 to be slightly lower than previously anticipated ................................................................................................................... 15 7. The focus of risks is shifting towards the medium-term 19 a. If demand exceeds supply capacity there may be a reemergence of inflation ........................ 19 b. There is a risk that Indonesia's economy will not grow as quickly as expected if the reform agenda is slowed ........................................................................................................................... 20 B. SOME RECENT DEVELOPMENTS IN INDONESIA'S ECONOMY 21 1. Update on the ASEAN-China Free Trade Agreement 21 a. The implementation of ACFTA tariff reductions ......................................................................... 21 b. The rising importance of China in Indonesia's imports reflects a range of factors ................ 22 c. ...but initial statistical analysis does not support the arguments that import growth from China has been impacted by the ACFTA tariff reductions......................................................... 23 d. Look at the ACFTA as the media to exchange market access .................................................. 25 2. Piecing together insights into the movements in household purchasing power 26 a. During the previous decade households experienced large increases in real consumption. 26 b. Consumer confidence indices and inflation measures provide real-time insights into developments in real consumption ............................................................................................. 27 c. Recent increases in some volatile food items particularly affected the purchasing power of poor and rural households ........................................................................................................... 29 3. Medium-term macroeconomic projections 30 a. Drivers of potential growth in the economy ................................................................................ 30 b. Growth accounting revisited ........................................................................................................ 31 c. Baseline projections ..................................................................................................................... 32 d. Advantages, disadvantages and further work ............................................................................ 33 e. Implications and conclusions ...................................................................................................... 34 C. INDONESIA 2014 AND BEYOND: A SELECTIVE LOOK 35 1. Improving Access to Financial Services: Analysis and ideas for policy-makers 35 a. What is access to finance/financial inclusion and where does Indonesia stand? .................. 35 b. Demand-side aspects: what products and services do people need? What is currently available to them? ......................................................................................................................... 36 c. Supply-side Aspects: Who are the market players and what is their role in providing access to financial services? .................................................................................................................... 37 d. The role of ICT in improving access to finance .......................................................................... 38 e. Policies to promote access to finance ........................................................................................ 39 2. Labor Reform in Indonesia ­ Balancing Job Creation and Employee Protection 40 3. Education, training and labor market outcomes for youth in Indonesia 44 a. Economic context: growth, transformation and the demand for skills .................................... 44 b. A difficult transition to the labor market ..................................................................................... 45 c. Senior Secondary Education, Skills and Labor Market Entrance ............................................. 46 4. PNPM Generasi final impact evaluation ­ the preliminary findings 49 5. Lumpy, evolving and constrained: Regional and urban economic development in Indonesia 51 APPENDIX: SNAPSHOT OF THE INDONESIAN ECONOMY 55 LIST OF FIGURES Figure 1: Quarterly GDP growth remains strong ................................................................................ 1 Figure 2: Strong trading partner growth dipped slightly in Q2 ......................................................... 1 Figure 3: Private consumption continues to drive growth ................................................................ 2 Figure 4: `Non-tradable' growth expands ahead of the growth of `tradables'.................................. 2 Figure 5: Imports values are growing strongly... ............................................................................... 3 Figure 6: ...while exports values continue to stagnate...................................................................... 3 Figure 7: Large recent foreign capital inflows were seen across asset classes ............................. 6 Figure 8: ...contributing to the appreciation of the Rupiah to a 3-year high against the USD in early August.................................................................................................................... 6 Figure 9: Net foreign purchases of SBIs fell sharply in July after BI's new measures, but climbed again in August... ............................................................................................ 7 Figure 10: ...while foreign investment in short-dated SUNs increased after BI's June measures 7 Figure 11: Base money increased from June to August ................................................................... 8 Figure 12: ...and M1 and M2 growth rates also picked up pace in Q2 after an unusually sluggish start to the year .............................................................................................................. 8 Figure 13: Quarterly credit growth and new loan approvals rose sharply in Q2... ......................... 8 Figure 14: ... driven by pick ups in working capital and investment loan growth........................... 8 Figure 15: Banks' LDRs rise considerably if recap bonds and SUNs are included on the asset side.................................................................................................................................. 9 Figure 16: ...and the recent sharp rise in undisbursed loans also impacts their liquidity ............. 9 Figure 17: Inflation has risen over the year but remains below historic averages........................ 10 Figure 18: Volatile and administered items drove the recent spike in headline inflation with core inflation rising steadily ................................................................................................ 10 Figure 19: Domestic and International Rice Price Comparison ...................................................... 12 Figure 20: Disbursements rates remain low in 2010 ....................................................................... 15 Figure 21. Indonesia's tariffs are low for both household... ........................................................... 18 Figure 22. ... and industrial users, even after the 2010 increase .................................................... 18 Figure 23: Scenario analysis of risks to GDP growth in 2010 and 2011 ......................................... 20 Figure 24: Indonesia's tariff rates on imported goods by trade agreement, simple average ....... 22 Figure 25: Share of Indonesia's non-oil and gas imports ................................................................ 22 Figure 26: Real consumption growth 2001-2009 .............................................................................. 26 Figure 27: Consumer confidence for the highest expenditure group broadly tracks their real consumption growth... ................................................................................................ 27 Figure 28: ... similarly for poor households ..................................................................................... 27 Figure 29: Consumer confidence measures fell in recent months... ............................................. 28 Figure 30: ...as inflation increased .................................................................................................... 28 Figure 31: Poorer households' consumer confidence is lower and fell more severely with recent price increases ............................................................................................................. 29 Figure 32: Rural households also have lower consumer confidence , again falling sharply during recent price rises.............................................................................................. 29 Figure 33: Potential GDP versus actual growth growth since 1980................................................ 32 Figure 34: Projections of Indonesia's potential growth ................................................................... 33 Figure 35: Share of the population with formal financial access .................................................... 35 Figure 36: Savers' financial inclusion in Indonesia ......................................................................... 36 Figure 37: Borrower's financial inclusion in Indonesia ................................................................... 36 Figure 38: Take-up of insurance is highly geared towards salaried employees ........................... 37 Figure 39: MSME lending relative to total loans ............................................................................... 37 Figure 40: The Manpower Law made it more costly for employers to terminate or reallocate employees ..................................................................................................................... 40 Figure 41: Indonesia's firing costs are high relative to regional comparators .............................. 40 Figure 42: Severance regulations have not been effective in protecting employees ................... 41 Figure 43: ...especially those who are more in need of protection ................................................ 41 Figure 44: Rising real income in Indonesia has been accompanied by a fall in the share of agricultural output ........................................................................................................ 44 Figure 45: The returns to education have remained largely flat over the past decade ................. 45 Figure 46: The share of skilled jobs has been constant during the past decade after rising in the 1990s ............................................................................................................................. 45 Figure 47: Unemployment rate by age group and education level ................................................. 46 Figure 48: Share of labor force in salaried jobs by age group and education level ...................... 46 Figure 49: Employers' opinion of the quality of employees with senior secondary education ... 46 Figure 50: Unemployment rates for SMA and SMK graduates, age 20-24 (1991-2007) ................. 47 Figure 51: Urbanization and economic development go hand-in-hand ......................................... 51 Figure 52: ...Indonesia also follows this trend ................................................................................. 51 Figure 53: UN Urban and rural population projections for Indonesia, 2010-2050 ......................... 52 LIST OF APPENDIX FIGURES Figure 1:GDP growth .......................................................................................................................... 55 Figure 2: Contributions to GDP(E) ..................................................................................................... 55 Figure 3: Contributions to GDP(P) ..................................................................................................... 55 Figure 4: Motor cycle and motor vehicle sales ................................................................................. 55 Figure 5: Consumer indicators .......................................................................................................... 55 Figure 6: Industrial partial indicators ................................................................................................ 55 Figure 7: Real trade flows................................................................................................................... 56 Figure 8: Balance of Payments .......................................................................................................... 56 Figure 9: Trade balance ...................................................................................................................... 56 Figure 10: International reserves ....................................................................................................... 56 Figure 11: Terms of trade and monthly export and import chained Fisher Price indices ............ 56 Figure 12: Inflation .............................................................................................................................. 56 Figure 13: Monthly breakdown of CPI ............................................................................................... 57 Figure 14: Inflation amongst neighboring countries........................................................................ 57 Figure 15: Sub-national government bank deposits since decentralization .................................. 57 Figure 16: Poverty, employment, and unemployment rates ............................................................ 57 Figure 17: Regional equity indices .................................................................................................... 57 Figure 18: Broad USD Index and Rupiah spot .................................................................................. 57 Figure 19: 5 Year local currency bond yields ................................................................................... 58 Figure 20: Sovereign USD bond EMBI spreads ................................................................................ 58 Figure 21: International commercial bank lending ........................................................................... 58 Figure 22: Banking sector financial indicators ................................................................................. 58 LIST OF TABLES Table 1: The outlook remains for robust growth in 2010 and 2011................................................ viii Table 2: Domestic demand is projected to continue to support a gradual rise in Indonesia's growth ............................................................................................................................. 4 Table 3: The BoP surplus is projected to reach record highs in 2010, then to narrow in 2011 ...... 5 Table 4: The importance of food and rice in per capita expenditures ............................................ 12 Table 5: A strong rise in the allocation to capital expenditure is proposed for the 2011 budget 15 Table 6: The Government of Indonesia's financing position for 2010 is favorable ....................... 17 Table 7: The tariff increase varies significantly across different customer groups, with most customers experiencing no increase ......................................................................... 18 Table 8: ACFTA tariff reductions and imports from China as a share of Indonesia's total imports by broad economic classification and significant products .................................... 23 Table 9: Estimated impact of tariff changes on the growth of Indonesia's imports by product line ................................................................................................................................. 24 Table 10: ACFTA tariff reductions and exports to China as a share of Indonesia's total exports by broad economic classification and significant products .................................... 25 Table 11: Sensitivity analysis ............................................................................................................. 33 Table 12: Indonesia's urban concentration is set to spread ........................................................... 52 Table 13: Regional GDP distribution by major urban agglomerations, other areas and rural areas, 1993-2006 ........................................................................................................... 53 Table 14: Budget outcomes and estimates ....................................................................................... 59 Table 15: Balance of Payments .......................................................................................................... 59 LIST OF BOXES Box 1: Recent rice price developments ............................................................................................ 11 Box 2: The July 2010 electricity tariff adjustment: a modest first step to better targeted energy subsidies at a lower burden on the budget................................................................ 17 Box 3: Focus on Micro, Small and Medium Enterprises (MSMEs) .................................................. 37 Box 4: Severance Reform Options .................................................................................................... 43 Box 5: PNPM Generasi target indicators........................................................................................... 49 Executive Summary: Looking forward Indonesia's robust Indonesia's economy continues to record robust growth, in contrast with the volatility and growth continues, with uncertainty characterizing major economies globally. The robustness of growth has the medium-term allowed the policy focus to shift from near-term uncertainty towards achieving the challenge being to investments and reform required to achieve sustained and strong growth over the longer- sustain and strengthen this performance term. Meeting the Government's target of over 7 percent growth by 2014 requires strong rises in investment, particularly in infrastructure, and in skills and productivity. Domestic demand Quarterly output accelerated in Q2, resulting in year-on-year growth of 6.2 percent, the continues to underpin highest since the onset of the global economic crisis two years earlier. Domestic demand, Indonesia's growth particularly private consumption, underpins the growth performance and has been performance associated with rising imports, particularly for investment and intermediate goods. Slow disbursement of government expenditures continue to act as a drag on growth, but less so than in Q1. Indonesia's trading partners also recorded stronger growth than expected, although the overall contribution of net external demand to growth was negative in Q2. Correspondingly, domestic-oriented sectors outperformed externally-oriented sectors. In the near term these patterns of growth are broadly expected to continue. Rising domestic consumption and increasing private and public investment will contribute to a contraction in the current account surplus. Investment is expected to pick up further, with strong growth in credit for investment and working capital seen in recent months, and with a reallocation of public spending towards capital expenditures planned for 2011. The projection for GDP growth in 2010 has been revised up slightly from the June IEQ, reflecting the improved strength of domestically-oriented sectors, while the GDP forecast for 2011 remains unchanged at 6.2 percent. Table 1: The outlook remains for robust growth in 2010 and 2011 2009 2010 2011 Gross domestic product (Annual per cent change) 4.5 6.0 6.2 Consumer price index (Annual per cent change) 4.8 5.1 6.4 Budget balance* (Per cent of GDP) -1.6 -1.5 -1.7 Major trading partner growth (Annual per cent change) -0.8 6.5 4.3 * Ministry of Finance projection, 2011 figure is proposed budget Sources: Ministry of Finance, BPS and other national statistical agencies via CEIC, Consensus Forecasts Inc., and World Bank Temporary factors drove Core inflation has risen in recent months along with expanding domestic demand. By the rise in headline August the monthly rate had returned to its historical averages, although it is too early to inflation with increasing assess whether this will persist as incipient capacity constraints become binding. Headline local rice prices moving inflation, as is often the case, has been much more volatile, rising strongly in June through poverty basket inflation up sharply August, mainly due to one-off supply-side shocks, particularly the extremely wet `dry' season. This initially impacted chili prices especially, but soon also other food prices, particularly rice prices, which rose significantly on the local market, creating a widening gap with the international price. These increases notably affected poorer households, whose largest single expenditure item is rice. Poverty basket inflation reached 11.3 percent in August, around 5 percentage points higher than the headline rate. These disruptions are expected to be transitory, and to the extent they do not feed into inflation expectations should have limited impact on underlying inflationary pressures. But, in the short-term, such rises in the prices of goods and services consumed by the poor may offset some of the future poverty-reducing effects of Indonesia's robust growth performance, which contributed to the fall in the poverty rate to 13.3 percent in March 2010 from 14.2 percent a year earlier. Capital inflows have Confidence in the country's prospects, combined with high yields relative to developed returned strongly, markets, has seen the strong return of capital inflows. Around USD 7.3 billion of net creating a challenge for foreign capital inflows were seen from June to August. But, the USD 5.7 billion of outflows monetary policy in May highlight the continued sensitivity of such flows to both domestic and international shocks to sentiment. The volume of recent inflows has put renewed upward pressure on the Rupiah which, if realized, and combined with rising domestic prices, adds to the viii competitiveness pressures facing export-orientated sectors. Productivity-enhancing policy reforms and investments, such as in infrastructure, can help to meet such pressures. A policy challenge in the near-term is how to address rising inflationary pressures while ensuring less volatile and more sustainable capital inflows. In light of these issues, Bank Indonesia tightened reserve requirements while leaving the policy rate unchanged. Managing the volatility of short-term flows motivated policy measures introduced in June, such as a one month minimum holding period for SBI investments. The Government plans to The Government has reduced its projected 2010 deficit, to 1.5 percent of GDP, reflecting boost capital factors including modified assumptions and the weakness in disbursement early in the expenditure to finance year. In July 2010 the Government increased average electricity tariffs by 8 percent. This infrastructure adjustment is the most substantial reform to electricity tariffs since January 2003, and is a development and continue pro-poor social small step towards the Government's 2010-2014 RPJMN goal of improving the targeting programs of energy subsidies while minimizing their burden on the budget. Such reforms can free up further fiscal space for investment and social protection. The Government's proposed 2011 budget takes a new direction, allocating significantly more resources to capital investment (28 percent more than allocated in 2010), while improving the efficiency of subsidy outlays, most notably by proposing to reduce the large electricity subsidies. Overall the Government does not expect this reprioritization to affect its fiscal position, with the projected 2011 budget deficit near that of 2010. Given the strength of the nominal economy, this is likely to see public debt levels continue to fall towards 25 percent of GDP. Assuming stronger growth in the nominal economy, due to growth in economy-wide prices returning towards the average rates of the last half- decade, and only a modest improvement in disbursement on the rates achieved in the first half of 2010, would put downward pressure on the projected deficit (see Part B of the June 2010 IEQ for a full discussion). Near-term risks remain The overall risks to Indonesia's near-term outlook remain sizeable, although have sizeable but have diminished somewhat since the last Quarterly reflecting the continued robust economic diminished, while policy growth. The risks to the inflation outlook on the other hand have increased to the upside, challenges remain especially if continued strong domestic demand and transitory supply shocks translate into more persistent price pressures. The focus is now on the aforementioned policy challenges associated with the volatility of capital inflows. In the medium-term, as capacity constraints start to bind in some sectors, creating inflationary pressures, the challenge will be to ensure the growth in available infrastructure and skilled workers, and improvements in the business climate, to support expanding production, employment and incomes. Meeting the The Government faces some immediate priorities to support growth and to ensure that it Government's medium- benefits all Indonesians, especially the poorest. Addressing the constraints on private term targets for growth sector activity due to Indonesia's stretched infrastructure will require substantial public and poverty reduction investment, and the 2011 budget starts to reflect this. Infrastructure improvements will requires progress across a range of structural help to promote growth going forward by enhancing connectivity between economic reforms, supported by regions, fostering the formation of industrial clusters and centers of economic activity, and targeted expenditures improving the efficiency of the structure of Indonesia's urban areas. All Indonesians need access to opportunities to benefit from the higher growth flowing from these investments. For example, at present, about half of Indonesia's population does not have access to formal financial services. Improving access to financial services for households and firms can not only enhance their ability to invest in productive opportunities but also allow them to build up savings to smooth shocks to their real incomes. In relation to labor markets, policy makers face challenges in identifying which policies and programs will spur the creation of good jobs while, at the same time, ensuring that workers are better protected from the risks that threaten their job security. Reforms to improve labor regulations, infrastructure and the investment climate are all important elements for a national "pro-jobs" strategy. Reducing the mismatch between the skills of school leavers and the skills demanded by employers will also help increase employment, and raise productivity. Improving education and health outcomes in lagging areas, where the PNPM Generasi block grant scheme has shown strong results, will, in line with the Government's priorities and the Millennium Development Goals, also help further reduce poverty, maternal and child mortality, and promote universal access to basic education. A. ECONOMIC AND FISCAL UPDATE 1. Indonesia's robust recent growth performance is expected to continue Annual growth in Indonesia's economy continued to grow strongly in the first half of 2010, with year on year Indonesia has now growth returning to rates not seen until prior to the global instability of 2008. Growth in Q2 returned to rates not seen once again exceeded the past decade's average (Figure 1). Year on year growth rose to since 2008, and is 6.2 percent, up from 5.7 percent in Q1. Indonesia's economy is expected to expand at expected to remain there until the end of 2011 around these rates for the rest of the year, bringing annual growth to 6 percent in 2010, and to 6.2 percent in 2011. Growth in Indonesia's Growth across Indonesia's major trading partners (MTP) continued to rebound very trading partners has been strongly in Q2, driven by Singapore, China, and Korea (Figure 2). Consensus forecasts revised up, while for Indonesia's major trading partners have been revised up to 6.3 percent, but most of commodity prices are this growth was already realized by the middle of 2010. International commodity prices expected to continue to grow strongly movements have been mixed since the large falls experienced in May. The World Bank Energy Commodity Price Index was flat, reflecting stable oil prices, while the Non-Energy Commodity Price Index has reached the highest level since September 2008. The increases were led by strong gains in base metals and food prices. Energy and non- energy commodity prices are expected to grow by 20 percent in 2010 and both indices are also projected to be around 4 percent lower in 2011. Figure 1: Quarterly GDP growth remains strong Figure 2: Strong trading partner growth dipped slightly in Q2 (percentage change) (average MTP growth, weighted by Indonesia's export shares) Per cent Per cent Per cent Per cent 4 8 4 8 Year on year YoY Growth (RHS) (RHS) 3 6 3 6 2 4 QoQ seas. adjust 1 2 2 (LHS) Average 4 (LHS)* 0 0 Quarterly Growth (sa, LHS) 1 2 -1 -2 -2 -4 0 0 Jun-03 Mar-05 Dec-06 Sep-08 Jun-10 -3 -6 Jun-02 Jun-04 Jun-06 Jun-08 Jun-10 * Average QoQ growth between Q1 2000 and Q2 2010 Sources: BPS, national statistical agencies via CEIC, Source: BPS, World Bank seasonal adjustment Consensus Forecasts Inc. projections, and World Bank a. Private demand accelerated in the first half of 2010, offsetting a slowdown in government spending Private consumption led Private consumption was the major contributor to expenditure growth in Q2, while growth in Q2 investment growth remained steady. Imports continued to outpace exports. Government spending remained weak in Q2 as slow disbursements continued (Figure 3). Strong consumption in Private consumption supported growth in Q2, with year on year growth returning to around the first half of 2010 was 5 percent. This has predominantly come from growth in expenditures on discretionary due to favorable trends consumption goods, as food expenditure growth has remained steady. The strength in including moderate price private consumption is consistent with rising domestic incomes and moderate price growth growth and rising incomes in the first half of the year. The robust private consumption growth appears to have continued into mid-year. Motorcycle and motor vehicle sales increased strongly into Q3, with sales at record highs (see Appendix). The Bank Indonesia (BI) retail sales index also remains at high levels. Credits to consumers grew very strongly in Q2, further supporting consumption. On the downside, consumer confidence has weakened slightly, perhaps reflecting rising expectations of inflation (see discussion in Part B). 1 Indonesia Economic Quarterly Looking forward Real investment spending Quarter-on-quarter investment growth slowed slightly in Q2, at 1.2 percent, down from has steadied after a 2 percent in Q1. The annual growth rate in investment expenditures is still recovering after recovery which began at a significant slowdown at end-2008 and during 2009, and remains below rates seen the beginning of 2009 during 2007 and the first half of 2008. Much of the rebound in investment over the last year has been from the small, but volatile investment in `foreign machinery and equipment'. The improvement in investment since mid-2009 can be attributed to factors including the favorable price impact of the appreciating exchange rate for imports of capital equipment, improved finance conditions as seen in the small fall in (nominal) lending rates and pickup in new loans, and more recently the recovery in inflows of foreign direct investment (FDI). Indicators of future investment in Indonesia, such as capital imports and cement production, remain robust (see Appendix). Figure 3: Private consumption continues to drive growth Figure 4: `Non-tradable' growth expands ahead of the (contribution to quarter-on-quarter growth, seasonally growth of `tradables' adjusted) (percentage change, year-on-year) Per cent Per cent Per cent Per cent 4 4 10 10 8 Non-tradable 8 2 2 6 6 0 0 4 4 2 2 -2 -2 Tradable Jun-07 Mar-08 Dec-08 Sep-09 Jun-10 Discrepancy Net Exports Investment 0 0 Government Private cons GDP Jun-01 Sep-03 Dec-05 Mar-08 Jun-10 Sources: BPS and World Bank staff calculations `Non-tradables' includes services sectors that may have tradable components such as finance; `tradables' are largely goods-producing sectors. Source: BPS and World Bank staff calculations Growth continues to be `Non-tradable' output continued to grow strongly in Q2, driven by wholesale and retail driven by domestically- trade, transport and communications, and finance and business services (Figure 4). oriented sectors... Growth rates across most other `non-tradable' sectors were stable in Q2. `Tradable' output improved slightly, with the manufacturing sector driving most of that improvement, while agricultural production and construction were steady in Q2. Higher frequency indicators suggest production growth remained robust into mid-2010, with the growth in cement sales at its highest levels since 2008 and industrial electricity consumption growing solidly (see Appendix Charts). ...with the slight increase Indonesia's manufacturing sector grew slightly in Q2, but year-on-year growth rates are in tradable output coming still below those of 2007 and 2008. Much of the recent recovery in manufacturing is driven from the manufacturing by transport equipment and machinery, consistent with the growth in private investment, sector while food and tobacco are a distant second. b. Import growth continues to outpace that of exports Import volumes continue In the second quarter of 2010, the growth in real imports of goods and services continued to recover more quickly to outpace that of exports. The demand for imports has been supported by robust than exports, but rising domestic economic activity and the strengthening Rupiah. The faster pace of real imports commodity prices saw growth saw net external demand subtract 0.1 percentage points from GDP growth in Q2. the trade surplus and current account surplus Rising commodity prices also supported increased trade values. Trade values, as remain stable in Q2 measured by Bank Indonesia (BI), had exports growing at a similar pace to imports, leaving the trade surplus stable in Q2. The stable trade surplus recorded by BI, combined with a widening in the income deficit (as oil & gas companies' profit repatriations increased), resulted in a slight contraction in the current account surplus for Q2. THE WORLD BANK | BANK DUNIA September 2010 2 Indonesia Economic Quarterly Looking forward Looking at the monthly trade data, the rise in import values has been driven by intermediate and capital goods, with machinery, electronics, vehicles and plastics realizing the strongest gains by sector (Figure 5). The recovery in exports has slowed, following the strong rebound during the earlier phase of the global recovery (Figure 6). Exports of vehicles, electronics, rubber, and clothing & footwear continue to rise, on the improved first half demand from the Euro zone. Mineral fuel and mining exports have stagnated, due to a pause in demand from China and India following restocking during the earlier phase of the global recovery. As a result of these trends, the monthly trade balance moved to a small deficit of USD 100 million in July (see Appendix), with the three month moving average surplus declining to USD 1 billion. Figure 5: Imports values are growing strongly... Figure 6: ...while exports values continue to stagnate (billions of USD, 3 month moving average) (billions of USD, 3 month moving average) USD bn USD bn USD bn USD bn 6 15.0 6 15.0 Manufacturing Total Exports (LHS) (RHS) 5 12.5 5 12.5 Intermediate Total Imports (LHS) (RHS) 4 10.0 4 10.0 Agriculture & Forestry (LHS) 3 7.5 3 7.5 Capital (LHS) 2 5.0 2 5.0 Oil & Gas Mining & (LHS) Minerals (LHS) 1 2.5 1 2.5 Consumption (LHS) Oil & Gas (LHS) 0 0.0 0 0.0 Jul-07 Jan-08 Jul-08 Jan-09 Jul-09 Jan-10 Jul-10 Jul-07 Jan-08 Jul-08 Jan-09 Jul-09 Jan-10 Jul-10 Source: BPS and World Bank staff calculations Source: BPS and World Bank staff calculations c. Indonesia's economic growth is likely to rise further Growth projections for Indonesia's economy is expected to grow by 6.0 percent in 2010, increasing to 2010 have been raised 6.2 percent in 2011 (Table 2). The forecast for 2010 has been raised slightly from slightly, reflecting the 5.9 percent due to improved domestic and international conditions into the second strong performance in Q2 quarter, while recognizing the downside risks associated with volatile financial markets and expectations that domestic demand will and uncertain developed economy growth prospects. continue to be supportive Private demand is projected to continue to support growth, offsetting a reduction in net external demand as imports outpace export growth. The growth rate of private consumption is expected to rise gradually, up from 5.2 percent in 2010 to 5.3 percent in 2011. This will continue to be supported by stable employment and improving incomes, offsetting the rise in prices in mid-2010. Sectors focused on domestic demand (transport, communications, various consumer and business services) are likely to continue to perform better than those more dependent on external markets (manufacturing and resource-related industries). However, the latter offers greater scope for upside surprises as the international environment improves. Investment is also expected to continue accelerating in 2010, stimulated by the increase in commodity prices, prospects for higher foreign direct investment, stronger credit growth, and the need to invest to expand capacity in some domestically-focused industries. Government consumption is likely to continue growing less quickly than the overall economy into 2011, although the weakness of core and developmental government spending in early 2010 suggests scope for some pick up later in the year. The current account Reflecting the relative strength of domestic to external demand, the outlook remains for surplus is expected to import growth to outpace export growth through 2010 and 2011. Imports are expected to narrow in 2010 before be led by capital and machinery imports, driven by robust growth in domestic investment. moving into deficit in In part these will contribute in due course to improving the competitiveness and capacity 2011, as the trade surplus narrows of exporters, but much will also expand production capacity for the domestic market, supporting domestic demand and incomes, and to a large extent these two processes are THE WORLD BANK | BANK DUNIA September 2010 3 Indonesia Economic Quarterly Looking forward likely to have offsetting impacts on the overall current account balance. Meanwhile, in the shorter term, increasing demand for raw materials from China and India is expected to see commodity exports reaccelerate through 2010-2011. The trade surplus is expected to contract by a little over USD 7 billion to around USD 14 billion in 2010, and by another USD 3-4 billion in 2011. Meanwhile, the income deficit is expected to widen, as higher commodity prices result in greater profit repatriation, particularly by oil & gas companies. Together these forecasts suggest the current account surplus is likely to register a small surplus in 2010, before moving into deficit in 2011. Table 2: Domestic demand is projected to continue to support a gradual rise in Indonesia's growth (percentage change, unless otherwise indicated) Annual Year to December quarter Revision to Annual 2009 2010 2011 2009 2010 2011 2010 2011 1. Main econom ic indicators Total Consumption expenditure 6.2 5.2 5.4 5.9 7.3 4.3 0.1 -0.1 Private consumption expenditure 4.9 5.2 5.3 4.0 5.6 5.1 0.1 -0.1 Government consumption 15.7 5.4 6.4 17.0 15.5 0.4 0.0 0.0 Gross fixed capital formation 3.3 8.3 9.2 4.2 9.0 9.3 -1.3 0.9 Exports of goods and services -9.7 11.5 11.0 3.7 4.3 13.7 -3.4 -0.1 Imports of goods and services -15.0 15.0 11.6 1.6 7.0 12.9 -2.4 -0.1 Gross Dom estic Product 4.5 6.0 6.2 5.4 6.3 6.1 0.1 0.0 Agriculture 4.1 3.0 3.1 4.6 2.9 2.0 -0.4 0.1 Industry 3.5 4.3 5.3 5.1 4.4 5.1 0.0 -0.1 Services 5.7 8.6 8.0 5.9 8.8 8.0 0.3 0.0 2. External indicators Balance of payments (USD bn) 12.5 19.0 8.8 n/a n/a n/a 12.9 3.7 Current account balance (USD bn) 10.5 1.8 -1.2 n/a n/a n/a -0.7 -1.7 Trade balance (USD bn) 21.0 13.9 10.5 n/a n/a n/a -0.3 -2.7 Financial account balance (USD bn) 3.6 16.4 9.7 n/a n/a n/a 13.6 5.1 3. Other econom ic m easures Consumer price index 4.8 5.1 6.4 2.6 6.4 6.1 0.0 0.1 Poverty basket Index 5.8 7.8 7.4 2.9 8.4 7.5 1.0 0.2 GDP Deflator 8.5 8.8 12.2 6.6 10.5 11.9 -0.6 0.0 Nominal GDP 13.4 15.4 19.2 12 17.4 18.8 -0.5 0.0 4. Econom ic assum ptions Exchange rate (IDR/USD) 10356 9091 9000 9475 9000 9000 -127.3 -200.0 Interest rate (SBI, 1 month) 7.3 6.4 6.5 6.5 6.5 6.5 0.0 0.0 Indonesian crude price (USD/bl) 61.6 77.5 77.0 75.1 77.0 77.0 -0.2 -1.0 Major trading partner grow th -0.8 6.5 4.3 3.3 5.4 5.0 1.5 0.0 Note: Projected trade flows relate to the national accounts, which may overstate the true movement in trade volumes and understate the movement in prices due to differences in price series. Source: MoF, BPS, BI, CEIC and World Bank projections THE WORLD BANK | BANK DUNIA September 2010 4 Indonesia Economic Quarterly Looking forward 2. The balance of payments surplus was near historic highs in Q2, but future narrowing is expected The BoP surplus The balance of payments (BoP) surplus remained near historic highs in Q2, at USD 5.4 remained near historic billion, down only slightly from the USD 6.6 billion surplus in Q1. The cumulative BoP highs in Q2, as strong surplus for the first half of 2010 is the highest on record. Further strong capital inflows in capital inflows more than Q3 have seen reserves rise to over USD 81 billion at the end of August. These record offset a slight contraction in the current account inflows have been driven by continued strong demand from foreign investors for surplus government bonds and short-term paper, as well as strong direct investment inflows. This reflects the strong domestic economic performance, the yield differential relative to developed economy assets and also expectations of continued exchange rate strength. The capital inflows more than offset the slight narrowing in the current account surplus (see Appendix). The balance of payments A continuation of net capital inflows into the second half of 2010 is expected to support an is expected to remain annual BoP surplus in excess of the historic high of USD 14.5 billion in 2006. However, high in 2010, before higher bond amortization and a move into current account deficit should see the BoP narrowing in 2011 surplus contract in 2011 (Table 3). The regular fluctuations in public debt issuances and loans drawings & repayments will continue to create volatility in the quarterly BoP. Table 3: The BoP surplus is projected to reach record highs in The underlying drivers of capital inflows are expected 2010, then to narrow in 2011 to continue into the medium-term, with the potential (billions of USD) for Indonesia's government debt to attain investment- 2007 2008 2009 2010 2011 grade rating also a supporting factor. Net capital Balance of Payments 12.7 -1.9 12.5 19.0 8.8 inflows are expected to continue to underpin the BoP surplus and to support reserve accumulation. But, Current Account 10.5 0.1 10.7 1.8 -1.2 portfolio inflows remain sensitive to shocks to Trade Balance 20.9 9.9 21.0 13.9 10.5 confidence and risk aversion. In conjunction with the Income Balance -15.5 -15.2 -15.1 -17.2 -17.4 strong portfolio inflows, firm net direct investment Transfers Balance 5.1 5.4 4.9 5.1 5.6 inflows are projected over 2010 and 2011, as the global economic recovery and improved financing Capital & Financial Accounts 3.6 -1.8 3.5 16.6 10.0 conditions see firms scale up investment. Government policies to improve the investment climate, and to Capital Account 0.5 0.3 0.1 0.2 0.3 expand private sector involvement in infrastructure Financial Account 3.0 -2.1 3.5 16.4 9.7 investment programs, should also support increased Direct Investment 2.3 3.4 1.9 5.4 4.6 direct investment. The BoP surplus, combined with Portfolio Investment 5.6 1.8 10.3 14.3 10.5 continued high roll-over rates on private short-term Other Investment -4.8 -7.3 -8.8 -3.3 -5.4 debt, support the funding of Indonesia's external Foreign Reserves (a) 56.9 51.6 66.1 81.3 financing requirements. Notes and sources: (a) 2010 foreign reserves value at end of August. BI and World Bank projections. 3. Indonesian financial markets strengthened with the return of strong capital inflows a. Indonesian asset prices strengthened in line with renewed capital inflows The return of capital After the strong outflows of May (USD 5.7 billion), foreign capital inflows to Indonesia inflows has more than have returned strongly in recent months, across equities, SUNs and SBIs (Figure 7). reversed the outflows Approximately USD 7.3 billion of net foreign capital inflows were seen over June-August. seen in May... This brings year-to-date net capital inflows to USD 11 billion. The high volume of recent inflows has contributed to gains across domestic asset classes. ...with Indonesian asset In terms of equities, the Jakarta composite index, after dropping 5.5 percent in May, rose prices rising across the by 13 percent from June to September. Around USD 1.3 billion of net capital inflows board... entered the equity market from June to August, compared with USD 1.5 billion in the whole of 2009, and USD 2 billion in 2008. Indonesian Rupiah and USD denominated sovereign bonds have also strengthened (see Appendix). For example, after reaching 8.75 percent in May, 5-year local currency sovereign yields fell to record lows in late June and have remained below 8 percent. Again, foreign purchases of SUNs have been a key supporting factor. June to August saw USD 3.9 billion in net foreign purchases of SUNs, bringing year-to-date net inflows to USD 7.9 billion. Foreign investors now own a record THE WORLD BANK | BANK DUNIA September 2010 5 Indonesia Economic Quarterly Looking forward 28 percent of SUNs outstanding. In addition to the inflows into SUNs and equities, there were inflows of USD 2.1 billion into SBIs. Indonesian EMBI USD bond spreads, which serve as a proxy for `country risk', have also narrowed. Strong economic growth, the potential for further sovereign ratings upgrades and relatively high yields are supporting foreign investors' demand for Indonesian fixed income assets. ... including upward The high volume of recent inflows has also renewed upward pressure on the Rupiah, pressure on the exchange reversing the weakening in May (Figure 8). The Rupiah has appreciated by 3.5 percent rate since the beginning of 2010 and reached a three-year high against the USD in early August. It is noteworthy that the appreciation would have likely been much sharper in the absence of central bank intervention to stem the rate of appreciation and currency volatility (BI's foreign currency reserves rose by nearly USD 6 billion in June-August, after plummeting by USD 4 billion in May when the bank intervened to support the Rupiah). Figure 7: Large recent foreign capital inflows were seen Figure 8: ...contributing to the appreciation of the Rupiah to across asset classes a 3-year high against the USD in early August (non-residents' net purchases of SBIs, equities and SUNs in (net foreign capital flow in trillions of IDR; IDR per USD) trillions of IDR; foreign reserves in billions of USD) IDR trillion USD billion IDR trillion IDR per USD 60 90 45 8,000 Foreign Reserves (RHS) 40 75 25 9,000 20 60 5 10,000 0 45 -15 11,000 Net Foreign IDR/USD -20 Net Foreign Purchases (LHS) 30 Capital Inflows (RHS) SBI JCI SUN -35 (LHS) 12,000 -40 15 IDR Appreciation -55 13,000 -60 0 Jan 09 Jun 09 Nov 09 Apr10 Aug10 Jan-07 Apr-08 Jun-09 Aug-10 Source: BI, CEIC and World Bank Source: CEIC and World Bank The real effective In addition to the nominal strength of the Rupiah against the USD, the real effective exchange rate has also exchange rate (a trade-weighted basket of key bilateral Rupiah exchange rates, taking continued to appreciate into account inflation) has appreciated markedly. It was 22 percent higher in July 2010 since March 2009 but the than a year earlier, 16 percent up on the pre-crisis levels of July 2008. This has recently IMF detects no major misalignment from fair led to some debate on whether the currency is overvalued at current levels. Based on the value at this stage IMF's latest Article IV report on Indonesia, their models do not suggest any major 1 misalignment at this stage. It will be important to continue monitoring the REER in coming months, and comparing its movement and trajectory to the real effective exchange rates of other countries in the region, as this can have a significant impact on the competitiveness of Indonesia's economy. The impact of The impact of large capital flows on the nominal exchange rate poses challenges for unpredictable capital central banks wishing to curb currency volatility through intervention in the foreign flows on exchange rate exchange market. Historically, the most volatile capital flows in Indonesia have tended to volatility triggered BI to be foreign investments in Bank Indonesia Certificates (SBIs), due to the short-term nature introduce new monetary measures in June, but of these money market instruments. In an attempt to manage the volatility of such flows, their effectiveness in BI announced new monetary measures in June (IEQ June 2010) such as a one month curbing short-term flows minimum holding period for SBI investments as well as a move towards reducing shorter into SBIs is still tenor SBIs and issuing 9- and 12-month paper. It is still early to gauge the effectiveness of uncertain... these measures; while net foreign purchases of SBIs fell significantly in July, August saw a resurgence of net inflows amounting to USD 1.4 billion. (Figure 9) Moreover, foreign holdings of SBIs as a percentage of total SBIs outstanding (a volatile series) went from 1 Available at http://www.imf.org/external/pubs/ft/scr/2010/cr10284.pdf. THE WORLD BANK | BANK DUNIA September 2010 6 Indonesia Economic Quarterly Looking forward 15.5 percent at the end of June to 20.5 percent at the end of August. Nonetheless, BI's introduction of a one month holding period may have helped limit daily volatility. ...although there does One consequence of BI's June monetary measures may have been to increase foreign seem to have been a rise investors' interest in short-term government bonds (in lieu of SBIs). At the end of May, in foreign investors' foreign investors' holdings of longer-term (over 5 year maturity) SUNs accounted for interest in short-term 77 percent of total foreign holdings of SUNs. This share fell to 68 percent by the end of government bonds since June, putting downward August. In contrast, the value of foreign holdings of short-term (0-2 year) SUNs almost pressure on short-end doubled over this period and as a share of total holdings went from 8 percent to 13 yields percent. (Figure 10) Given the Ministry of Finance's intention to reduce short-term bond issuance in favor of longer-term tenors, foreigners may not be able to continue this trend. However, temporarily at least, the increased interest appears to have put downward pressure on short-end yields which dropped more aggressively than 5 year and longer- end bond yields between June and August. Figure 9: Net foreign purchases of SBIs fell sharply in July Figure 10: ...while foreign investment in short-dated SUNs after BI's new measures, but climbed again in August... increased after BI's June measures (non residents' holdings of SBIs and net purchases of the (non-residents' holdings of SUNs by maturity, in trillions of same, in trillions of IDR) IDR) IDR trillion IDR trillion IDR trillion 200 90 45 Foreign Holdings of SUNs 0-2 yr 2-5 yr >5 yr 75 30 160 Net Foreign Purchases of 60 SBI (RHS) 15 120 45 0 80 30 -15 40 15 Foreign holdings -30 of SBI outstanding 0 0 (LHS) -45 Jan-07 Apr-08 Jun-09 Aug-10 Source: BI, CEIC and World Bank Source: CEIC and World Bank b. With money supply growth and inflation picking up, BI has raised the statutory reserve requirement for commercial banks Rising foreign reserves The USD 7.3 billion of net capital inflows between June and August led to a rise in foreign have been accompanied reserves of over USD 5 billion. Base money expanded by USD 3 billion while SBIs by an expansion of base outstanding remained relatively flat (a USD 4 billion increase in 6 month and 3 month SBIs money over June-August, was offset by an equivalent reduction in 1 month SBIs) indicating that BI did not sterilize with SBIs outstanding remaining flat the reserve increase to the same extent as in previous months. (Figure 11). Money growth also In addition to the increase in base money, M1 and M2 growth picked up significantly in picked up steam in Q2, Q2, after a very subdued start to the year. M1 grew by 10.3 percent in Q2, corresponding after an unusually slow to an annualized growth rate of 41 percent, while M2 grew by 5.6 percent, or 22 percent in start to the year annualized terms. If growth in broad and narrow money were to continue at this pace, it would top the 2006-08 annual growth averages for M1 and M2, of 24 percent and 18 percent, respectively. In year-to-date terms, the growth numbers are much more muted given the slow start to the year (Figure 12). With money supply Given the increase in money supply growth and the recent pickup in headline and core growth and inflation inflation figures, in early September BI raised its statutory reserve requirement for rising, BI raised its commercial banks from 5 to 8 percent, effective November 1. The increase in the reserve statutory reserve requirement will drain approximately IDR 53 trillion from banking system liquidity. At this requirement from 5 to 8 percent in early stage, most market participants expect the central bank will raise its policy interest rate September from 6.5 percent by the end of this year or early next year, depending on the evolution of prices and credit in coming months. THE WORLD BANK | BANK DUNIA September 2010 7 Indonesia Economic Quarterly Looking forward Figure 11: Base money increased from June to August Figure 12: ...and M1 and M2 growth rates also picked up (base money and SBI outstanding in trillions of IDR; reserves pace in Q2 after an unusually sluggish start to the year in billions of USD) (M1 and M2 growth in year-on-year percentage change) IDR trillion USD billion Percent YoY Percent YoY 450 90 40 40 M0 Base Money (LHS) 350 75 M1 growth 30 30 M2 growth 250 60 20 20 Total Reserves (RHS) 150 45 10 10 Total SBI Outstanding (LHS) 50 30 0 0 Jan-08 Nov-08 Sep-09 Aug-10 Jan-07 Mar-08 May-09 Jul-10 Source: BI, CEIC and World Bank Source: BI, CEIC and World Bank c. Credit growth increased significantly in Q2, boosted by working capital and investment loans Credit growth rose to Overall, credit growth was up 18.8 percent year-on-year in June and average lending almost 20 percent year- rates were down to 13.62 percent, the lowest level this decade. (Figure 14). Quarterly on-year in June, driven by credit growth was positive for a fifth consecutive quarter and increased significantly in Q2, investment and working as quarterly new loan approvals continued to rise. (Figure 13). Pickups in investment and capital loans working capital loan growth were the main drivers of the credit acceleration, while consumer loan growth dropped off from Q1 levels (although it remains high in year-on- year terms). Investment loan growth was up 10 percent month-on-month, 25 percent year-on-year in June, while the corresponding growth numbers for working capital loans (which constitute 48 percent of total loans) were 5 percent and 13 percent, respectively. To the extent that these loans actually finance new projects, they are supportive of productive investment, and the outlook for growth going forward. Figure 13: Quarterly credit growth and new loan approvals Figure 14: ... driven by pick ups in working capital and rose sharply in Q2... investment loan growth (new loan approvals in IDR trillion; credit growth in quarter- (investment, consumer and working capital loan growth in on-quarter percentage change) year-on-year percentage change; lending rate in percent) Percent QoQ IDR trillion (quarterly) Percent YOY Percent 12 1150 50 18 Investment Loan Lending Rate Credit Growth (LHS) 10 1000 (RHS) Growth 40 15 (LHS) New Loan 8 850 Approvals 30 12 (RHS) 6 700 20 9 4 550 Consumer Loan Growth (LHS) 10 6 2 400 Total Loan Growth (LHS) 0 250 0 3 Working Capital Loan Growth (LHS) -2 100 -10 0 Mar-07 Apr-08 May-09 Jun-10 Jan-07 Feb-08 Apr-09 Jun-10 Sources: BI and World Bank Sources: BI and World Bank THE WORLD BANK | BANK DUNIA September 2010 8 Indonesia Economic Quarterly Looking forward In order to support In an attempt to promote continued loan growth, in early September BI announced a lending further, BI target range for commercial bank loan-to-deposit ratios (LDRs) of 78-100 percent. introduced a target LDR Effective March 2011, penalties involving an increase in statutory reserve requirements range of 78-100 percent, will be imposed upon banks with LDRs outside this range (although for banks above the below or above which, banks will be penalized stated range, penalties will be waived as long as their capital adequacy ratios exceed 14 percent). This policy change is in addition to the previously mentioned rise in commercial banks' statutory reserve requirements from 5 to 8 percent. Although the two policies may look contradictory at first glance, they are actually complementary if BI's goal was for banks to meet the increased reserve requirement by reallocating their loanable funds, rather than by reducing their lending. There are different views BI's policy change is set in the context of the 75 percent LDR for the overall banking on the actual level of sector, and the considerably lower LDR of some of the largest banks. However, a closer commercial bank liquidity look at the numbers suggests that banks may not be as liquid as headline LDR figures based on alternative LDR 2 indicate. For example, including banks' recapitalization bonds on the asset side in calculations ... addition to loans, gives an adjusted LDR of 96 percent for the overall banking sector. (Figure 15) Going one step further in adding banks' investments in SUNs to the "loan" assets raises the LDR even further to 107 percent. Although commercial banks are not obligated to invest in SUNs, if they were to liquidate a large portion of their SUN holdings (currently 36 percent of SUNs outstanding) to free up funds for loans to corporates or households, this would have a wider impact on yields. While the addition of SUNs to loans and recapitalization bonds may result in an overstatement of banks' true LDR, the point is that the system wide figure of 75 percent may be an understatement. The actual liquidity of banks may be somewhere between this figure and the different adjusted values. ...and rising undisbursed Another noteworthy factor is the significant increase in undisbursed loans in the year-to- loans date (Figure 16). At the end of June, undisbursed loans on commercial bank balance sheets were 50 percent higher than in December 2009, and the ratio of undisbursed loans to total loans climbed from 22 percent to 30 percent. This has a major impact on banks' available liquidity for lending as they must set aside funds to provide for undisbursed loans. Figure 15: Banks' LDRs rise considerably if recap bonds and Figure 16: ...and the recent sharp rise in undisbursed loans SUNs are included on the asset side... also impacts their liquidity (loan to deposit ratio and adjusted loan to deposit ratios in (liquid assets to deposits ratio and undisbursed loan to loan percent) ratio, in percent) Percent Percent Percent Percent 140 140 50 50 Loan+Recap Bond+SUNs to Liquid Assets to Deposit Ratio deposits ratio (LHS) 120 120 40 40 100 100 Undisbursed loan Loan+Recap Bond to to loan ratio (RHS) 80 Deposit Ratio 80 30 30 60 Loan to Deposit Ratio 60 20 20 40 40 20 20 10 10 0 0 Dec-07 Jun-08 Dec-08 Jun-09 Dec-09 Jun-10 0 0 Dec 07 Jun 08 Dec 08 Jun 09 Dec 09 Jun 10 Sources: BI and World Bank Sources: BI and World Bank 2 Recapitalization bonds were introduced by the Indonesian government to support commercial banks' balance sheets at the time of the 1997-98 crisis, by replacing written off loans. They still exist on banks' balance sheets. THE WORLD BANK | BANK DUNIA September 2010 9 Indonesia Economic Quarterly Looking forward 4. Core inflation continues to recover while headline inflation accelerated sharply due to temporary food supply shocks a. CPI inflation rose in mid-2010 because of weather-related supply disruptions Consumer prices have The headline annual inflation rate almost doubled from 3.4 percent in March to 6.4 percent picked up sharply in August largely because of volatile food prices (Figure 17). Core inflation (a measure of following their slow underlying inflation that excludes volatile items and administered items) steadily increased recovery in the first over the same period from 3.6 percent to 4.2 percent. Inflation outcomes outstripped months of 2010 market expectations in June and July but were lower than expected in August. Indonesia's headline inflation rate is now well above its major trading partners' (see Appendix). The growth in the GDP deflator, a broader measure of prices in the economy, fell from 7.5 percent in Q1 to 6.9 percent year-on-year in Q2. Food prices were a key Headline CPI inflation accelerated sharply in August, reaching its highest level in 16 source of the acceleration months driven by the negative food supply shocks of mid-2010. An unusually wet `dry' in consumer prices over season impacted various spice and, later, rice prices. From March to July spice prices this period although most grew by 54 percent, contributing more than one-fifth of the total increase in CPI inflation of the increases are expected to be temporary over this period despite a weight of only 1.5 percent in the CPI basket. There was significant variability in spice price growth with the average raised by considerably higher growth throughout Sumatra. These increases are expected to be temporary with chili prices already retreating and by mid-September had almost fully unwound. Higher food prices drove The unusual growing season and weather-related supply disruptions also affected the poverty basket Indonesian domestic wholesale rice prices which remain well-above international price inflation rate above levels (see Box 1 on recent rice price developments). Increases in such basic food prices 10 percent and led to the particularly affect poor households. Food items make up 63 percent of poor households' largest gap between the poverty basket inflation consumption basket on average, and the relatively strong growth in these prices lifted the rate and the headline rate poverty basket inflation rate above 10 percent for the first time since 2008. At 11.3 percent since 2007 the poverty basket inflation rate is around 5 percentage points higher than the headline rate ­ the largest gap between the two series since 2007 (see Appendix). Several administered In July and August there were also price adjustments to several administered items. For price increases in recent example, rising costs of vehicle registration contributed 0.21 percentage points to headline months, including inflation and electricity tariffs, which were increased by an average of 8 percent in July electricity tariffs and 2010, directly contributed 0.35 percentage points (see Box 2 on electricity tariff vehicle registration, also contributed to the sharp adjustments). The impact of the electricity rate increase on industry and its subsequent acceleration in prices pass-through to consumer prices will become clearer in Q3. Figure 17: Inflation has risen over the year but remains Figure 18: Volatile and administered items drove the recent below historic averages spike in headline inflation with core inflation rising steadily (year-on-year percentage change) (contributions to monthly inflation) Per cent Per cent Per cent Per cent 32 32 1.8 1.8 Volatile Wholesale Administered Prices Index Core 24 24 Headline Inflation 1.2 1.2 GDP deflator 16 16 0.6 0.6 8 8 Consumer Price Index 0.0 0.0 0 0 CPI (average 2001-2009) WPI (average 2001-2009) GDP Deflator (average 2001-2009) -0.6 -0.6 -8 -8 Feb-09 Aug-09 Feb-10 Aug-10 Sep - 06 Sep - 07 Sep - 08 Sep - 09 Sep - 10 Sources: BPS and World Bank Sources: BPS and World Bank THE WORLD BANK | BANK DUNIA September 2010 10 Indonesia Economic Quarterly Looking forward Core inflation, generally In contrast to the sharp rise in the headline rate, core inflation continued to increase unaffected by the recent steadily (Figure 18). This gradual increase suggests that underlying price pressures within volatility, picked up, with the economy, while rising, remain stable despite the abrupt acceleration in the headline monthly outcomes rate. The monthly core inflation outcomes have gradually returned to historical rates in returning to historic averages July and August reflecting four consecutive quarters of above average GDP growth and rising credit growth. The effects of the food supply shock do not appear to have led to broader price adjustments as yet, reflecting recognition of the temporary nature of the supply shock. Initially inflation expectations increased sharply in July but by August this rise had been fully unwound. The strong Rupiah, which appreciated for the sixth consecutive quarter against the USD, has also provided some dampening of inflationary pressures from rising import prices. Economy-wide price GDP deflator growth was weak in Q2 relative to the past decade, growing by 6.9 percent growth was also below year-on-year, reflecting broad-based trends in implicit price deflators. Investment prices trend growth continued to be around 8 year-lows, as construction costs fell for a second quarter while transportation prices growth continued to recover but remain well below pre-crisis levels. The sluggish recovery in the growth of investment prices kept the gap between growth in the CPI and the GDP deflator relatively low in late 2009 and the first half of 2010. b. The inflation outlook for 2010 remains unchanged, but stronger price growth is expected in 2011 The inflation outlook for The outlook for consumer prices growth in 2010 remains unchanged relative to the June 2010 remains unchanged IEQ at 5.1 percent. This reflects an expectation that most of the food price shocks in the as recent food price last three months will be temporary and unwind. However, the large price increases of low increases are expected to quality rice in the past three months appear less likely to fully retract by the end of 2010 unwind by the end of the year and will significantly affect poorer households. Accordingly, the outlook for the poverty basket inflation rate was revised up to 7.8 percent in 2010. The forecast for the GDP deflator for 2010 was revised lower to 8.8 percent as the Q2 outcome was weaker than expected. The faster recovery in import prices compared with export prices pulled down the terms of trade to near record lows in Q2 which led to a downward revision in the GDP deflator growth forecast. Inflation is expected to Looking ahead to 2011, most of the short-term supply disruptions are likely to have abated. rise in 2011 However, inflation rates are expected to remain above BI's 4-6 percent target band as the recovery in new lending and acceleration in the money base experienced since late 2009 support domestic demand, and the adjustment in energy prices towards their economic price continues, and the exchange rate stabilizes. There is expected to be some rise in imported inflation, as the outlook for Indonesia's major trading partners improves in 2011. This, combined with a more favorable investment outlook and stronger credit growth, led to a small upward revision in the headline inflation rate to reach 6.4 percent in 2011. Box 1: Recent rice price developments Domestic rice prices rose significantly and reached a record high in August 2010, due to unusually wet weather which caused crop losses and distribution disruptions. The average wholesale rice price for the lowest quality (IR64-III) was IDR 5,848 per kg in August, compared with IDR 4,882 per 3 kg in August last year. Rice price growth on a year-on-year basis has increased by almost 20 percent in August ­ far faster than the general rate of inflation (which reached 6.4 percent in August). Prices remained elevated until early September. The rise in prices was caused by an unexpected drop in rice supply in several regions, as unusually prolonged wet weather caused various kinds of pestilences to break out in several main production areas. Moreover, continuous rain also disrupted the inter-city and inter-island rice distribution channels. Although historically more stable, the domestic rice price is currently much higher than the world rice price. As seen in Figure 19, the domestic rice price is relatively stable but has shown an increasing trend. The Government of Indonesia has succeeded in stabilizing the domestic rice price through a combination of rice procurement and imports. 3 IR64-class III is the lowest quality rice price in the Jakarta Wholesale Rice Market (PIBC). THE WORLD BANK | BANK DUNIA September 2010 11 Indonesia Economic Quarterly Looking forward After a sharp rise and fall in 2008 during the global food price crisis, international rice prices have been on a declining trend, in contrast to domestic price movements, and are currently at a much 4 lower level. The fall in prices in international markets between December 2009 and July 2010 was due to an abundant supply of traded rice. For example, the Vietnam 25 percent broken rice price 5 fell from USD 488 per metric ton in December 2009 to USD 325 per metric ton in July 2010. There were some upward movement in August this year, up 2.2 percent to USD 332 per metric ton, on the back of strong demand and concerns that erratic weather across major food grain growing regions in the world would damage rice crops. As a result of these divergent price trends, the Figure 19: Domestic and International Rice Price Comparison price that Indonesians pay (Price difference between domestic and international prices, percent; IDR/Kg) for their rice has been rising 100 Price difference between 9,050 considerably above the domestic and international international price. This 80 price (%), LHS 7,900 positive price differential has not always been the case, 60 Domestic with domestic prices below 6,750 IR64-III the elevated international 40 rice, RHS prices of 2008. However, 5,600 over the course of 2010 the 20 gap has widened markedly. 4,450 For example, the domestic 0 price was only 4 percent 3,300 more expensive than the -20 international price in December 2009, but this -40 2,150 Vietnam 25% gap increased to broken, RHS 77.2 percent in August 2010. -60 1,000 This stark difference between domestic and international prices occurs while rice imports are prohibited. This not only Source: PIBC, FAO and staff calculation encourages illegally imported rice to enter the domestic market, but also does not facilitate price convergence, leaving most rice consumers in Indonesia to buy their staple food at prices higher than otherwise. Rice is the most important commodity in Indonesia. It is the main staple of the majority of the population and also plays an important part in Indonesia's rural economy. Although the trend is declining, in 2009 around 25 percent of all rural households in Indonesia worked in rice planting, and many more are connected to the rice economy through services, trade and labor. While food constitutes more than 50 percent of expenditure for half of the population in Indonesia, rice constitutes around 17 percent of all expenditure for the poorest 20 percent of the population, in contrast with only 3.8 percent for the highest quintile (Table 4). As a result increases in the domestic rice price significantly impact the purchasing power of the poor and near poor, with more adverse effects on those who are net rice consumers. Table 4: The importance of food and rice in per capita expenditures Expenditure Total Food Rice Share of Share of Share of rice to quintile expenditure expenditure expenditure food rice food expenditure expenditure expenditure (IDR/month) (IDR/month) (IDR/month) (%) (%) (%) Lowest 178,920 100,511 30,521 56.18 17.06 30.37 267,225 155,504 39,308 58.19 14.71 25.28 358,602 202,234 42,976 56.40 11.98 21.25 504,500 262,269 43,881 51.99 8.70 16.73 Highest 1,084,432 388,894 40,999 35.86 3.78 10.54 Source: World Bank staff calculations based on Susenas 2009 4 Insurance, freight, transportation and other costs have been added to the international price to convert it into a wholesale domestic price equivalent. 5 The price of Vietnam 25 percent broken rice traded in the international market is chosen due to its similar quality to the IR64 class III rice traded in domestic market. THE WORLD BANK | BANK DUNIA September 2010 12 Indonesia Economic Quarterly Looking forward Unlike beef or other food prices, rice price fluctuations are not affected by festive events such as Ramadhan, Idul-Fitri or the New Year. Rice price movements are more affected by the availability of domestic rice supply which is related to rice crop cycle. In general, Indonesia has three crops per year: the biggest in March/April, followed by July/August and December. The rice price commonly peaks before harvest time (in February/January, and September) when there is a shortage in rice. In contrast with the previous rice harvest season, the rice harvest in July/August 2010 did not result in a decrease in rice price. According to rice traders, supply in main production areas remained weak in August, driven by crop failures in several regions due to pests and wet weather. Therefore, the current rice price increases in many regions in Indonesia are mainly due to shortage of rice supply and the disruptions to distribution caused by heavy rains related to the La Nina phenomenon. While prolonged rains could benefit areas which rely on rain in rice planting there is evidence that pest attacks have been more severe in 2010 than in previous years. Improvements of the quality of domestic infrastructure, as well as the provision of reliable data related to rice are important aspects for domestic rice price stabilization. A trade policy which allows the private sector to trade in rice while including a specific tariff could also maximize benefits for the country and protect farmers. Distribution of rice from main production areas to other areas should be supported by appropriate infrastructure to avoid high distribution costs. The role of Bulog in distributing rice to remote areas remains significant, but the availability of reliable infrastructure is very critical to transport rice and other food commodities in an efficient and effective way. Moreover, availability of accurate information related to rice such as production, consumption and value of stocks would allow the Government to take appropriate, and timely, action to manage rice price stability. 5. Poverty continues to trend downwards The 2010 Susenas survey The robust recent growth performance contributed to the improvements seen in the March points to a continuation 2010 Susenas poverty numbers. The absolute number of urban poor and rural poor fell of the trend of poverty below 11.5 million and 20 million (respectively) for the first time since 2004. This meant reduction in Indonesia... year-on-year declines in urban and rural poverty rates of approximately 0.85 and 0.80 percentage points (respectively). The overall poverty rate declined to 13.3 percent, from 14.2 a year earlier. Going forward, poverty reduction performance in urban areas will account for an increasing share of overall poverty performance: preliminary figures from the 2010 census indicate that more than 50 percent of Indonesians now live in cities and that the long migration of the labor force from rural to urban areas continues (see Part C for a discussion of the relationship between development and urbanization and the associated policy challenges). ... but with regional The greatest year-on-year percentage point declines in poverty headcount rates were variation in the found in Sulawesi, particularly Sulawesi Barat, Sulawesi Tenggara, and Gorontalo. These reductions achieved provinces were top performers on all three measures: urban, rural, and combined. Among these top performers, Gorontalo managed relatively large poverty rate reductions in both urban and rural areas even though it began 2009 with one of the largest divides between urban and rural: 2009 headcount poverty rates approached 35 percent in rural areas but were less than 8 percent in urban areas. The worst performers in Kalimantan and Sumatera had essentially no year-on-year reduction in poverty, but those provinces began 2009 with already-low overall headcount rates of under 10 percent. Looking forward, the The poverty rate is expected to continue to decline over the next year. Although overall benefits of growth for economic growth is expected to be strong, at 6.2 percent, growth in the agriculture sector, poverty reduction may be where many of the poor are employed, is projected at 3.6 percent. More importantly, the dampened by the impact annual increase in the cost of living for the poor is expected to be 7.8 percent, nearly of rising living costs 3 percentage points higher than non-food non-fuel inflation. Such increases in the prices of goods and services consumed by the poor dampen the poverty-reducing effects of growth with preliminary estimates projecting a fall in the poverty rate to 12.75 percent by March 2011, a 0.6 percentage point decrease from 2010. Labor force growth has In terms of recent labor market developments, in May 2010, BPS reported that the labor been accompanied by a force had grown by 2.2m people in the six months to February 2010, reaching 116m, of fall in the open whom 107m were working, resulting in an open unemployment rate of 7.4 percent, down unemployment rate from 7.9 percent in August 2009 and 8.1 percent in February 2009. Employment in the agricultural sector actually fell by 0.5 percent and by 2.2 percent in transportation. THE WORLD BANK | BANK DUNIA September 2010 13 Indonesia Economic Quarterly Looking forward 6. Reallocating and improving the quality of fiscal spending is on the Government's agenda a. The Government proposes a reallocation of expenditures in the 2011 budget The Government The Government has reduced its projected 2010 deficit, to 1.5 percent of GDP, reflecting announced the 2011 both modified assumptions, strong revenue realizations, and the weakness in budget proposal in mid disbursement early in the year. The 2011 budget proposal was announced in mid-August August ... and includes an increase in expenditures of approximately 6.5 percent relative to the 2010 revised budget. The projected revenues are up by 9.5 percent (by slightly less relative to the latest 2010 forecast). The deficit in the Government's proposed 2011 budget is 1.7 percent of GDP. Given the strength of the nominal economy, this is likely to see public debt levels continue to fall towards 25 percent of GDP. ...which plans to increase The proposed budget is designed to achieve 10 strategic goals; higher economic growth expenditure to address and international competitiveness, increased economic stability and improvements to the Government's domestic finance, higher employment, lower poverty levels, increased per capita income, strategic goals while improved food, water and energy security, and strengthened efforts to protect the targeting improved quality of spending environment. While the headline deficit figure is only slightly higher that that projected by the Government for 2010, the proposed 2011 budget takes a new direction, allocating significantly more resources to capital investment, while improving the efficiency of subsidy outlays, most notably by reducing the large electricity subsidies. Such improvements in the quality of spending are a key priority and a number of other steps to make improvements in this regard have been outlined. These include, amongst other things, speeding up the implementation of performance based budgeting and the medium- term expenditure framework (MTEF), expanding the implementation of bureaucracy reform and improving budget disbursement through revision of Presidential Decree no. 80/2003 regarding procurement procedures. The Government In an effort to improve the country's severe infrastructure weaknesses, the Government proposed to boost capital has allocated a significant increase to capital expenditure in 2011, of 28 percent (Table 5). expenditure to finance Energy, irrigation and transport sectors are to receive the highest budget increase next infrastructure year. This is in line with targets set by the Government to build 2,800 km of toll roads, development and continue pro-poor social connect more households to the electricity grid, and to raise rice production. Given the program experience of recent years, realizing such increases in budget allocation for infrastructure investment, if approved, would likely require improvements in absorption capacity and disbursement rates. The Government will also continue with the bureaucratic reform initiative in part funded by an 11 percent increase for personnel expenditure in 2011. There is a commitment to continue pro-poor programs such as BOS, Jamkesmas, PKH, and PNPM with a poverty reduction target also outlined. However, some sectors are experiencing budget reductions. For example, the reduction in central spending for education is due to a movement of the BOS program (IDR 16.8 trillion) from central spending to transfers to sub national government. The Government has also proposed further reforms to improve the effectiveness of its energy subsidies in 2011. It has proposed to increase the average electricity tariff by 15 percent starting January 2011, although, at the time of press, this was still being negotiated with members of Parliament, who were prioritizing cuts in the monopoly electricity suppliers' (PLN) operating budget ahead of increasing tariffs towards the economic cost of supply. These plans follow the Government's July 2010 increase in average electricity tariffs by 8 percent. This adjustment is the most substantial reform to electricity tariffs since January 2003, and is a small step towards the Government's 2010- 2014 RPJMN goal of improving the targeting of energy subsidies while minimizing their burden on the budget. In addition, the Government has indicated that it plans to improve the targeting of fuel subsidies. Such reforms can free up further fiscal space for investment and social protection. THE WORLD BANK | BANK DUNIA September 2010 14 Indonesia Economic Quarterly Looking forward Table 5: A strong rise in the allocation to capital expenditure Figure 20: Disbursements rates remain low in 2010 is proposed for the 2011 budget (percent of total full-year budgeted expenditure disbursed in the (level and percentage change by expenditure categories) first half of the year) 2010 2011 60 Per cent Per cent 60 (Revised (Proposed Growth Low 2010 disbursements 2008 2009 2010 Budget) Budget) (Per cent) 45 45 IDR trillion IDR tillion Total Expenditure 1,126 1,202 6.7 30 30 Central Govt 782 824 5.4 Personnel 163 181 11.0 Material 113 132 16.8 15 15 Capital 95 122 28.0 Int. payments 106 116 10.2 0 0 Subsidy 201 185 -8.2 Capital Transf. Total Social Total CG Personal Materials Int. payments Others Subsidies Social assistance 71 62 -13.6 Others 33 26 -20.1 Transfers 345 378 9.8 Sources: MoF Sources: MoF, 2010 data are estimates b. Slow disbursement rates have continued through 2010 Disbursement rates In the first half of 2010 the ratio of realized to budgeted expenditures was 35 percent, continue to be low, slightly below 2008 and 2009 spending rates of 37 percent. The low level of central particularly for capital spending was mainly driven by capital, personnel, `others', and non-energy subsidy spending expenditures which were particularly weak (Figure 20). On capital spending, the top 10 line ministries, which account for 91 percent of total budgeted capital expenditures in 2010, spent only 16.5 percent of their budgeted expenditure compared with 30 percent in 2009. By end-August, cumulative disbursements were only 45 percent of budgeted allocations. The Government has Some long-standing issues constraining timely budget disbursement still persist. These identified a range of include, for example, delays in the appointment of heads of implementing units (Satkers), bottlenecks to improper project planning, incomplete project information, and a lack of procurement disbursement plans by implementing unit. Early budget revisions in Q1-2010 also contributed to the slow disbursement whereby most line ministries held the budget execution until the budget revision was approved. Other factors contributing to slow disbursements in the first half of 2010, as identified by 6 the Government include the following. Late procurement and a lack of understanding by implementation units regarding the disbursement procedures, which hampered materials and capital spending. For personnel expenditures factors included the low realization of a new remuneration scheme in piloted line ministries, and the late disbursement of the 13th month salary, restrained expenditures. Some technical and incomplete payments procedures led to payment delays for non-energy subsidies. Low levels of disbursements were also recorded for "other" expenditures due to low levels of contingency expenditures which are included in this category. c. World Bank projections are for the budget deficit for 2010 to be slightly lower than previously anticipated The World Bank projects The Government's revised 2010 budget projects a deficit of 1.5 percent of GDP. Using an a lower deficit for 2010 alternative methodology based on an assumption of stronger growth in the nominal than the Government due economy, and that disbursements improve only modestly over the rates achieved in the to a stronger assumed first half of 2010, the World Bank projection is for a deficit of 0.9 percent of GDP (see growth in the nominal economy and lower Appendix). forecast expenditures 6 Perkembangan Realisasi Belanja Pemerintah Pusat semester I dan Prognosis Semester II Tahun 2010 THE WORLD BANK | BANK DUNIA September 2010 15 Indonesia Economic Quarterly Looking forward This difference reflects both a lower expenditure projection and higher revenue forecast by the World Bank. On the expenditure side, the forecast is revised down relative to the Government's reflecting the consistent trend of under 100 percent disbursement of planned expenditures. On the revenue side, the World Bank continues to project higher revenues relative to the Government which are largely responsible for the difference in the deficit estimates. The key factor driving the divergence in revenue estimates is the difference in the nominal price index used to inflate real growth to project the level of nominal GDP. Recent trends show the economy-wide prices growing faster than the CPI, and this implies faster growth in nominal GDP, and hence in tax revenues (see Part B, June 2010 Indonesia Economic Quarterly for further detail). This higher projected level of nominal GDP also reduces the budget deficit to GDP ratio. Expenditures are Spending rates in the second half of 2010 are expected to improve, in particular since the expected to rise in the budget revision has been completed in the second quarter. However, the first half second half of 2010, but spending performance and recent developments in macroeconomic assumptions suggest overall annual spending that the 2010 budget outcome will likely be below initial projections. The implementation of is likely to be below budget the new remuneration scheme may not fully materialize within this fiscal year, and the remaining challenges with budget execution may contribute to low spending of core government programs. A stronger Indonesian Rupiah and relatively stable SBI rates suggest that interest payments will also be lower than budgeted. The lower central government expenditures, offset slightly by higher transfers to sub-national governments, mean that the World Bank projection for total spending in 2010 is in line with that made in the June IEQ; a minor negative revision of just under IDR 1 trillion has been made. Government revenues in Aside from some compositional changes, the World Bank projections for government 2010 are expected to be revenues in aggregate have changed little on those of the June 2010 IEQ. Revenue slightly stronger than the growth of 18.2 percent is forecast for 2010. Revenues are projected to fall as a share of June IEQ projections nominal GDP. This fall as a proportion of GDP is explained through various changes to revenue policy. Tax revenue growth Tax revenue is expected to grow by 16.6 percent, a little less than projected in June 2010, projections for 2010 have which is in line with the slight slowing in growth for nominal GDP. Over the last quarter tax been lowered slightly in collections in most categories have slowed. Non-oil income tax and weak VAT revenues line with revisions to are responsible for most of the revision, offset by some strength in excise duties. The nominal GDP... reduction of the corporate income tax rate in 2010 from 28 to 25 percent is likely to explain some of the weakness in income tax, along with the continued recovery of taxable incomes from the economic slowdown in 2009. ...offset by mildly The outlook for non-tax revenues has improved slightly from the June 2010 projections, stronger projections for with annual growth expected of 22.8 percent. Most of this revision is due to mild strength non-tax revenues in the oil and gas natural resources sector and to continued strength in the `other non- taxes' category in recent months, such as visa fees, drivers' license fees, unclaimed tax revenue etc., which have consistently been performing ahead of expectations all year. In terms of financing of With regard to the Government's financing needs for 2010, the Government entered the the 2010 deficit, the year with a financing surplus rolled over from 2009's smaller-than-expected deficit and Government has already anticipatory financing strategy (Table 6). In addition, the downward revision in the met 87 percent of its projected deficit by the Ministry of Finance has led to a reduction in the Government's revised gross bond issuance target for 2010 gross bond issuance target for the year from IDR 175 trillion to IDR 158.3 trillion. By early September 2010, the Government had issued IDR 140 trillion of government bonds (IDR and non-IDR denominated), equivalent to 88 percent of the revised gross bond issuance target. Given the continued expected strong international demand for Indonesian sovereign debt, both local and foreign currency, the Government is unlikely to face difficulty in issuing the remainder. The Government of Indonesia plans to make its second samurai bond issuance in Q4. THE WORLD BANK | BANK DUNIA September 2010 16 Indonesia Economic Quarterly Looking forward Table 6: The Government of Indonesia's financing position for 2010 is favorable (trillions of IDR, unless otherwise indicated) 2010 (p) 2011 (p) 2008 2009 APBN-P WB RAPBN Net financing needs: A Primary deficit -84.3 -5.2 0.0 -35.9 -0.7 B Total interest payments 88.4 93.8 105.7 95.1 116.4 of which:[1] Total commercial bonds: 66.8 63.8 71.9 77.4 80.4 Official external loans -63.4 30.1 33.8 33.4 36.0 A+B Overall deficit: 4.1 88.6 105.7 59.2 115.7 Amortizations: C Commercial bonds [2] 46.8 49.1 64.1 64.1 60.5 D Official external loans 57.9 68.0 54.1 53.5 60.1 E Subsidary Loan Agreement 3.0 16.8 16.8 12.0 C+D+E Total am ortization: 104.7 120.1 135.0 186.4 132.5 Gross financing needs: A+B+C+D+E Total gross financing needs: 108.8 208.8 240.7 245.6 248.2 (in billions of USD) 11.2 20.1 26.2 27.0 26.7 Financing sources: [3] Official borrow ing 50.2 69.3 70.8 69.9 57.1 Total commercial bonds: 85.9 142.4 158.3 156.4 186.0 Domestic banking 16.2 56.6 45.5 19.2 7.7 Other 2.9 3.1 Total gross financing sources: 155.2 271.3 274.6 245.6 250.7 (in billions of USD) 15.9 26.1 29.8 27.0 27.0 M emo items: Variable interest rate (SBI-90 day rate) 9.47% 6.59% 6.50% 6.44% 6.50% IDR/USD exchange rate 9,757 10,408 9,200 9,091 9,300 Share USD bonds (prevailing exchange rate) 11.8% 14.8% Note: [1] Interest payments by component may not sum to totals due to different data sources, timing and rounding issues. [2] No USD bonds are due over this period. [3] World Bank deficit projection is derived from revenue forecasts which are based on a different methodology to the Government to project the nominal GDP (see Part C of the June 2010 IEQ for a full discussion). Sources: CEIC, Ministry of Finance, World Bank projections. Box 2: The July 2010 electricity tariff adjustment: a modest first step to better targeted energy subsidies at a lower burden on the budget In July 2010 the Indonesian Government increased average electricity tariffs by 8 percent. This modest adjustment is the most substantial reform to electricity tariffs since January 2003, and is a small step towards the Government's 2010-2014 RPJMN goal of improving the targeting of energy subsidies while reallocating their burden on the budget to investment in Indonesia's physical infrastructure and human development. The July 2010 increase of 8 percent on average was modest. Between January 2003 and July 2010, prices across the board increased by 71 percent for the average urban household, and by 84 percent for the average poor household. Prices economy-wide (including investment goods and trade prices) rose by 121 percent. Even in the year leading up to the July increases, a period of relatively low inflation by Indonesia's historical standards, the 8 percent increase does not appear excessive, given CPI inflation of 6.2 percent and growth in the GDP deflator of 6.9 percent. Meanwhile, since 2003, the economic cost of PLN's fuels to generate electricity and the resources required to distribute it have also risen substantially. Crude oil prices rose almost two-and-one-half fold and gas prices tripled in IDR terms, while Indonesian construction materials prices (a very rough indicator of the cost of maintaining and building distribution lines) increased by 175 percent. Indonesia's electricity tariffs are generally lower than in neighboring and other major economies, for both residential and industrial users. In many cases the difference is very large. Tariffs have increased substantially in most other economies over the past decade; Indonesia's have been fairly stable. (Figure 21 and Figure 22) THE WORLD BANK | BANK DUNIA September 2010 17 Indonesia Economic Quarterly Looking forward Figure 21: Indonesia's tariffs are low for both Figure 22: ...and industrial users, even after the household... 2010 increase (US cents / kWh) (US cents / kWh) US cents / kWh US cents / kWh 21 21 2003 2008 2010 2003 2008 2010 18 18 15 15 12 12 9 9 6 6 3 3 0 0 * 2007. ^ 2000. Sources: International Energy Agency, OLADE for Latin America Yet this modest and long-delayed increase was controversial. The controversy may largely be due to different customers experiencing very varied tariff movements. Indonesia's electricity customers pay an electricity connection charge and per-unit tariff based on what sort of customer they are (household, business, etc) and their connection capacity (i.e., the maximum amount of electricity they can consume at once), and may receive a discounted or higher usage tariff depending on their actual use. About 87 percent of households Table 7: The tariff increase varies significantly across experienced no change in tariff ­ most of different customer groups, with most customers these have the lowest capacity connections, while 1.3 percent or almost a half million, experiencing no increase with the highest capacity connections, are Average tariff already paying above the economic cost of Rp / kWh Change Share of Share of supply, and further increases may make it 2009 2010 (%) use customers cheaper for these households to switch to TOTAL 676 740 9.4% 100% 100% Households 597 637 6.6% 41% 92% their own supply. PLN bills make up about s/d 450 VA 418 418 0.0% 33% 52% 3 percent of most households' total 900 VA 609 609 0.0% 32% 34% spending (a little less for the very highest 1300 VA 672 793 18.0% 15% 10% spending households) and have a base 2.200 VA 675 797 18.1% 10% 3% 2.200 - 6.600 VA 755 891 18.0% 7% 1% weight of 3.05 percent in the current CPI > 6.600 VA 1,330 1,330 0.0% 3% 0% series, so even a 10 percent increase Business 926 979 5.7% 17% 4% directly raises a household's cost of living by <450 VA 538 538 0.0% 1% 19% about 0.3 percentage points. Yet, <900 VA 634 634 0.0% 2% 19% <1,300 VA 685 795 16.1% 3% 18% surprisingly, the BPS reported that the far <2,200 VA 782 907 16.0% 5% 17% smaller July 2010 increased household fuels 2.200 - 200 KVA 1,104 1,104 0.0% 44% 26% by 5.8 percent, with the increase in PLN >200 KVA 811 908 12.0% 45% 0% Industry 628 718 14.4% 34% 0% tariffs contributing 0.35 percentage points to <450 VA 486 486 0.0% 0% 1% the 0.8 percent increase in the CPI from July <900 VA 604 604 0.0% 0% 1% to August. <1,300 VA 724 767 5.9% 0% 2% <2,200 VA 746 790 5.9% 0% 23% 2.200 - 14 KVA 840 916 9.0% 0% 0% The tariff adjustment faced by businesses 14 - 200 KVA 805 878 9.1% 8% 55% varied widely, depending on their use, and > 200 KVA 641 737 15.0% 68% 18% often on when they established their supply > 30,000 KVA 529 608 14.9% 24% 0% arrangement with PLN. Many newer Social 591 642 8.7% 2% 2% Government 802 901 12.3% 4% 1% business and industrial customers Other 915 1,089 19.0% 1% 0% negotiated special supply agreements with Note: based on the draft of tariff adjustment. The PLN, whereby they agreed to pay a increase actually implemented in July was somewhat premium tariff in exchange for premium smaller, especially among the groups subject to the quality of supply, and larger users have also larger increases. Source: PLN been subject to higher tariffs. Older customers, particularly those with supply contracts dating from 2003, continued to enjoy the lower 2003 regulated tariffs plus any excess usage charges. Such agreements may mean that two otherwise identical business customers pay markedly different electricity tariffs. Businesses were complaining against the discriminatory nature of this system. So a goal of the July 2010 increases was to consolidate the tariffs for commercial users into a common tariff for like commercial customers. It simplifies commercial tariffs, by unifying the separate connection and usage charges, and the higher tariffs for higher users, into one usage tariff for each connection capacity. For some customers, this means significant increases in their overall electricity bill; other users' bills fell. Even for most businesses experiencing the larger increases, the impact on overall costs appears likely to be moderate. PLN bills make up about 5 percent of most medium and large manufacturing firms' total costs, although the share is notably higher for most non-metallic mineral product THE WORLD BANK | BANK DUNIA September 2010 18 Indonesia Economic Quarterly Looking forward producers (e.g., cement) (11.9 percent), textiles (7.4 percent) and electronics manufacturers (6.2 percent), according to firms' responses in the Industry Survey. For the average manufacturer, this means a 15 percent higher electricity bill increases total costs by less than 1 percent. In the extreme case of a firm being charged only the 2003 base tariff, a 90 percent increase in overall tariffs would increase total costs for paper & paper products and apparel industries by 4 to 5 percent. More significant for many of these firms is the cost of unreliable electricity supply ­other fuels make up 5.4 percent of the average manufacturer's costs and much of this may be for fuel for in-house generation used to fill in when the grid supply fails. Even after the July 2010 increase, electricity tariffs remain a significant burden on the budget, consuming 5 percent of total spending or about IDR trillion. To further address this, the Government has announced that it will introduce another round of modest tariff increases at the start of 2011, with its proposal of a 15 percent average increase overall being negotiated with Parliament at the time of press, with Parliamentarians arguing for a smaller increase and reduced operational budget for PLN. At the same time, the realization of the first stage of the 10,000 megawatt generation crash construction program will stabilize and start lowering PLN's generation costs. 7. The focus of risks is shifting towards the medium-term While short-term risks The distribution of risks to Indonesia's near term outlook has diminished over the last three have stabilized , there are months following the continued period of robust economic growth and relatively stable risks around whether financial markets. The focus is now on Indonesia's capacity to meet further robust growth Indonesia can sustain, or in domestic demand into the medium-term and the policy challenges associated with increase, current levels of growth into the medium ongoing significant capital inflows. Without further investment in physical and human term, capital, as well as other reforms to enhance productivity, there are risks around whether Indonesia can average growth rates above 7 percent by mid-decade. a. If demand exceeds supply capacity there may be a reemergence of inflation The short-term risks on The risks around a rapid slow-down emanating from Europe have decreased since the last the downside have Quarterly, with the focus shifting towards the weakness in the recovery in the US. Given diminished... that Indonesia's external exposure is relatively low, the impact from a more protracted slowdown is likely to be limited. In terms of the risks to domestic demand, there is a downside risk for growth if government disbursement rates continue to be less than projected for the remainder of 2010 and into 2011. ... however the continued The sustained growth in domestic demand may feed through into increases in domestic growth in domestic prices with some increase in core inflation seen in recent months. The uncertainty over the demand and capital international outlook for commodity prices further creates risks around inflation prospects inflows is creating policy (see Section B-3 in the June IEQ) while further reforms of administered prices toward their challenges economic cost would also put upward pressure on prices, and the challenge will be for policy makers to again succeed in ensuring that an adjustment in price levels does not translate into a persistent rise in inflation expectations and inflation outcomes. At the same time, the strength of the domestic economy has been one factor which has driven the strong capital inflows of recent months, along with the perceived prospects for exchange rate appreciation. The interest rate differential with developed economies has also played an important role, as it has elsewhere in the region. Monetary tightening through the policy rate in response to inflationary pressures may encourage further capital flows into Indonesia. In turn this would put increasing upward pressure on the exchange rate which could have a dampening effect on externally orientated industries, and BI has indicated that it is seeking to manage this constraint. At the same time, capital inflows are highly sensitive to international financial market conditions, as seen in May's sharp withdrawal of non-resident funds, and to the domestic policy stance. A balance is required in policies so as to address any incipient signs of domestic inflationary pressures while at the same time limiting the impact on the real economy of volatile external flows. THE WORLD BANK | BANK DUNIA September 2010 19 Indonesia Economic Quarterly Looking forward Figure 23: Scenario analysis of risks to GDP growth in 2010 Scenario analysis is a useful way to examine the and 2011 distribution of risks around a central forecast. In this (GDP growth, percent) Quarterly two scenarios were conducted, a high and a low Per cent Per cent scenario (Figure 23). In the high scenario it was assumed 8 8 that global growth picks up faster than anticipated, putting upward pressure on commodity prices aiding Indonesia's external sector. It is also assumed that in this scenario 6 6 that credit growth expands quicker than currently anticipated leader to stronger domestic demand. The higher-than-anticipated strength in Indonesia's economy may also put upward pressure on the exchange rate 4 4 through stronger capital inflows. In this scenario GDP growth is expected to be one-quarter of a percentage point higher in 2010, and one-half of a percentage 2 2 point percent higher in 2011 compared to the baseline forecast. In the low scenario it is assumed that there is a protracted weakness in the high income indebted 0 0 economies, hurting Indonesia's external sector. It is also assumed that commodity prices are weaker than currently 2002 2004 2006 2008 2010 anticipated. Investor sentiment towards Indonesia could Notes: BPS and World Bank staff calculations weaken in this scenario, leading to an exchange rate weakening relative to the baseline. In this scenario, GDP is projected to be Œ percent lower in 2010 and 1 percent lower in 2011 compared with the baseline forecast. b. There is a risk that Indonesia's economy will not grow as quickly as expected if the reform agenda is slowed The pace of reform needs While the short-term risks remain, the robust quarterly growth of recent quarters leads to to continue for growth the question of the sustainability of such growth rates, or indeed if they can be increased, rates above 7 percent to into the medium-term. There are sizeable reform challenges for Indonesia to sustain be achieved by the average growth rates above 7 percent in the medium-term. Indeed, the medium-term middle of the decade development plan (RPJMN) outlines an ambitious reform agenda towards achieving stronger and more inclusive growth. Given that the typical financial market investor appears to have a strongly positive outlook for Indonesia's reform efforts, should outcomes fall short of these expectations there is the risk of a sharp reversal of sentiment with significant outflows responding, again destabilizing domestic market conditions and the exchange rate. Strong investment To illustrate the need for Indonesia to expand the supply potential of the economy through growth, improvements in greater investment in human and physical capital, section B-3 focuses on Indonesia' human capital and medium-term macroeconomic projections, and finds that without strong investment and productivity enhancing productivity growth it is going to be difficult to reach the growth projections outlined in the reforms will be required to reach a higher Government's medium-term development plan. Various policy areas for enhancing potential rate of growth productivity and investment in human and physical capital are analyzed in Section C, for example in relation to health and education, access to finance and labor market regulations. Finally, achieving sustained and inclusive growth going forward must also address the challenges of increased urbanization which is the focus of the final piece in Section C. THE WORLD BANK | BANK DUNIA September 2010 20 B. SOME RECENT DEVELOPMENTS IN INDONESIA'S ECONOMY 1. Update on the ASEAN-China Free Trade Agreement Full implementation of The ASEAN-China Free Trade Agreement, which took force on 20 July 2005, has resulted ACFTA in 2010 drove a in a gradual reciprocal lowering in the import tariffs amongst the signatory countries. On 1 publicized concern that it January 2010, the near complete implementation of ACFTA was achieved by the ASEAN- would prompt a surge of 6 economies, including Indonesia, and China, with zero tariffs applied to 90 percent of cheap Chinese imports but the evidence products or almost all of their tariff lines7. The remaining four countries (Cambodia, Laos suggests that this has not PDR, Myanmar and Vietnam) are to achieve this objective by 2015. been the case with previous tariff reductions The implementation of ACFTA in 2010 has led to a publicized concern over the potential under the agreement for a surge of cheap Chinese imports which would undercut domestic producers. However, the tariff cuts under the ACFTA were relatively small in 2010, and historically the tariff reductions made by Indonesia on Chinese imports are in line with cuts made by other signatories of the agreement which benefit Indonesian exporters. Furthermore, preliminary statistical analysis in this section does not support the view that ACFTA tariff reduction in itself increased the rate of Indonesia's import growth from China, once common trends are controlled for. Instead the increasing trend of imports from China, which are dominated by capital goods, appears to be rather a reflection of robust domestic demand. a. The implementation of ACFTA tariff reductions Tariff reductions Tariff reductions under the ACFTA ­ including by Indonesia ­ began in 2005, with gradual, implemented under the annual cuts in tariff rates. The reductions implemented under the ACFTA are consistent ACFTA since 2005 are with Indonesia's reductions in its tariffs on the imports of other major trading partners consistent with those made under a host of regional agreements ­ the ASEAN FTA (AFTA), the ASEAN-Korea implemented by other partners to the agreement FTA (AKFTA), the Indonesia-Japan Economic Partnership Agreement (IJEPA) ­ and with and also with reductions the unilateral reductions in Indonesia's Most Favored Nation (MFN) preferential tariff rates made by Indonesia under (Figure 24). Indonesia's tariff reductions are also consistent with those implemented by other regional trade other countries, including China, under the ACFTA which are of benefit to Indonesian agreements and via exporters. unilateral reductions in its MFN rates Indonesia cut tariff rates substantially in 2007 and 2009, in line with other economies under the ACFTA. However, Indonesia's average tariff reductions under ACFTA in 2010 were comparatively small. In addition, evidence suggests that historically the majority of goods imported from China have not utilized the preferential tariff rates under the ACFTA. Instead they have applied to export using the Most Favored Nation (MFN) tariff rates, which are ­ on average ­ marginally higher than the preferential rates (Figure 1). It has been suggested that the administrative costs associated with applying under the ACFTA mean that the financial incentives to use the ACFTA rates can be diluted. In particular, most of the preferential tariffs require complicated rules of origin terms. Under the ACFTA, a product imported into the territory of a member country from another member country must satisfy one of the following conditions: the product must be wholly produced or obtained in the exporting member country or that at least 40 percent of the product's value content must originate from a member country. Figures for Indonesia's exporters illustrate the low utilization of preferential tariffs under ACFTA. Of the total USD 11.5 billion exports from Indonesia to China in 2009, only just under a quarter, or USD 2.6 billion, used the ACFTA facility (Antara, March 2010). This is consistent with related studies on Free Trade Areas. For example, Baldwin (2006) noted that the utilization rates under AFTA were about five to ten percent. But, with the differential tariff rates applying under MFN and ACFTA increasing in 2009 and 2010, there may now be sufficient incentive for Chinese importers to increasingly utilize ACFTA rates. In turn, this could lead to increased competition for some domestic industries. However, greater utilization should also lead to benefits for Indonesian consumers, producers and exporters, in terms of lower prices for final and intermediate products. 7 The ASEAN 6 consists of Brunei, Malaysia, Singapore, Indonesia, Thailand, and Philippines. 21 Indonesia Economic Quarterly Looking forward Figure 24: Indonesia's tariff rates on imported goods by Figure 25: Share of Indonesia's non-oil and gas imports trade agreement, simple average (percent) (percent) Per cent Per cent 12 jan-jul 2010 10 2009 8 2008 MFN 6 2007 IJEPA AFTA 2006 4 ACFTA 2005 2 AKFTA 2004 0 2004 2005 2006 2007 2008 2009 2010 0 20 40 60 80 100 china asean japan korea euro usa Source: Ministry of Finance, World Bank calculation Sources: BPS. b. The rising importance of China in Indonesia's imports reflects a range of factors Strong growth of capital China's share of Indonesia's imports has increased steadily since 2004, growing the goods, especially fastest among Indonesia's other FTAs and major trading partners (Figure 25). This is personal computers, similar to the trend observed for the share of China in other countries' total imports. In the laptops, and cellular- first seven months of 2010, China's share of Indonesia's non-oil and gas imports was 18.2 phones, has driven China's rising share of percent, double the 9.5 percent level in 2004. In addition, since 2007, China has become Indonesia's imports Indonesia's largest source of imports, replacing the long-time lead contributions of the Japan and US. The continuing increases in Indonesia's imports from China were mainly driven by capital goods. Products under this category, especially personal computers, laptops, and cellular- phones, expanded most rapidly during 2007 and 2009 when there were also significant tariff cuts. The total share of those three products was only 0.2 percent in 2005. This jumped significantly to 0.9 percent in 2007 and continued rising to reach 2.6 percent in 2010 (Table 8). The increasing imports of these products reflect the strong growth of Indonesia's telecommunications and digital media industry. Sector-specific factors High demand from local consumers for cheap information technology (IT) and shed light important on telecommunication products has opened a big opportunity for China's producers to enter the overall rise in the large Indonesian domestic market. As an example of how large the market is for these Indonesia's imports from products, Buddecomm consulting group (March 29, 2010) reported that Indonesia's China mobile phone service penetration was around 65 percent of the population, meaning that Indonesia's mobile service subscribers had passed 150 million by the end of 2009, up from only 12 million subscribers in 2003. The penetration is predicted to grow to 76 percent or 180 million subscribers in 2010. Meanwhile, the number of internet connections grew strongly by 44 percent in 2009, and estimated to grow by 38 percent in 2010. All these figures give an illustration of the scale of the hardware required to support this industry, which must all be imported due to the absence of local products. The import share of assorted machines and parts from China, including pumps, lifts, conveyors, machinery for plants, and power generating machines also increased from 2 percent in 2005 to above 4 percent in 2009 and 2010. The 10,000 Megawatt electric power projects undertaken by the state electricity company (PLN) were one of the reasons why the power generating machine imports increased in 2009. So, again, this part of the rise in imports from China was not necessarily triggered by the ACFTA tariff reduction. THE WORLD BANK | BANK DUNIA September 2010 22 Indonesia Economic Quarterly Looking forward The import of consumer goods such as textile, clothing and footwear (TCF) from China, which may compete directly with local producers, has actually stagnated as a share since 2007 although the tariff was lowered continuously up to 2010. Meanwhile, the expanded share of fruits imports were also not necessarily because of the tariff reductions, because their tariffs were already zero percent since the initial reductions made in "Early Harvest" 2004. All these cases suggest the need for further empirical investigation into whether the tariff reduction under ACFTA indeed influenced the import growth from China. Table 8: ACFTA tariff reductions and imports from China as a share of Indonesia's total imports by broad economic classification and significant products (percent) Im ports from China as share of Indonesia's total ACFTA tariff im ports ITEM (%) (%) 2005 2007 2009 2010 2005 2007 2009 2010* Raw Material 6.6 5.2 2.1 1.2 4.2 5.2 4.7 4.6 Chemical products 6.2 5.0 2.0 1.8 1.3 1.6 1.8 1.8 Irons & steels 9.9 6.9 4.4 3.2 1.5 1.6 1.0 1.1 Fuels 3.5 2.6 0.6 0.6 2.2 0.8 0.6 0.6 Capital goods 3.6 2.9 0.9 0.2 2.3 3.6 7.1 6.9 Data processing parts 0.5 0.2 0.1 0.0 0.1 0.2 0.3 0.4 Computers & laptops 0.0 0.0 0.0 0.0 0.0 0.1 0.6 0.8 Pow er generating machines 7.0 6.3 3.8 0.0 0.1 0.1 0.4 0.3 Telecom & internet parts 8.8 1.2 0.6 0.0 0.1 0.2 0.2 0.2 Cellphones 7.7 1.2 0.8 0.0 0.1 0.6 1.4 1.4 Lifts & conveyors 1.5 1.7 0.9 0.0 0.1 0.0 0.1 0.1 Pumps 2.8 3.3 1.3 0.9 0.0 0.1 0.1 0.1 Machinery for plants 5.4 3.6 0.6 0.0 0.0 0.1 0.1 0.1 Other machines and parts 3.5 3.0 0.9 0.2 1.7 2.2 4.0 3.4 Transports 25.7 15.1 14.6 14.4 0.3 0.4 0.8 0.7 Consum er goods 11.7 9.0 6.5 5.0 1.1 1.5 1.8 1.8 TCFs 13.4 8.9 6.3 3.7 0.1 0.2 0.2 0.2 Onions 0.0 0.0 0.0 0.0 0.1 0.2 0.2 0.2 Fresh fruits (apple, orange) 0.0 0.0 0.0 0.0 0.2 0.3 0.4 0.4 Other consumer goods 11.4 9.3 6.8 5.6 0.7 0.9 1.0 1.0 TOTAL 9.5 6.3 3.8 2.9 10.1 11.5 15.1 14.6 * Imports in non bounded zones up to May 2010. Sources: Ministry of Finance, BPS, World Bank staff calculation c. ...but initial statistical analysis does not support the arguments that import growth from China has been impacted by the ACFTA tariff reductions A simple statistical model To investigate the effect of ACFTA on Indonesia's import in more detail, we run a simple is used to analyze the statistical model following the approach of Pardo et al (2009)8 in examining the impact of relationship between past AFTA on regional trade. This exercise uses Indonesia's tariff data that are applied for ACFTA tariff reductions unilateral partners (MFN) and partners under preferential agreements. The data is highly and Indonesia's import growth by product disaggregated ­ at the 6 digit Harmonized System classification level ­ with annual data used from 2005 through 2009. The baseline model is designed to test the relationship between import growth and tariff changes at the product level. The impact of changes in tariffs is estimated relative to the trend in import growth from different groups of trading partners and the overall growth of Indonesia's imports in different periods (i.e. after inclusion of group and year fixed effects):9 dln(Mjtg) = g + t + 1 dlnPREFjt +2 dlnMFNjt + jt (1) Mjtg corresponds to the volume of imports of product j to Indonesia, at period t, from group g where dln indicates the growth rate (calculated as the log difference across periods). 8 Pardo, Hector Calvo, et al (2009). The ASEAN Free Trade Agreement: Impact on Trade Flows and External Barriers 9 Fixed effects estimation is a method of estimating parameters from a panel data set. This approach is relevant when one expects that the averages of the dependent variable will be different for each cross-section unit, or each time period, but the variance of the errors will not. THE WORLD BANK | BANK DUNIA September 2010 23 Indonesia Economic Quarterly Looking forward There are five groups of countries. There are four groups representing members of Indonesia's preferential partners ­ ACFTA, AFTA, IJEPA, AKFTA ­ and the fifth group is all other countries. PREFjt indicates Indonesia's preferential tariffs (in percentage points) enjoyed by the exporters of product j in period t, while MFNjt corresponds to the unilateral tariff (in percentage points) that exporters of product j have to incur. Finally, g is the group fixed-effect and t the year fixed effect. 10 As the model variables are expressed in terms of percentage change , the coefficients of the independent variables are interpreted as elasticities. Coefficient 1 represents the elasticity of the growth of imports from group g with respect to changes in PREF tariff, controlling for changes in the MFN tariff. A negative 1 means that falls in preferential tariff rates are associated with growth in imports from the preferential trading partner. The coefficient 2 has a similar interpretation as the elasticity of imports with respect to changes in the MFN tariff, controlling for changes in preferential tariffs. To examine the impact of changes in tariffs under different preferential tariff agreements, particularly the ACFTA, equation (1) is modified to equation (2): dln(Mjtg) = g + t + 1a chn*dlnPREFjt + 1b asean*dlnPREFjt + 1c jpn*dlnPREFjt +2dlnMFNjt + jt (2) Now the coefficient 1a represents the elasticity of imports from China with respect to changes in preferential tariffs under the ACFTA, and similarly for 1b and 1c for the AFTA and Japan FTA; all controlling for changes in the MFN tariff. (The interaction with Korea for the AKFTA is not included for reasons of multicollinearity). Controlling for general The results of the exercise are shown in Table 9. The first two columns include the group import trends and fixed effect to control for the average growth of imports by Indonesia from each group. In changes in MFN tariffs, column one the coefficient on changes in tariffs under ACFTA is associated with falling evidence does not imports from China relative to trend. This runs contrary to the claim of ACFTA leading to support the view that ACFTA tariff cuts were rising imports. However, the coefficient becomes insignificant when controlling for associated with import changes in MFN tariffs (column 2). As expected, Indonesia's import growth by product is growth from China significantly negatively correlated with changes in MFN tariffs. Table 9: Estimated impact of tariff changes on the growth of Indonesia's imports by product line (1) (2) (3) (4) chn*dlnpref 3.86*** 1.74 3.97*** 1.95 afta*dlnpref -0.09 -2.3 -0.02 -2.14 jpn*dlnpref -2.92 -5.28 -2.99 -5.25 dlnmfn -2.66*** -2.54*** group fe* yes yes group-prod fe** yes yes group-time fe obs 180838 180829 180838 180829 R2 0.0009 0.001 0.0015 0.0015 * p<0.10, **p<0.05, ***p<0.1 Sources: World Bank staff calculations, BPS, Ministry of Finance. Notes: * Group fixed effect, ** Group-category product fixed effect. Year fixed effects included in all regressions. Similar outcomes are observed with other specifications. In columns 3 and 4 the regression includes group-product effects, i.e. controlling for general trends in imports of different products from partner groups. This attempt to avoid misspecification and misattributing the impact of changes in the preferential tariffs with possibly correlated group specific trends in the imports of individual products (for example, the fall in tariffs on capital goods may coincide with the rise in demand for capital goods from China). 10 For tariffs, the dln means dln(1+PREF/100) and dln(1+MFN/100). THE WORLD BANK | BANK DUNIA September 2010 24 Indonesia Economic Quarterly Looking forward Similar results were also observed when looking at sub-samples of products, such as capital goods and consumer goods; and also when looking at annual growth rates of Indonesia's monthly import data up to May 2010. However, it should be noted that these are initial results. However, these preliminary results are in line with the findings of Pardo et.al. (2009) on AFTA that suggested that imports appear to be more negatively affected by changes in MFN tariffs than changes in preferential tariffs. d. Look at the ACFTA as the media to exchange market access ACFTA should be These initial findings suggest that, on the basis of previous tariff cuts, fears of an viewed as an aggressive influx of cheap China's products into country because of tariff reduction under opportunity to ACFTA do not appear to be statistically proven. The pattern of imports from China, which strengthen and enhance is dominated by capital goods and rising as a share of Indonesia's total imports, is rather a economic, trade and investment co-operation reflection of robust domestic demand. between Indonesia and China Similarly, the increasing Indonesia's exports to China in the last five years largely reflects China's enormous hunger for natural resources to fuel its economy rather than the utilization of tariff reductions made in 2005. The structure of Indonesia's exports to China has been relatively unchanged, dominated by resource-based commodities, such as energy, including oil, gas, and coal; metals; and estate crops (Table 10). Thus the ACFTA should be viewed as an opportunity to strengthen and enhance economic, trade and investment co-operation between Indonesia and China. Among other things this includes greater technological transfer and greater specialization and investment within the region. Many of the benefits from the agreement are expected to accrue over the long-term ­ and as such are difficult to quantify, for example as Indonesian firms realize new market opportunities in China or improve their competitiveness relative to an expanded set of imports.. Table 10: ACFTA tariff reductions and exports to China as a share of Indonesia's total exports by broad economic classification and significant products (percent) China's tariff to Export to China as Indonesia share of w orld export ITEM (%) (%) 2005 2007 2009 2010 2005 2007 2009 2010* Oil & gas 4.7 4.6 2.4 2.1 3.2 2.6 2.2 0.7 Agriculture com m odities 12.0 7.9 3.3 0.4 1.3 1.9 2.5 1.5 Rubber 5.4 5.1 0.2 0.0 0.4 0.6 0.6 0.6 Palm Oil 0.0 0.0 0.0 0.0 0.6 0.8 1.4 0.5 Mining & m ineral com m odities 3.4 3.4 0.1 0.0 0.6 1.3 2.7 3.4 Copper 3.9 3.9 0.0 0.0 0.4 0.3 0.3 0.3 Coal 4.2 4.2 0.0 0.0 0.1 0.4 1.8 2.6 Forestry products 2.7 2.6 2.1 1.2 1.0 0.8 0.7 0.6 Paper and Paper Products 7.5 7.5 5.0 0.0 0.2 0.2 0.1 0.1 Pulp and Waste Paper 0.0 0.0 0.0 0.0 0.4 0.4 0.4 0.3 Manufacture products 8.7 7.0 3.4 1.3 1.7 1.8 1.8 1.4 TCF 12.3 9.4 5.7 2.4 0.2 0.2 0.2 0.2 Electronic & parts 7.1 5.5 3.0 1.4 0.3 0.4 0.4 0.2 Chemicals 6.2 5.6 1.4 0.7 0.9 1.0 0.7 0.7 TOTAL 8.3 6.6 3.0 1.0 7.8 8.5 9.9 7.5 TOTAL Non-oil & gas 5.9 7.2 8.7 10.0 * Export up to May 2010. Sources: Ministry of Finance, BPS, World Bank calculation THE WORLD BANK | BANK DUNIA September 2010 25 Indonesia Economic Quarterly Looking forward 2. Piecing together insights into the movements in household purchasing power Recent increases in The sudden price increase of several food items in recent months, accompanied by inflation and particularly declines in consumer confidence, have raised concerns about their impact on household poverty basket inflation consumption levels, particularly poor households, due to their adverse effect on have raised questions household purchasing power. A household's purchasing power reflects its ability to about the impact on households' consume goods and services. It is determined by household income and the price of the consumption, particularly goods and services that the household consumes. The discussion below sets these that of poor households recent developments in the context of previous episodes of falling consumer confidence and rises in inflation and within the longer-term trends of rising household consumption levels. a. During the previous decade households experienced large increases in real consumption Real consumption over Recent movements in purchasing power are better understood in the context of longer- the longer term has term trends in consumption growth as well as the medium-term outlook. Real consumption increased for all levels and how they evolve each year can be observed through the annual Susenas households but the household survey. Figure 26 illustrates that between 2001-2009 real consumption growth was less for poor households increased across all ventiles (20 population groupings of 5 percent each ordered by consumption level). However, consumption growth was lower for poorer households. The average real consumption growth for the bottom ventile (the poorest group) was 1.6 percent, compared with the average for the entire population of 2.3 percent. The growth rate for the bottom ventile was around one-third of the growth for the highest consumption ventile. Nevertheless, real consumption growth for poorer households contributed to a fall in the poverty rate, which has dropped from 18.4 percent of the population in 2001 to a low of 13.3 percent in March 2010 ­ a reduction of 6.9 million people. The profile for real consumption growth across the different ventiles does however vary depending on the time-period. For example, over the period of 2006-2009 the highest expenditure ventile experienced the lowest growth rate while the middle ventiles had the highest growth rate. Over the same period as the Susenas data, aggregate real private consumption in the national accounts grew at an annual rate of 4.2 percent. This trend is expected to strengthen in the medium-term with real private consumption growth forecast to be above 5 percent in 2010 and 2011 (see Table 2). Figure 26: Real consumption growth 2001-2009 (annual average growth by ventiles ordered lowest consuming household to highest) Per cent Per cent 5 5 4 4 3 3 Average 2 2 1 1 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Lowest consuming Ventiles Highest consuming Source: Susenas (2001-2009) THE WORLD BANK | BANK DUNIA September 2010 26 Indonesia Economic Quarterly Looking forward b. Consumer confidence indices and inflation measures provide real-time insights into developments in real consumption Consumer confidence The Susenas data depicted above only provides an annual snapshot of real consumption indices and monthly and is published with several months delay. The once-a-year reporting is not sufficient for inflation movements policy makers who require more timely and frequent insights into developments in provide additional household purchasing power. Accordingly, alternative data sources are required to garner insights into developments in insights into real consumption growth between the annual data points reported in the household real Susenas. Consumer confidence indices (CCI) are helpful for this purpose as they act as a consumption proxy indicator of households' likelihood of undertaking consumption and are produced monthly with only a few weeks lag by BI and Danareksa. Similarly, the change in the cost or price to consume a set of goods or services is a key determinant in consumption decisions, and this is summarized in the monthly inflation series produced by BPS. CCI and inflation broadly The degree of their correlation with the detailed Susenas real consumption data is a map real consumption gauge of the high frequency CCI's and inflation measures' usefulness as proxies for movements in the developments in real consumption. The correlation (or lack of) between the CCI and real Susenas, making them consumption can be examined by overlaying the monthly movements in the CCI on the useful as real-time and high frequency proxies of annual Susenas real consumption growth (Figure 27 and Figure 28). Over the past eight real consumption years, it appears that the CCI of the different income groups broadly reflects the real consumption growth profile of the relevant households, particularly for the highest consumers of the Danareksa survey. The correlation of the lowest expenditure group in the CCI was not as strong in 2007 and 2008 and this may reflect government programs such as the temporary unconditional cash transfer BLT (Bantuan Langsung Tunai) for poor households or the piloted conditional cash transfer program PKH (Program Keluarga Harapan) which improved real consumption but may not equally affect respondents views on economic conditions ­ a key component in the CCI. Overall, the higher frequency data are a useful indicator for real consumption developments as they provide real-time insights and intra-year movements. Figure 27: Consumer confidence for the highest expenditure Figure 28: ... similarly for poor households group broadly tracks their real consumption growth... (consumer confidence index, real consumption growth) (consumer confidence index, real consumption growth) 24 Per cent Index 120 Per cent Index 8 120 Consumer Confidence Consumer Confidence (700k-1.5 m IDR) (> 1.5 million IDR) (RHS) (RHS) 12 100 4 100 0 80 0 80 -12 60 -4 60 Real consumption Real consumption growth (highest 10% growth (lowest 10% of consumers) (LHS) of consumers) (LHS) -24 40 -8 40 Jan-02 Jan-04 Jan-06 Jan-08 Jan-10 Jan-02 Jan-04 Jan-06 Jan-08 Jan-10 Sources: Danareksa and Susenas Sources: Danareksa and Susenas Note: The consumer confidence index is smoothed to Notes: The consumer confidence Index is smoothed to reduce monthly volatility (two month moving average) reduce monthly volatility (two month moving average) THE WORLD BANK | BANK DUNIA September 2010 27 Indonesia Economic Quarterly Looking forward Consumer confidence Bank of Indonesia and Danareksa each produce CCI, which generally track in the same levels produced by BI and direction but the outlook for Danareksa is usually more pessimistic than the BI index Danareksa move in (Figure 29). This likely reflects the larger representation in the Danareksa survey of similar directions but the households who spend less than IDR 5 million a month, including respondents who spend Danareksa CCI is generally lower due to a less than IDR 1,000,000 a month, which the BI survey does not capture at all. The BI higher proportion of poor survey also excludes rural households who comprise around 30 percent of the Danareksa households and the survey. Both CCI are highly sensitive to movements in inflation with a negative inclusion of rural relationship between the two, whereby price shocks (temporary or permanent) are households in their followed by large contemporaneous declines in consumer confidence (Figure 30). The sample greater coverage of poor and rural households in the Danareksa CCI leads to a significantly greater correlation with the poverty basket inflation rate than the headline rate while the BI Consumer Confidence Index is more highly correlated with the headline inflation rate. A key feature of the relationship between CCI and both inflation measures is CCI rebound just as quickly when the effects of a temporary inflation increase unwind. Figure 29: Consumer confidence measures fell in recent Figure 30: ...as inflation increased months... (consumer confidence index inverted; inflation year-on-year (index) percent) Index Index Index YoY 120 120 55 18 Inflation BI Consumer (RHS ) Confidence 100 100 75 12 80 80 95 6 Average consumer Danareksa confidence* Consumer Confidence (LHS Inverted) 60 60 115 0 Aug-02 Aug-04 Aug-06 Aug-08 Aug-10 Aug-02 Aug-04 Aug-06 Aug-08 Aug-10 Sources: BI and Danareksa Sources: BPS, Danareksa, BI Note: CCI are constructed by a Balanced Score Method (net Note: The average consumer confidence series is the simple balance + 100), where the index is above 100 points indicate average (midpoint) of the BI and Danareksa consumer optimism (positive responses) and vice versa confidence indices shown in Figure 29 Consumer confidence As was the case with real consumption growth (Susenas data), there are also differences levels vary according to in consumer confidence when examined by expenditure level. This is apparent when income group and setting comparing the headline BI and Danareksa CCI, but also explicitly by their series for with lower levels for rural different expenditure levels. A breakdown of the Danareksa CCI by expenditure levels and poor households reveals that poorer household's confidence levels are consistently lower (Figure 31). Similarly, consumer confidence levels are generally lower for rural consumers (Figure 32). This is partially explained by the higher share of poor households in rural settings and may also reflect broader connectivity related issues associated with rural and remote areas including lower job prospects, poor infrastructure and less competition for goods and services. These factors negatively impact rural consumers' purchasing power as well as businesses ability to produce products at competitive prices. THE WORLD BANK | BANK DUNIA September 2010 28 Indonesia Economic Quarterly Looking forward Figure 31: Poorer households' consumer confidence is Figure 32: Rural households also have lower consumer lower and fell more severely with recent price increases confidence , again falling sharply during recent price rises (consumer confidence index levels by household expenditure) (consumer confidence index levels by setting) Index Index Index Index 100 100 100 100 > 1.5 million IDR Urban 90 90 90 90 80 80 80 80 Rural 70 70 70 70 < 500,000 IDR 60 60 60 60 Aug-06 Aug-07 Aug-08 Aug-09 Aug-10 Aug-06 Aug-07 Aug-08 Aug-09 Aug-10 Source: Danareksa Source: Danareksa c. Recent increases in some volatile food items particularly affected the purchasing power of poor and rural households BI and Danareksa CCI in Susenas data for 2010 will not be available until early 2011, which make the CCI and recent months indicate inflation measures crucial to gain a timely understanding of the impact of the recent price sizeable falls and appear increases in volatile food items on real consumption. In recent months, BI and Danareksa to be largely associated CCI indicate sizeable falls (Figure 29) and appear to be largely associated with the spike with the recent food price increase in inflation (Figure 30). The Danareksa Consumer Confidence Index fell for the third consecutive month in August and remains below the peak of August 2009. According to BI, consumer confidence fell because of rising consumer expenditure on basic needs reducing their ability to consume durable goods. The recent food price The recent food price hike led to a greater decline in the confidence of households who increase had a more spend less than IDR 500,000 a month (Figure 31). This is not surprising given the higher severe affect on poor share of food that poorer households consume in their basket of goods and services. This households' consumer reflects a similar divergence between the headline inflation rate and the poverty basket confidence levels inflation rate in recent months (see Appendix) which indicates that poor households faced higher inflation rates and subsequently experienced larger falls in purchasing power, feeding into the greater drop in consumer confidence. However, consumer confidence levels should recover over the next few months given the temporary nature of the recent increase in certain volatile food items. In conclusion, several Overall, a number of different indicators point to a fall in household purchasing power and high frequency indicators consumer confidence in recent months on the back of higher food prices but as those suggest the recent food prices unwind purchasing power is expected to continue improving. The impact was most price increases led to sharply felt by poor and rural households, which consume a greater share of food in their declines in consumer confidence and consumption basket and generally possess fewer savings to withstand the temporary purchasing power but as shock. Longer-term increases in real consumption provide a necessary context to those price rises unwind understand recent developments, as previous falls in purchasing power and consumer and strong economic confidence triggered by sudden movements in prices were temporary. Over the previous growth continues decade strong economic fundamentals drove rising household consumption levels, purchasing power is despite periods of volatile inflation. Therefore, the fall in purchasing power in recent expected to continue months should also be temporary, considering the temporary nature of the recent price improving in the medium- hike. The challenge for policy makers in the medium-term is to help develop greater term market mechanisms to deal with temporary supply side shocks whilst providing temporary and targeted relief in the short-term for those most adversely affected. THE WORLD BANK | BANK DUNIA September 2010 29 Indonesia Economic Quarterly Looking forward 3. Medium-term macroeconomic projections Raising longer-term One of the aims of policy makers is to raise the longer term growth trajectory of their economic growth is a key economy in order to improve the living standards of the population. This is particularly objective of policy important for countries such as Indonesia where income levels are relatively low. In order makers to raise the potential growth of the economy it is important to understand what are its main drivers and how these are expected to evolve in the medium term. Credible medium-term projections of economic growth may also be useful to feed into the private sector outlook and investment decisions and for government planning in other areas, for example the potential fiscal balance in the medium term. A supply-side framework A useful way of quantifying Indonesia's potential medium-term output is through the use of is useful in estimating the a supply-side framework based on the inputs to production. Major inputs include physical medium-term growth capital, human capital, land, energy and materials. A supply-side framework focuses on potential of the economy the way in which these inputs interact with one another, and the efficiency at which they are used. This framework can help policy makers by highlighting the need to improve constraints facing investment in infrastructure, investment in human capital through better targeted education spending, and other regulatory or microeconomic reforms which may lead to stronger productivity growth. Part C of this publication deals with a number of such reform challenges facing Indonesia, for example, in the areas of access to finance, education, health and labor regulations. This note focuses on the This rest of this section sets out a supply-side framework for estimating and forecasting estimation and the potential GDP. This note outlines the methodology for estimating and projecting potential projections of Indonesia's growth in Indonesia. The note attempts to draw policy implications from the resulting potential GDP estimates, and then outlines the advantages and disadvantages of this particularly approach. This note finds that for the Government to reach its growth targets laid out in its medium term development plan (RPJM) then strong rates of investment growth are required, combined with improving productivity. a. Drivers of potential growth in the economy The potential rate of The potential rate of economic growth can be defined as the trend rate of growth in the growth in the economy is productive capacity of the economy. It can also be interpreted to be the trend rate of the trend rate of growth in growth that is consistent with stable inflation. Therefore when the economy is operating productivity capacity above trend growth, or when output growth exceeds that of productive capacity, there tends to be inflationary pressures. The potential rate of The potential rate of growth in the economy depends on the quantity and quality of inputs growth in the economy to production, which are available in the economy, and the efficiency with which they are depends on the quantity used. The inputs can include physical capital such as plant and equipment, labor, material and quality of inputs to and energy. The more of these inputs available in the economy, and the higher their production which are available in the economy, quality, the greater the potential output of the economy. A simple way to analyze potential output is to focus on two key factor inputs, namely labor and capital (as in the famous "Solow" model of economic growth). The relationship between these two inputs and real GDP growth is assumed to be summarized by a simple mathematical production function. While it is possible to have production functions for each sector of the economy, and then aggregate the results to obtain a potential GDP estimate, the approach here is a top-down estimate where there is only a production function for the whole economy. The production function can be manipulated to decompose the sources of growth using a technique known as "growth accounting". Total factor productivity The efficiency with which inputs are used in production, referred to as total factor is a major driver of productivity (TFP), can be the driver of potential growth. Examples of more efficient uses growth of resources are numerous. Through research and development (R&D) firms may be able to produce more or new products by rearranging their existing capital and labor. When new technology is transferred from overseas, for example via foreign direct investment, domestic firms may be able to adopt and adapt this technology to enhance the efficiency of their production processes to produce more from their existing inputs. When distortions are removed from sectors of the economy by the lifting of inefficient market regulations or changes in inefficient taxes then resources may shift between sectors leading to a more efficient allocation of production. When measured TFP often captures more than these efficiency improvements because it is often calculated as a residual. THE WORLD BANK | BANK DUNIA September 2010 30 Indonesia Economic Quarterly Looking forward b. Growth accounting revisited Total factor productivity A growth accounting exercise uses an assumed production function to decompose is measured as being the measured output growth into the contribution from the measured growth of inputs and a difference between input residual component which is taken to represent growth in total factor productivity. This and output growth framework can then be used to obtain an estimate for potential output going forward. To obtain projections for potential output, projections for the factor inputs and total factor productivity are required. The Organisation for Economic Co-operation and Development (OECD) and the Congressional Budget Office in the United States use this approach for their estimates of potential output.11 The Cobb-Douglas To compute estimates for potential output using the growth accounting framework, a production function is standard production function (Equation 1) is employed, where , , , represents output, commonly employed in total factor productivity, capital inputs, and labor inputs respectively. One version of this economics function in Equation 2, known as the constant returns-to-scale "Cobb-Douglas" function is commonly employed in economies due to its convenient mathematical properties. (Equation 1) (Equation 2) If the natural log is taken of Equation 2 followed by the total differential, then Equation 3 outlines the core equation behind growth accounting % % 1 % % (Equation 3) Where % , % , % , % are the percentage change in GDP, total factor productivity, labor force and capital stock respectively. 1 is the labor share of economy and is the share of the economy that is returned to capital. This equation states that growth in GDP is the weighted share of labor and capital plus total factor productivity growth. Therefore to project GDP in the medium term using this framework only three variables are required, capital stock, labor force and total factor productivity. Unfortunately there are Capital stock data for Indonesia is very limited, however fortunately a number of no official estimates of researchers have attempted to calculate their own in the absence of an official estimate capital stock from BPS. This note uses a capital stock series for 1980-2007 that was constructed by 12 Van Der Eng (2008) using the perpetual inventory measure (PIM). The PIM is calculated by taking an existing measure of the capital stock, then sums annual investment for each year, while adjusting for depreciation. This series was extrapolated to 2009 using gross fixed capital formation data. The labor force data comes from the Survei Angkatan Kerja Nasional (Sakernas) but, for reasons of simplicity, is not adjusted for labor quality, for example in terms of educational attainment. Real GDP data is obtained from the BPS quarterly National Accounts. Given there are no income accounts in Indonesia it hard to obtain an estimate of the labor and capital shares of output so a 67 percent labor share has been assumed, this is consistent with OECD (2008) which estimated potential GDP in Indonesia.13 Using the data for real GDP, capital and labor, the growth in total factor productivity can be calculated as a residual from Equation 3. This means that any growth that is not accounted for in movements in the weighted share of labor and capital is recorded in TFP. Total factor productivity Total factor productivity grew by around 0.8 percent in 2009. The 2000s decade average averaged around 2.2 for TFP growth was around 2.3 percent.14 Apart from 2009 when TFP growth was 0.8 percent during 2000s percent, growth was trending up over most of the decade (Figure 33). Smoothing the estimates for TFP, capital and labor using the Hodrick-Prescott filter removes the effects 11 See, for example, Congressional Budget Office (CBO) (2001), `CBO's Method for Estimating Potential Output: An Update'. 12 Van Der Eng, P. (2008), `The sources of long-term economic growth in Indonesia, 1880-2007', Australian National University, Working Papers in Economics & Econometrics 13 OECD (2008), `OECD Economic Surveys ­ Indonesia Economic Assessment', pp 49-51. 14 As an illustration of the sensitivity of the estimates to the model assumptions, increasing the labor share by 10 percent, increases the decade average TFP growth estimate by around 0.2 percent. THE WORLD BANK | BANK DUNIA September 2010 31 Indonesia Economic Quarterly Looking forward of the business cycles and allows an estimate for potential GDP. Using this methodology, in 2009 Indonesia's potential GDP grew by around 5.2 percent , compared with actual growth of 4.5 percent. In terms of contributions to total GDP growth, consistent with other studies, it was found that input accumulation was responsible for much of Indonesia's growth from the 1980s. For all of the 2000s input accumulation contributed around 55 percent of total GDP growth. Figure 33: Potential GDP versus actual growth growth since 1980 (percent annual growth) Per cent Per cent 15 15 10 10 Real GDP 5 5 0 0 -5 -5 TFP Potential GDP -10 -10 -15 -15 -20 -20 1981 1985 1989 1993 1997 2001 2005 2009 Sources: World Bank staff calculations and BPS c. Baseline projections The short-term This framework can be used to produce projections for potential GDP into the medium projections from Part A term. For 2010 and 2011 projections in Part A for gross fixed capital formation are used to are used to extend the calculate the capital stock until 2011. ILO projections for the working age population are projections for total also used to calculate employment growth for 2010 and 2011. This is combined with the factor productivity and hence potential GDP World Bank near-term real GDP projections of 6.0 percent and 6.2 percent for 2010 and 2011 to calculate a total factor productivity series which extends until 2011 (Figure 4). If the Government's From 2012 until 2014 long-term assumptions have to be made regarding the capital stock, RPJM projections for the labor force and total factor productivity. For the baseline projections, it is assumed GDP, and investment are that GDP grows at the Government's projections assumed in the medium term assumed, this implies development plan or RPJM. it is assumed that annual investment increases at the rate strong TFP growth in the medium term consistent with the Government's RPJM. By the end of the projection period investment is assumed to grow above 10 percent. World Bank projections for the working age population are then used to calculate employment growth until 2015. Total factor productivity is assumed to change to meet the Government's RPJM projections. Growth In Indonesia's The RPJM projects growth to increase from around 6.0 percent today, to around 7œ economy relies heavily percent by 2014 (Figure 4). The RPJM relies on significant increases in investment. This on the RPJM investment will however require a continuation and deepening of reform efforts undertaken by the targets Government to encourage private sector investment, both domestic and international, and to address the country's significant infrastructure needs. THE WORLD BANK | BANK DUNIA September 2010 32 Indonesia Economic Quarterly Looking forward Figure 34: Projections of Indonesia's potential growth (percent annual growth) Per cent Per cent 8 4 WB Forecasts RPJM 6 3 projections Real GDP (LHS) TFP (RHS)* 4 2 2 1 0 0 2000 2002 2004 2006 2008 2010 2012 2014 Notes: RPJM projections are mid-points of range. *Implied TFP projections based on RPJM GDP projections and RPJM investment Sources: World Bank staff calculations and BPS Scenario analysis Scenario analysis is important to understand the possible variations around the central illustrates the sensitivity projection. Scenarios were conducted around the central RPJM projection, where of the growth projections investment and total factor productivity growth were able to vary. In the low investment to assumed investment growth scenario it was assumed that investment growth averages around its 2000 rate and TFP growth rates over the projection period, while in the high scenario it is assumed investment growth averages around rates prior to the Asian Financial Crisis in 1998. The low TFP growth scenario assumes that TFP grows around 1.5 percent on average over the projection period, while the high TFP growth scenario assumes that TFP accelerates to around 3 percent over the projection period (Table 11). Table 11: Sensitivity analysis (annual average growth rate over 2012-14 unless otherwise specified) Investment Varies (TFP as in RPJM)TFP varies (Investment as in RPJM) Average 20122014* Low RPJM High Low RPJM High Investment 8.0 11.0 12.0 11.0 11.0 11.0 TFP 2.2 2.2 2.2 1.5 2.2 3.0 GDP 6.6 7.0 7.3 6.0 7.0 7.4 GDP/cap (USD 2010 prices)* 3470 3530 3570 3410 3530 3570 *GDP/cap at 2014 Sources: World Bank staff calculations and BPS. If TFP is assumed to stay at the rate required to meet the baseline RPJM projections, then under a low investment scenario, GDP could average around 6.6 percent, whereas if investment grows at a more rapid pace, GDP could average as high as 7.3 percent over 2012-2014. If investment is assumed to stay at the rate required to meet the RPJM projections then under a low TFP scenario GDP could be as low as 6 percent, while under a high scenario it could grow as strong as 7.4 percent. d. Advantages, disadvantages and further work The simplicity of this One of the key attractions in using this framework to make growth projections is that it is approach is both an based on a very simple supply side representation of the economy, where potential output advantage and is the sum of growth rates of only three factors of supply, labor, capital and total factor disadvantage productivity. However, this simplicity can also be a drawback, for example the labor and factor shares were assumed to remain constant in this approach rather than econometrically estimated or sourced from input-output data. More sophisticated approaches can be used to estimate the production function econometrically but are not without their own technical problems. THE WORLD BANK | BANK DUNIA September 2010 33 Indonesia Economic Quarterly Looking forward This approach rests on Another difficulty for this approach is that it relies on data for capital stock, when there are the quality of the data for no official estimates of capital stock in Indonesia. Estimating the capital stock using the factor inputs, particularly perpetual inventory method relies on accurate fixed investment figures and the usage of the capital stock data appropriate assumptions on depreciation. It also does not control for the quality of the capital stock. Even in countries where official capital stock data are available such problems remain. Similar issues relate to the data on labor inputs and the difficulty of adjusting for the quality of human capital. Disaggregated In this exercise potential GDP was estimated at the aggregate level. This framework can approaches for be extended to estimate potential output for each of the economic sectors. Estimating estimating potential sectoral output can provide further insight into the drivers of economic growth which are output can be employed not captured at the aggregate level. For example, capital stocks for different sectors may be of different quality and vintage. This is especially important given that some sectors would have a higher information technology penetration requiring different depreciation rates. However, the data difficulties mentioned at the aggregate level are amplified at the sectoral level in terms of obtaining reasonable capital stock and labor data. e. Implications and conclusions The Government needs to There are a number of key points to take away from this exercise. For the Government to maintain its commitment reach its growth targets laid out in its medium term development plan (RPJM) very strong to reform for the rates of investment growth are required, combined with improving productivity. For this to economy to expand happen it requires that the government maintain its strong reform commitment outlined in toward 7 percent by the middle of the decade the development plan. Reforms such as lowering entry barriers for new business by reducing the time and cost to start a business as well as encouraging more foreign investment through fewer restrictions on foreign ownership are especially important. Polices that spur job creation in the formal non-agriculture sectors which tend to have higher productivity growth could also contribute to advance aggregate economy productivity. Without these reform measures the growth of potential output may not meet the government's targets. This also has implications for inflation looking forward. If domestic demand continues its strong recent growth then inflationary pressures may eventuate in the coming years as the economy is estimated to grow more than the growth in potential output in both 2010 and 2011. The Government in its Considerable attention is given in the RPJM to both the better mobilization of all main medium term types of economic inputs as well as measures to improve productivity. The advantage of development plan (RPJM) this approach is that it sends a strong signal to both public sector agencies as well as the set out a strong private sector that the government gives high priority to creating the environment for a commitment to undertake these reforms sound expansion of economic activity. On the inputs side, it is recognized that for many reasons, economic inputs such as land, labor, capital, and entrepreneurship are often not mobilized effectively in Indonesia. The RPJM notes that numerous problems relating to the acquisition of and the efficient use of land are holding back much investment. Labor is seen to be underutilized as well. In recent years the bulk of new entrants to the labor force have been absorbed in low productivity activities in the informal sector in small and micro enterprises. Further, it is noted that large amounts of investment will be needed to underpin faster economic growth. Measures to improve the investment climate for both domestic and foreign investors are therefore listed as a priority. And the crucial role that entrepreneurs play in promoting, especially, private sector growth is acknowledged in the growth strategy set out in the RPJM. THE WORLD BANK | BANK DUNIA September 2010 34 C. INDONESIA 2014 AND BEYOND: A SELECTIVE LOOK 1. Improving Access to Financial Services: Analysis and ideas for policy-makers Access to formal financial Access to formal financial services is widely recognized as critically important to increase services is fundamental economic and financial inclusion, reduce income inequality and alleviate poverty around to increasing financial the world. However, only about half of Indonesia's population has access to formal and economic inclusion financial services. Commercial banks, which dominate the Indonesian financial sector, and helping to reduce poverty and inequality serve a relatively small proportion of Indonesian households. Less than half of Indonesia's population saves at banks, while a mere 17 percent of Indonesians borrow from banks. Increasing access to formal financial services does not only lead to social and economic benefits for households but it is also advantageous to the Government, as well as the commercial banks. Owing to a number In Indonesia, there are large areas of unmet demand where consumers want formal factors, there are large financial services but are unable to obtain them, owing to issues such as lack of areas of unmet consumer appropriate products and geographic isolation. Past policies have focused on maintaining demands for financial overall stability while neglecting accessibility, even though both can be achieved services in Indonesia simultaneously. This section provides an overview of the data, analysis, and recommendations for policy-makers that can assist in improving access to financial services in Indonesia. a. What is access to finance/financial inclusion and where does Indonesia stand? Access to finance and Microfinance has gained immense popularity in recent years and micro credit operations financial inclusion refers are flourishing worldwide. Small-scale loans given to entrepreneurs from poor households to an individual's or have been instrumental in helping them generate income and exit poverty. However, enterprise's accessibility contrary to common perceptions, access to finance is not limited to micro credit. From to financial products and services such as savings, microfinance is often linked to the broader concept of access to finance and financial loans and insurance inclusion, which refer to an individual's or enterprise's accessibility to financial products and services such as savings, loans, transaction channels and insurance. It includes financial asset-building in the form of savings products, as well as access to products providing financial risk mitigation of unexpected events through insurance schemes. About half of Indonesia's Research indicates that about Figure 35: Share of the population with formal financial population does not have half the population in access access to formal financial Indonesia does not have services Percent access to formal financial 100 services. Indonesia has lower levels of financial access than 80 regional comparators such as 60 Malaysia, Thailand and Sri Lanka. However, Indonesia is 40 better placed than Bangladesh 20 and the Philippines (Figure 35). To look into the degree of 0 access to finance in Indonesia Pakistan Indonesia Philippines Bangladesh India Thailand Malaysia Korea Singapore Nepal China Sri Lanka in more detail, the World Bank has conducted a study of the demand-side and supply-side aspects of access to financial services, along with the Sources: World Bank (2008, 2009b) current regulatory 15 framework. 15 See World Bank (2009a), "Improving Access to Financial Services in Indonesia" (available at http://go.worldbank.org/ZYCJR4KHE0). For an international perspective see, for example, World Bank (2008), "Finance for All: Policies and Pitfalls in Expanding Access" (available at http://go.worldbank.org/7BAOC2SP90) and World Bank (2009b), "Banking for the Poor, Measuring Access in 54 Economies". 35 Indonesia Economic Quarterly Looking forward b. Demand-side aspects: what products and services do people need? What is currently available to them? The single most Results of household surveys indicate demand for a wide range of financial services. Of important financial these, the single most important financial service is a bank account; more than 40 percent service is a bank account of respondents have a bank account compared with only 15 percent who have a bank loan. The primary reason stated for wanting or holding a bank account was `security'. In terms of access to the services, commercial banks, while having a wide regional reach, do not penetrate deeply into the poorer strata of Indonesian society. Relatively high levels of Sixty eight percent of Indonesians save, but only 47 percent do so using formal banks informal saving reflect (Figure 36). These data point to a large potential market for commercial banks. Limited poor perceptions of reach to bank customers is not caused by physical accessibility to bank branches/ATMs. formal savings products The problem lies in the poor perception of saving products such as bank accounts offered by commercial banks. These are viewed as costly to maintain and there is low financial literacy regarding formal means of saving. High monthly fees and high minimum account balance requirements are also impediments to access. Figure 36: Savers' financial inclusion in Indonesia Figure 37: Borrower's financial inclusion in Indonesia (means of saving, percentage out of total population) (Means of borrowing, percentage out of total population) 68% Financially Included 60% Financially Included 32% Financially Excluded 40% Financially Excluded 0% 20% 40% 60% 80% 100% 0% 20% 40% 60% 80% 100% Banks Other Formal Only Informally Don't save Banks Semi-Formal Informally Voluntarily Excluded Can't Borrow Source: World Bank (2009a) Source: World Bank (2009a) Sixty percent of Household surveys on the demand for loans indicate that a fairly large proportion (60 Indonesians borrow percent) of Indonesians borrow money (Figure 37). However, only 27 percent of the money, but only population does so from a formal bank or microfinance institution. The rest borrow using 27 percent use formal informal means, such as friends and family. Survey results also show that formal sources institutions of finance are used for business loans, whereas informal sources are tapped for consumption purposes. ...with wide variation on Interest rates charged on loans vary widely. Loans from commercial banks can be loan interest rates obtained at a rate of about 25 percent per annum. This is followed by loans from microfinance institutions (MFIs) and community welfare schemes at over 40 percent. High interest rates are also viewed as one of the main obstacles to finance for small and medium enterprises which are a main source of employment in the economy (see discussion in Box 3). It is noteworthy that banks and some MFIs charge markedly less interest if the borrower has a bank account. Thus, a simple way to reduce costs for borrowers is to open bank accounts, as this serves as an indicator for creditworthiness. THE WORLD BANK | BANK DUNIA September 2010 36 Indonesia Economic Quarterly Looking forward There is a dramatic Turning to insurance Figure 38: Take-up of insurance is highly geared towards difference in insurance services, there is a salaried employees take-up between dramatic difference in take- (insurance ownership as percentage of sample group) agricultural workers and up between agricultural salaried employees workers and salaried employees (Figure 38). 30 Private health Life The latter are more than 10 Asset insurance insurance times as likely to buy insurance insurance as the former. In 20 Indonesia, insurance sales Education are heavily tilted towards insurance the urban, upper-income groups. The bulk of 10 insurance is accounted for by the compulsory coverage of government 0 workers and travel insurance. Thus, voluntary Agriculture Sector Worker Salaried Employee insurance, where customers are willing to Source: World Bank (2009a) pay the premium, is low. Survey results suggest a demand for micro-insurance products for poor households that are geared towards low-cost protection from illness and poor business performance, such as harvest failures. Box 3: Focus on Micro, Small and Medium Enterprises (MSMEs) MSMEs are widely recognized as crucial for the economic development of countries. International evidence indicates that they generally account for about 98 percent of all enterprises and employ about 60 percent of the private sector workforce. Figure 39: MSME lending relative to total loans However, they face numerous barriers in accessing financing. Percent The results of a Bank Indonesia survey show that the main obstacle for micro businesses is collateral and for SMEs are 54 high interest rates. Their greatest needs from the 52 Government are for credit, training and market information. 50 52 51 Loans to MSMEs in Indonesia comprise a significant 48 50 proportion of banks' total loan portfolios. The proportion of 49 49 such loans is currently at around 50 percent (Figure 39). 46 47 44 People's Business Credit or KUR (Kredit Usaha Rakyat) is a newly established government program to help farmers and 42 43 MSMEs with credit accessibility. KUR provides risk mitigation 40 for commercial banks and thus accelerates primary sector development and empowers small-scale businesses. 38 Through KUR, commercial banks can lend to MSMEs with a 2002 2003 2004 2005 2006 2007 2008 70 percent credit guarantee. As of January 2008, KUR disbursement was Rp 1,397 trillion through six executing Sources : Bank Indonesia banks. There is great potential for this program to expand and improve credit accessibility for poorer households. c. Supply-side Aspects: Who are the market players and what is their role in providing access to financial services? The different players in Indonesia has various market players in the banking system. In summary, these include the banking sector each the first-tier banks, which are the commercial banks, and second-tier banks, such as have a potential role to people's credit banks (BPRs) and BKDs (Badan Kredit Desa, Village Credit Institutions). play in improving access Other players include cooperatives, LDKPs (Lembaga Dana Kredit Pedesaan, Rural Fund to finance and Credit Institutions), and other non-banking financial institutions (NBFIs). Commercial banks can take two important steps in extending wider access to financial services. As innovators, the banks are the most likely institutions to introduce new technologies to deliver relatively high-cost services to low-income clients in costly, remote THE WORLD BANK | BANK DUNIA September 2010 37 Indonesia Economic Quarterly Looking forward areas. The banks will also be helpful in extending access because they will put competitive pressures on the other service providers, thereby holding down prices and improving the quality of services. In terms of micro-finance, BRI's Unit Desa is Indonesia's premier micro-finance provider and has a large client base. There is great potential for it to better serve lower income clients. A policy decision on the part of management may be needed regarding the current fee structure and lending policies, which can be viewed as hindering Unit Desa's ability to reach potential customers with low incomes. People's Credit Banks (BPRs) offer great opportunities for wider access to finance for poorer households and MSMEs. BPRs are relatively low-cost operations and are close to the community, as they have better local knowledge than regular commercial banks. However, lack of human-resource capacity hinders the extent to which BPRs can serve low-income households. Also, these rural banks have limited operations, as they are restricted to certain geographic areas. The Know-Your-Customer In 2001, Bank Indonesia (BI) mandated Know-Your-Customer (KYC) principles for policy was designed to commercial banks and BPRs in an effort to strengthen transparency and increase enhance transparency customer information. KYC regulations that demand the presentation of a full set of but hinders access to documentation to access financial services reduce access to financial services for lower- finance for lower-income households income households. Identification requirements such as identity cards (KTP), driver's licenses and taxpayer numbers are difficult to obtain, especially for those with minimal resources. To address such concerns over their impact on access to finance, a flexible approach in complying with KYC and Anti-Money Laundering issues has been implemented in other countries, such as South Africa. d. The role of ICT in improving access to finance Information and The role of information and communication technology (ICT) is essential in improving communication access to financial services without increasing costs for the institutions. Banks have technology (ICT) has an numerous ICT options to replace transactions using bank branches. Of these the most essential role in common are ATM machines that can accept, store and dispense cash, as well as point of improving access to financial services source (POS) devices such as mobile phones. However, as ATMs must be regularly re- filled and emptied, it is more cost effective to place them in densely populated areas with a large number of customers. Thus, banking through mobile phones is one of the best 16 low-cost options to reach potential customers in remote villages . Mobile phone banking Recent years have seen a boom in mobile phone banking operations to improve financial has been adopted outreach in developing countries. There is a transformation from cash to electronic value, successfully in other stored and conveyed by mobile phones. In Kenya, the mobile wallet service offered by developing countries Safaricom attracted 1 million registered users in 10 months (in a country where fewer than 4 million people have bank accounts). In the Philippines, the country's two leading mobile network operators offer small-scale transactional banking to an estimated 5.5 million 17 customers . High mobile phone With Indonesia's wide geographic spread, reaching the unbanked poor in remote areas penetration rates offer using cost-efficient means is an issue. Because of its wide reach, mobile-phone banking much promise in offers much promise in providing financial services to the poor in rural regions. Mobile- providing financial phone penetration in Indonesia is at 37 percent of the population, whereas only services to the poor in rural Indonesia 8.4 percent of Indonesians have fixed-line telephones. But, mobile financial Indonesia is moving ahead with mobile phone banking services and as of end-2007, 23 service provision to the banks offered different kinds of mobile banking services to their customers. But these poor remains relatively services are limited to existing customers and do not reach out sufficiently to the limited and is constrained unbanked population. by certain regulatory factors 16 See Consultative Group to Assist the Poor, CGAP, "Using Technology to Build Inclusive Financial Systems" (2006). 17 See CGAP's "Regulating Transformational Branchless Banking: Mobile Phones and Other Technology to Increase Access to Finance" (2006) THE WORLD BANK | BANK DUNIA September 2010 38 Indonesia Economic Quarterly Looking forward Indonesia is also among the few countries with regulations allowing non-banks to issue e- money through popular cellular provider Telkomsel. However, on the whole, the services that mobile-phone providers can offer do not meet all the needs of the poor. Specifically, these limitations include no cash-out and no person-to-person fund transfer capabilities. In countries such as the Philippines, remittances from one person to another can be made easily using mobile phones. However, in Indonesia the current regulatory framework does not permit this. Currently, fund transfers are only permitted from one bank account to another. e. Policies to promote access to finance Various policy reforms There are a number of policy areas on both the demand- and supply-side which could be could be employed to addressed in order to promote and facilitate access to finance in Indonesia: promote access to finance In terms of regulations, changes in the policy framework for commercial banks will help in improving access to finance for poor households. For example, Know-Your-Customer could be simplified for small accounts. Similarly, the requirements for taxpayer numbers could be waived for small loans below a pre-specified threshold. Other areas for policy focus could include the provision of incentives for insurance companies so that they actively pursue micro-insurance among rural populations. Suitable insurance products should be available to meet demand from lower-income groups. The TabunganKu (My The Government has placed high importance on the issue of improved access and the Saving) product is a authorities are initiating policies aimed at overcoming constraints to formal financial welcome move to services. The launch of a new saving product called TabunganKu (My Saving) in early improve access to 2010, is a step in the right direction as this provides access to banking services for the savings millions of Indonesians who have the financial capacity to save money, even if only small amounts18. However, a continuation of these efforts on a larger scale is needed to provide rural populations with the incentive to save using formal sources such as the commercial banks. The economic impact of TabunganKu should also be assessed. TabunganKu is currently limited to being a good entry point for unbanked populations. However, ideally, incentives should be provided for customers to begin banking with normal commercial banks. Facilitating the expanded Finally, there are a range of regulatory issues related to the promotion of financial services usage of ICT to improve through improved and expanded usage of ICT. To use ICT such as mobile phones and access to finance is provide cheap services to improve outreach, the economies of scale offered by retail another important area agents are vital to reduce unit costs to a commercially viable level. As Indonesia has the for policy world's fourth largest population, there is great potential to promote mobile-phone banking operations and reduce costs. Thus, a necessary condition for success in mobile banking is a regulatory framework that permits commercial agencies to operate on a large scale. As mentioned earlier, the regulatory framework should also allow for a wider range of financial services, such as person-to-person money transfers. 18 TabunganKu is a `no frills' account with zero administrative fees and a minimal initial deposit of Rp 20,000. THE WORLD BANK | BANK DUNIA September 2010 39 Indonesia Economic Quarterly Looking forward 2. Labor Reform in Indonesia ­ Balancing Job Creation and Employee Protection 19 The pace of job creation Indonesia has enjoyed a demographic dividend over the past forty years. The working must be accelerated to population has been growing faster than the population of non-working dependents. This employ Indonesia's presents a major opportunity to employ a workforce, which will grow by an estimated growing workforce. 20 million workers over the next ten years. The pace of job creation in the formal and non- agricultural sectors, however, needs to be accelerated in order to provide better employment opportunities for the growing workforce. Today's policy makers face challenges in identifying which policies and programs will spur the creation of good jobs while, at the same time, ensuring that workers are better protected from the risks that threaten their job security. Reforms to improve labor regulations, infrastructure, and the investment climate are all important elements for a national "pro-jobs" strategy. Attempts to reform labor regulations, however, have been deadlocked, slowing Indonesia's ability to generate better jobs. The Manpower Law The Manpower Law (No. 13/2003) substantially increased the rigidity of Indonesia's labor increased the rigidity of regulations. It increased severance rates for workers with three or more years of service Indonesia's labor and added a 15 percent gratuity payment (Figure 40). As a result, severance pay is regulations estimated to be equivalent to a "hiring tax" of around one third of a worker's annual wage. This compares to a "hiring tax" of around 2 months of wages in 1996 and 3.4 months in 2000.20 While the Manpower Law made it more difficult for employers to terminate or reallocate employees, it also restricted firms' ability to hire temporary employees. The use of fixed-term contracts (FTCs) and outsourcing services was restricted to non-core positions and the maximum length of a temporary contract was reduced from five to three years. The law also introduced some welcome changes. The minimum wage-setting process was improved by reforming the use of price surveys and strengthening the role of local wage councils. Indonesia's current Cross-country surveys of regulations indicate that severance pay rates in Indonesia and severance pay rates are above the levels in China, Malaysia, Philippines and Vietnam for example (Figure 41). high, both by While most Asian economies limit the use of FTC work to certain activities and stipulate international and both the duration of the contract and the conditions for renewal, Indonesia falls into the regional standards group of countries with more restrictive regulations governing FTC work, alongside Cambodia, the Philippines, and Vietnam.21 Figure 40: The Manpower Law made it more costly for Figure 41: Indonesia's firing costs are high relative to employers to terminate or reallocate employees regional comparators (amount of severance pay received after years of service) (firing costs in number of weeks of salary) Calculated months of salary Number of weeks 30 120 1996 Law 2000 Law 2003 Law 100 25 80 20 60 40 15 20 10 0 Phillipines Thailand Vietnam East Asia & China Indonesia Malaysia 5 Pacific 0 <1 3 5 10 20 Max Years of service Source: UNPAD/GIAT (2004). "Indonesia's employment Source: World Bank, Doing Business 2010. For more details Protection Legislation: Swimming Against the Tide?" p.18 see www.doingbusiness.org 19 This section is based on Chapter 4 of World Bank (2010), "Indonesia Jobs Report ­ Towards Better Jobs and Security for All". 20 LP3E FE UNPAD and GIAT (2004), "Indonesia's Employment Protection Legislation: Swimming Against the Tide?" Faculty of Economics, University of Padjadjaran Bandung and Growth through Investment, Agriculture and Trade (GIAT) Project. 21 Based on country-specific regulation data from the ILO (LABORSTA on-line database, 2008) on the duration of temporary contracts and the conditions under which they are permitted. THE WORLD BANK | BANK DUNIA September 2010 40 Indonesia Economic Quarterly Looking forward The current stalemate Recent analysis has shown that the regulations have not been effective in protecting traps workers and employees who are terminated and face unemployment. Only 34.4 percent of all eligible employers in a lose-lose employees who separated from a job in the last two years reported that they received any situation that leaves severance pay (Figure 42). Of those employees who received severance pay, employees inadequately protected and constrains 78.4 percent reported that they collected less than amount to which they were legally job creation. entitled. The burden of non-compliance falls disproportionately on workers who, arguably, need more income protection: women, temporary staff and low-wage employees (Figure 43). Small firms are more likely to avoid compliance because they are too small to form unions and fall under the radar of labor inspectors. These current regulations not only leave workers unprotected but may dampen job creation too. Indonesia's high de jure severance rates deter foreign investment and discourage entrepreneurs from creating new businesses. The business community in Indonesia has argued that excessively generous severance regulations stunt job creation by worsening the investment climate and discouraging the creation of new businesses. To avoid high firing costs for permanent employees, firms are increasingly relying on fixed- term contract arrangements, including temporary staff and outsourced employees that are significantly less likely to be eligible for both severance pay and employer-provided pensions. Figure 42: Severance regulations have not been effective in Figure 43: ...especially those who are more in need of protecting employees protection (percentage of employees who received severance pay) (eligible workers who report not receiving severance pay by group, percent) Compliant: Percent employees received full 100 entitled amount or 80 more 8% 60 Non- compliant: employees Non- 40 received compliant: less than employees received no 20 entitled amount severance 25% pay 0 67% Male Female <250 250 - 500 - 1,000 - >15,000 500 1,000 15,000 Sex Monthly wage ('000 Rp) Source: Sakernas, 2008 Source: Sakernas, 2008 International evidence Empirical evidence of the actual impact of increased rigidity of job creation is not available indicates that rigid labor for Indonesia because data on severance payment and contract status has not been regulations hamper job consistently collected. International research, however, generally finds that developing growth, worsening labor countries with onerous labor regulations also experience lower rates of investment, market outcomes for 22 workers productivity and investment in manufacturing. Rigid labor regulations can hamper job growth by limiting the benefits of trade openness and discouraging entrepreneurs from starting new businesses. This has a direct, negative impact on workers. Developing countries with rigid labor regulations are more likely to experience lower (male) participation in the labor force, lower employment rates, and high unemployment rates ­ especially among women and young people. 22 Djankov, S. and Ramalho, R. (2008), "Employment Laws in Developing Countries". World Bank. THE WORLD BANK | BANK DUNIA September 2010 41 Indonesia Economic Quarterly Looking forward Labor reform efforts There is scope to find a win-win solution that works both for employers and employees. need to focus on finding This section presents reform options that may be considered in an effort to reach a win-win solution for conciliatory agreement between employers and employees. The reform options are firms and unions, while presented not only with the interests of employers and unions in mind, but also consider ensuring that vulnerable workers are better the voice of vulnerable workers who have been excluded from the debate. Although poor protected. and low-income workers ­ especially those working in small-sized firms ­ are among the most vulnerable to unemployment shocks, they are the least likely to be protected through the current system. Employer associations and unions represent their members around the negotiating table, but not necessarily informal workers and contract-less employees who would benefit from policies that expand job creation in the formal sector. By negotiating a bargain First, simplifying the legal complexities of current severance regulations and adjusting ­ lowering severance rates downward will bring Indonesia in-line with regional standards. This will not only rates in exchange for improve labor market flexibility but it will also improve Indonesia's investment climate and introducing global competitiveness. At the same time, efforts should be made to simplify severance unemployment benefits ­ the Government can calculations in order to make it easier for employees to understand what they are entitled improve labor market to and easier for firms to know what they are responsible for paying. flexibility while increasing protection for Complementary new programs are needed, however, to provide effective protection for employees unemployed formal sector workers in compensation for reduced severance rates. Shifting towards a monthly contributory approach ­ where firms make monthly contributions to an account that is managed centrally with government oversight ­ will improve the predictability of firms' labor costs without influencing their hiring and firing decisions. It will also likely increase the level of compliance by employers, which will help reduce reliance on the Industrial Relations Courts that are backlogged with disputes concerning employment termination. This will be a relief to both employees and employers who find the dispute resolution process costly and time-consuming. There are a range of Indonesia is ready to begin following the lead of other middle-income countries that have unemployment benefit adopted unemployment benefit systems. There is a range of reform options that can systems that should be improve the predictability of labor costs and compensate workers for lowered severance considered and assessed rates. The options include a pooled fund from which terminated employees can draw, an for inclusion in the future National Social Security individual account severance system, or a flat-benefit unemployment assistance program system (Box 4). Each option has its respective advantages and disadvantages, and varies in terms of the level of institutional complexity to manage the program. The reform process can be started by initiating the necessary analytical studies to identify which option best fits Indonesia. Simulation studies are needed to assess the anticipated impact of the alternative systems and the institutional implications and demands associated with each reform option. Based on the best-suited model, a reform roadmap is needed to lay the foundations for the future system, which should be linked to the future National Social Security (NSS) system mandated by the Social Security Law (No. 41/2004). THE WORLD BANK | BANK DUNIA September 2010 42 Indonesia Economic Quarterly Looking forward Box 4: Severance reform options Option One: Severance Fund: Firms make regular severance payments to a pooled fund that is administered by a central government agency or private firm(s). Terminated employees receive severance according to years of service. Individual pooled funds may be created for each firm, or shared among all contributing firms. Option Two: Individual accounts: Employers and employees regularly deposit contributions in individual accounts managed and disbursed by a central agency. Unemployed contributors draw from their own accounts upon confirmation of unemployment status. Option Three: Flat-benefit unemployment assistance: Create a fund that eligible workers can draw from when unemployed. Funds are managed and disbursed by an appointed agency, not the employer. Unemployed workers receive a low benefit from a common fund for a specified period of time. Eligibility based on active job search and availability of suitable work. Possibility to make benefits contingent on means-tested family income. Source: Revenga and Rigolini (2007) and Vroman (2007).23 Easing the rigidity of Even with major labor regulation reforms, many workers will still remain employed without current regulations can a formal contract or in the informal sector for the foreseeable future. For this reason, while contribute to spurring regulatory reform is necessary, it is not sufficient to improve the prospects for most job creation in the formal Indonesian workers. Additional strategies are therefore needed to empower and protect sector, but complementary Indonesia's excluded and vulnerable workers, including developing a comprehensive skills strategies are still development strategy and designing effective safety nets to protect vulnerable workers needed. from wage and employment shocks. The World Bank's Indonesia Jobs Report (2010) examines the performance of Indonesia's labor market over the last twenty years and discusses how regulatory reform, skills development and safety nets can support the acceleration of job creation while providing more effective protection for all workers. 23 See Revenga, A. and Rigolini, J. (2007), "International Evidence on Severance Pay Reforms: Some Food for Thought for Indonesia's Current Reform Proposal", World Bank, and Vroman, W. (2007), "Reforming Social Protection for Workers in Indonesia", World Bank Office Jakarta. THE WORLD BANK | BANK DUNIA September 2010 43 Indonesia Economic Quarterly Looking forward 3. Education, training and labor market outcomes for youth in Indonesia Demand for skilled Indonesia's robust growth performance and prospects is likely to be accompanied by an workers in Indonesia has increasing demand for educated workers as production away from agriculture towards the increased in recent increasingly sophisticated manufacturing sector and the services sector. At the same time decades but there is a the education levels of those leaving their studies and entering the labor market has mismatch with the skills of new workers entering increased. However, surveys indicate that the skills of secondary school leavers do not the labor force meet the expectations of employers. Concern over the quality of graduates includes those from both the General and Vocational secondary education streams. Meeting the challenge of matching the supply with the demand for skills is a key ingredient in enhancing labor market outcomes for the youth of Indonesia and improving productivity in the economy more generally. a. Economic context: growth, transformation and the demand for skills Fast growth and rapid Over the last two decades Indonesia has experienced fast sustained growth and rapid urbanization over the last urbanization, significantly increasing the share of non-agriculture GDP despite an two decades significantly important setback during the 1997 financial crisis. The achievements are impressive­GDP increased the share of per capita in real terms increased 160 percent from 1990 to 2008, and the share of non-agriculture GDP agriculture in GDP decreased from 20 to 13 percent by 2006, although this rose slightly to just over 14 percent in 2008 (Figure 44). Figure 44: Rising real income in Indonesia has been accompanied by a fall in the share of agricultural output (GDP per capita, LHS, and share of agriculture in GDP, 1990-2008, RHS) 4,500 22 4,000 20 3,500 3,000 18 2,500 16 2,000 1,500 14 GDP per capita, PPP (current international $) 1,000 Agriculture (% of GDP) 12 500 0 10 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 Source: World Development Indicators (1990-2008). Service and Employment by economic sector followed a similar pattern as the share of GDP, but the manufacturing are low lower labor intensity in the education-intensive manufacturing and service sectors has labor-intensity sectors resulted in slower job creation for educated youth. Almost 70 percent of workers in the but their growth is service sectors who are employed for wages have completed senior secondary education. expected to result in sustained demand for The share in manufacturing is lower ­ 40 percent ­ but this is still double that of skilled workers agriculture (based on Sakernas data 1994-2007). The growing importance of the service sector in the economy and the increased sophistication of the manufacturing sector are expected to result in a sustained demand for skilled workers. The overall trend should ensure sustained, and potentially increasing, demand for skills in the economy. However there is At the macro-economic level, there is evidence of sustained demand for skilled workers evidence that integration but there are signs that integration of educated workers into the labor market is becoming of educated workers into more difficult. Despite the large influx of more educated youth into the labor market, the the labor market is returns to education have remained largely constant in the last decade, indicating that the becoming more difficult, even with the expansion demand for educated workers has so far been able to absorb the increased supply without of education decreasing returns (Figure 45). The supply of educated workers, however, is likely to continue increasing significantly with the expansion of education, so it is critical to ensure that new graduates can access good quality jobs. THE WORLD BANK | BANK DUNIA September 2010 44 Indonesia Economic Quarterly Looking forward Figure 45: The returns to education have remained largely Figure 46: The share of skilled jobs has been constant flat over the past decade during the past decade after rising in the 1990s (wage premium relative to primary education) (skilled, unskilled and skilled production jobs as share of total jobs, 1994-2007) 160% 100% 1994 2001 2007 140% 80% 120% 100% 60% 80% 60% 40% 40% 20% 20% 0% 0% Junior Senior Tertiary 1994 1997 1999 2001 2003 2005 2007 Secondary Secondary Skilled Skilled production Unskilled Source: Indonesia Skill Report 2010, Sakernas 1994-2007. Source: Sakernas Indonesia Skills Report (2008). Note: Only includes salaried employees. Note: Only includes salaried employees. Younger generations are Each year over 3.3 million youth leave the formal education system to enter the labor more educated than ever, market, and younger generations are more educated than ever before. In 2008, the however the occupational proportion of labor market entrants who have completed senior secondary or higher share of skilled jobs has education (that is those considered "skilled") surpassed 50 percent, confirming the trend not kept pace, resulting in difficulties in entering the of increased educational attainment in the population. However, the occupational share of labor market skilled jobs has not kept pace with the increase in education levels. The share of jobs considered "skilled" increased during the 1990s but has remained broadly constant during the last decade (Figure 46), questioning the ability of the labor market to absorb these new graduates at their appropriate education level. This has resulted in over-qualified entrants and a difficult transition into the labor market, especially for senior secondary school graduates who are the least skilled workers. b. A difficult transition to the labor market Youth unemployment rate In Indonesia, the unemployment rate of youth in the 20 to 24 age group is about two and a is higher than the half times that of the overall population, and it is higher for more educated youth, population especially senior secondary school graduates (Figure 47). Over 40 percent of 15-24 year unemployment rate olds with completed senior secondary school in the labor market are unemployed and, although the rate decreases for older age groups, it does not converge to the country average until the 35-39 year age group. Rural areas, with less educated workers and greater availability of unskilled jobs have lower unemployment rates; stressing the current dichotomy in the types of jobs in the urban and rural areas in the country. Even though there is a In terms of type of employment, there is a clear positive association between education clear positive association and formality, but access to salaried jobs is difficult for senior secondary school between education and graduates. Only 60 percent of young senior secondary school graduates in the labor force formality, access to have a salaried job and this share falls with age. Unlike the case with unemployment salaried jobs is difficult for senior and junior patterns, the outlook for workers with junior secondary school or lower qualifications in secondary school terms of the quality of jobs is especially bad, with only one-third of junior secondary school graduates graduates in the labor force holding a salaried job (Figure 48). But while senior secondary school graduates fare better, on average only about 50 percent of senior secondary school graduates are salaried employees. THE WORLD BANK | BANK DUNIA September 2010 45 Indonesia Economic Quarterly Looking forward Figure 47: Unemployment rate by age group and education Figure 48: Share of labor force in salaried jobs by age group level and education level (unemployment rate as percentage of age group, 2007) (percentage of age group, 2007) 60% 100% 50% 80% 40% 60% 30% 40% 20% 10% 20% 0% 0% 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 Primary JSS Primary JSS SSS Tertiary SSS Tertiary Source: Sakernas (2007) Source: Sakernas (2007) Note: JSS­junior secondary school; SSS­senior secondary Note: JSS­junior secondary school; SSS­senior secondary school. school. c. Senior Secondary Education, Skills and Labor Market Entrance Lack of access to The share of workers employed for wages that have a senior secondary education or information about labor higher qualification has increased significantly during the last 15 years (from 35 percent to markets may be a reason over 50 percent), but the growth mainly occurred during the 1990s and has remained for difficulties in constant in the past decade. A lack of mechanisms to access information about the labor transition to work and can lead to inequality in market, returns and types of work available may contribute to the difficulties in transition to labor market outcomes work for young graduates. As senior secondary education expands, the new students progressing through the education system are more likely to come from disadvantaged backgrounds and may have less access to job networks or information about job possibilities from peers. In the absence of efficient mechanisms to gain access to this information, this may lead to inequality in labor market outcomes even if all graduates meet the skill requirements of existing jobs. Also, the observed differences in returns by parental education may also reflect differences in the quality of the education received by students. As an illustration, according to PISA (the Programme for International Student Assessment), average cognitive skills in math and language have increased in recent years, but there are vast differences by socio-economic characteristics, which are largely driven by parental education. Employers are concerned The results from a recent Figure 49: Employers' opinion of the quality of about the quality of Employer Skills Survey confirm employees with senior secondary education graduates from both that the skills of senior secondary (percentage of respondents) senior secondary graduates do not meet the vocational (SMK) and expectations of employers. A Very Good Fair Poor Very Poor general stream (SMA) 100 quarter of recent hires with a senior secondary education are considered of poor or very poor 80 quality (Figure 49). Only 7 percent 60 of them are considered very good, and most of them are considered 40 "fair". While employers are concerned about the quality of 20 graduates from both streams of schools, the types of skills and 0 jobs of graduates vary. Average Senior SMA SMK Secondary Source: Employer Skills Survey (2008) THE WORLD BANK | BANK DUNIA September 2010 46 Indonesia Economic Quarterly Looking forward Despite the differences in In principle, the vocational school (SMK) stream is geared towards more job-specific skills, the skills provided by the seeking to equip graduates with the skills necessary for a quick and effective transition to two education streams, the labor market. The general stream (SMA) provides a more general education, which the unemployment rates serves as the basis for further education. As such, one might expect that if demand for for recent SMA and SMK graduates with no higher specific skills is high, SMK graduates would be better suited for the labor market in their education are high and initial years. However, the unemployment rate of recent SMA and SMK graduates 20-24 very similar years of age is very high (30 percent), with SMK graduates faring only slightly better (Figure 50). In the overall population, SMK graduates used to have significantly lower unemployment rates, but it has converged to the level of SMA graduates. Comparing the labor In terms of income, there is a small positive wage premium for the general stream, market outcomes for the especially when considering the higher transition rate to higher education. The transition two educational streams rate to higher education is higher for the general SMA stream than for vocational school points to the need to graduates (30 percent compared to 15 percent), which, combined with the very high improve the quality of both streams and pay returns from higher education, results in a higher adjusted salary for SMA graduates. attention to equity in There is also a large wage penalty for vocational school for males, especially those with access low ability (as measured by test scores), which has become larger in recent years, indicating either a deterioration of the quality of graduates or a lower demand for their skills. A comparison of the labor market outcomes of both streams does not support a drastic guided expansion of vocational school enrollment, suggesting instead that the priorities should be to improve the quality of both streams and pay special attention to equity in access. Figure 50: Unemployment rates for SMA and SMK graduates, age 20-24 (1991-2007) (unemployment rate as percentage of total graduates from stream) 36 34 32 30 28 26 24 SMA SMK 22 20 1991 1993 1995 1997 1999 2001 2003 2005 2007 Source: Sakernas (1991-2007) Cognitive skills matter Cognitive skills matter more than the education stream for future earnings and better more than the education cognitive skills are associated with higher wages­even more so for general stream stream for future graduates. The exit exam in junior secondary school is a good predictor of future earnings earnings, which implies that a strong knowledge base in basic cognitive skills is critical for success in higher levels of education and ultimately in the labor market. Since cognitive skills are usually correlated with socio-economic characteristics and also affect the future education path (better cognitive skills are more likely to attend the general stream), the inequalities in cognitive skills observed in PISA scores need to be addressed through specific interventions targeting low-performing students and schools in early grades. Employers complain In line with the observed earnings, employers value basic mathematics and reading skills about inadequate generic as very important, but complain about inadequate generic skills, which may be a cause of skills such as thinking the difficult transition to the labor market. Core skills are in high demand, especially basic and behavioral skills mathematics and reading, thinking and behavioral skills, which are the cornerstones of general education. Few employers consider that their employees have a significant gap in their basic skills. However, 40 percent of employers consider their staff to lack thinking and behavioral skills, which points to the need to strengthen generic/life skills for THE WORLD BANK | BANK DUNIA September 2010 47 Indonesia Economic Quarterly Looking forward Indonesian graduates. Vocational skills that are transferable between jobs like computer literacy and English language proficiency are also noted as important gaps in employees' skills. The private sector should While transferable vocational skills should be part of the vocational school curriculum, the participate in the education sector cannot aim to provide all the skills necessary to perform any job in the provision of job-specific productive sector. Private firms should participate in the provision of job-specific skills, skills both through on-the-job training and through copayment of institution-based training for their employees. It is clear from employer surveys that the productive sector values practical on-the-job training more than theoretical training, which is difficult to adapt as quickly as the shifting demands in the market. Non-formal vocational A strong system of non-formal vocational training should serve two purposes: provide training can help reduce tangible skills to unskilled workers and meet the increasing demand for sector/industry- the mismatch between specific training for individuals with completed formal education. The effectiveness of the skills supply and demand system, however, depends on the provision of quality training, the crucial connection with the productive sector and, ideally, some form of interaction with the formal education sector that allows dropouts to return to the path of formal education upon completing basic competencies. Standards for different training courses and providers are not developed and the quality assurance systems are still in very early stages. Resources devoted to training are still inadequate and the coordination between agencies is limited. THE WORLD BANK | BANK DUNIA September 2010 48 Indonesia Economic Quarterly Looking forward 4. PNPM Generasi final impact evaluation ­ the preliminary findings 24 The Government of In 2007, the Government of Indonesia launched a large pilot of two Conditional Cash Indonesia launched two Transfer (CCT) programs applying different approaches: a conditional cash transfer to large pilots of conditional households, and an incentivized block grant scheme to communities. These two pilot cash transfer programs in projects began in six provinces. They are designed to achieve the same objectives and 2007... goals, in line with the Indonesian Government's priorities and the Millennium Development Goals: to reduce poverty, maternal and child mortality, and to ensure universal coverage of basic education. ...one of them is PNPM The community block grant scheme, Box 5: PNPM Generasi target indicators Generasi Sehat dan PNPM Generasi Sehat dan Cerdas, Health Cerdas, a community seeks to improve twelve basic 1. Four prenatal care visits for pregnant women block grant scheme 2. Iron tablet intakes during pregnancy health and education indicators (See which concentrates on 3. Assisted delivery by a trained professional health and education Box 5). Applying the principles of 4. Two postnatal care visits community-driven development, 5. Complete childhood immunizations communities, with the assistance of 6. Monthly weight increases for infants trained facilitators, decide how best 7. Regular weighing for children under five to use the block grants to reach the 8. Vitamin A twice a year for under-fives education and health targets. Education PNPM Generasi is currently 9. Primary school enrollment of all children implemented in 178 sub-districts in between 6 to 12 years old five provinces as part of the 10. Minimum attendance rate of 85% for all primary Government's flagship poverty school-aged children 11. Junior secondary school enrollment of all program, the National Community children between 13 to 15 years old Empowerment Program or Program 12. Minimum attendance rate of 85% for all junior Nasional Pemberdayaan secondary school-aged children Masyarakat (PNPM- Mandiri). Incentives based on past PNPM Generasi takes the idea of performance incentives from CCT programs and performance are used to applies it in a way that allows communities the flexibility to address supply or demand motivate communities' constraints, or some combination of both. As an incentive for communities to focus on choice of the intervention selecting the most effective intervention method, the distribution of 20 percent of the sub- methods to be funded by the block grants district PNPM Generasi block grant for the subsequent year is determined by the village's performance on each of the twelve indicators. Surveys have been used The final PNPM Generasi evaluation survey examines the program's impact after two- to evaluate the impact of years of implementation as well as the effect of community incentives within the program. the pilot program PNPM Generasi was designed to be rigorously evaluated by randomly selecting which kecamatans would receive Generasi and which would remain as a control group. The impact of community incentives can also be assessed by comparing the two versions of PNPM Generasi: the incentivized version and the non-incentivized version that does not link village performance to fund distribution. The baseline survey was conducted in June­ September 2007, followed by an interim evaluation survey in October­December 2008, and a final evaluation survey in October­December 2009, after the program had been in operation for 27-30 months. Generasi had a The final impact evaluation found that PNPM Generasi had a statistically significant statistically significant positive impact on the 12 indicators it was meant to address. The strongest improvements positive impact on the 12 among the health indicators were in the frequency of weight checks for young children. health and education The program also increased the number of iron sachets received by pregnant mothers indicators it targeted through antenatal care visits. These improvements were supported by dramatic increases of mothers and children's' participation in Posyandu activities to receive the targeted maternal, neonatal and child health services. Education indicators also saw improvements in the final evaluation, reversing the zero or negative impact found at the interim evaluation, most notably in the increased school participation rate among the primary school-age group. 24 This section draws on the forthcoming Final Impact Evaluation Report and the Interim Impact Evaluation Report on the PNPM Generasi Program by Benjamin A. Olken, M.I.T. Department of Economics, Junko Onishi, World Bank and Susan Wong, World Bank. THE WORLD BANK | BANK DUNIA September 2010 49 Indonesia Economic Quarterly Looking forward ...particularly in relation The primary long-term impact was a decrease in childhood malnutrition. On average, over to childhood malnutrition the two year implementation, childhood malnutrition was reduced by 2 percentage points or about a 10 percent reduction from the baseline level. This reduction in malnutrition was stronger in areas with higher malnutrition rates prior to project implementation, most notably in the Nusa Tenggara Timur (NTT) Province. However, there was no evidence of an improvement in infant and childhood mortality rates. Although reductions in infant and child mortalities were observed in the interim evaluation, the same rates of decrease were not sustained in the final evaluation. The impact of Generasi The impact of Generasi was most apparent in those areas with low baseline health and was most apparent in education indicators. Lower pre-project levels of health and education indicators imply areas with low baseline more room for improvements. The greater impacts in areas with lower baseline values health and education appeared more prominently in the final evaluation survey than in the interim results, with indicators stronger improvements found in education indicators in such areas. However, these improvements in health and education indicators in areas with low baseline coverage did not appear to have resulted in improving long-term health and education outcomes in these areas. Furthermore, the greater impacts observed in health and education indicators were not simply correlated with pre-project levels of poverty, but instead driven by the level of health and education indicators in the area. The use of incentives Making grants conditional upon performance improves program effectiveness in health. improved program On average, the incentivized group outperformed the non-incentivized group in improving effectiveness in health indicators, particularly in the increase of antenatal care services and improving promoting health coverage of childhood immunization coverage. However, the incentivized group did not outcomes but not for education appear to do better, or worse, than the non-incentivized group on the education indicators. The overall mixed results for the performance of community incentives may suggest that the incentives are effective in certain sectors where communities have greater control over improvements, but not in others. Further expansion should The findings of the impact evaluation point to areas where policies may be adapted or prioritize areas with low modified to further improve performance. First, as mentioned, the results show that health and education Generasi had strong impacts in areas where health and education indicators are low. This indicators suggests that future expansion of Generasi implementation should prioritize areas where these indicators are lagging behind and not necessarily in areas identified as poor. ...which will require Second, although Generasi has proven to be effective in improving performance in stronger coordination lagging areas through community mobilization and strong partnership with local health with the relevant and education service providers, an expansion into less developed areas will pose ministries challenges. In particular it will require stronger coordination with the Ministries of Health and Education. The use of indicators is The planning, targeting, and delivery mechanisms for PNPM can be used as a platform for an effective mechanism other forms of local assistance. Generasi proved that communities are able to understand, in community embrace and effectively work towards improving targets provided by the project. Providing development and they communities with target indicators within the framework of PNPM is an effective delivery can potentially be used to address other issues mechanism for the Government to guide community development. The target indicators must be relevant to communities yet reflect development priorities of the Government. Although it is important not to overload the project, additions to the existing target indicators may allow Generasi to address other government development priorities such as early childhood development or higher education. Further evaluation is Finally, two years of project implementation is a very short timeframe for a development needed to monitor the project to show sizeable impacts. The current Generasi project set up allows for future long-term impact of the evaluation (e.g., after 5 years of project implementation) if the current randomization is project maintained. If the government considers identifying the impact of Generasi in the long- term, planning is required at this time to ensure that the control areas are maintained for future evaluations. THE WORLD BANK | BANK DUNIA September 2010 50 Indonesia Economic Quarterly Looking forward 5. Lumpy, evolving and constrained: Regional and urban economic development in Indonesia Spatially concentrated Over the past two decades, there has been a resurgence of interest in economic urbanization and geography. Policy researchers in both developed and developing countries have launched economic development a range of studies examining the relationship between urbanization, economic are intimately linked 25 development and geography. Spatially concentrated urbanization and economic development go hand-in-hand (Figure 51). Indonesia's growth in recent decades has also been accompanied by increasing urbanization (Figure 52). As the World Bank World Development Report of 2009 noted: "No Country has grown to middle income without industrializing and urbanizing. None has grown to high income without vibrant cities". High levels of population density, and efficient and low cost access to production inputs, have allowed economies to transform from agrarian to industrial and to service activities. Sub- national regions that can manage the rural to urban transition successfully are able to rapidly grow their economies, increase incomes and raise living standards. Those that cannot do so, tend to lag economically and socially. Figure 51: Urbanization and economic development go Figure 52: ...Indonesia also follows this trend hand-in-hand (urban population percentage and GDP per capita across (urban population percentage and PPP per capita real GDP, countries, 2005) 1960-2007, Indonesia) Percent Percent 120 60 y = 14.149ln(x) - 69.619 y = 0.008x + 6.2242 RČ = 0.5274 RČ = 0.9646 100 50 80 40 60 30 40 20 20 10 0 0 0 20,000 40,000 60,000 80,000 0 1,000 2,000 3,000 4,000 5,000 6,000 GDP per capita, US dollars PPP GDP per capita (2005 international dollars) Source: UN DESA World Urbanization Prospects and World Source: UN DESA World Urbanization Prospects and the Bank World Development Indicators University of Pennsylvania Penn World Table, version 6.3 Note: The line is the fit from a regression of the urban Note: The line is the fit from a linear regression model population share on the log of per capita GDP Addressing the policy Uneven patterns of economic development across geographic areas have raised policy challenges of challenges for governments for decades. Many countries have aggressively pursued a urbanization can involve variety of policies to foster more inclusive and spatially balanced growth across their a focus on `people' or territorial space. These include, for example, attempts to control internal migration, the `place' formation of growth poles, financial incentives to induce firms to locate in lagging regions, the creation of economic zones, or establishment of new towns. The overall impact of these efforts is very mixed. Based on their focus, there are two ways to characterize regional development policy. One is on "people" prosperity - the use of policy instruments in areas such as education and human capital development, or promoting labor utility and migration. The other is on building "place" prosperity by spending scarce resources to build up lagging regions. 25 See, for example, the World Bank's 2009 World Development Report: Reshaping Economic Geography, available at www.worldbank.org/wdr2009. THE WORLD BANK | BANK DUNIA September 2010 51 Indonesia Economic Quarterly Looking forward Indonesia has urbanized Indonesia has urbanized Figure 53: UN Urban and rural population projections for rapidly over the past 40 rapidly over the past 40 years, Indonesia, 2010-2050 years, with two thirds of and is on track to join the Million its population projected "urban club"--countries with 350 to live in urban areas by most of their population living Urban Rural 2050 in urban areas­by 2025 300 (Figure 53). The United 250 Nations estimates that by 2050, two thirds of Indonesia's 200 population will live in urban 150 areas, and between 2025 and 2030, the country's rural 100 population will start to decline 50 in absolute terms. 0 2010 2015 2020 2025 2030 2035 2040 2045 2050 Sources: UN DESA, World Population Prospects, 2009 Urbanization can promote Overall, Indonesian policymakers should not be concerned about increasing urbanization. growth through There is strong evidence that urbanization will drive economic growth through the agglomeration economics formation of agglomeration economies. Many policymakers are concerned about the dominance of the Jakarta Metropolitan Region (JMR). But while it is indeed quite large, population projections of the JMR and other large cities between 2010 and 2025 indicate that most population growth will take place in secondary and smaller cities (Table 12). Although Jakarta is still expected to grow, its relative share of Indonesia's total urban population is set to decline due to the faster growth of smaller cities. Large medium agglomerations will grow slowly as well and their share of the total urban population will decline slightly or remain stable. A notable trend is that smaller cities will account for nearly 83.2 percent of Indonesia's total urban population, up from 81.7 percent in 2010. What this means is that while Indonesia will still continue to urbanize, urban concentration will spread to other centers. Table 12: Indonesia's urban concentration is set to spread (projected population and percentage of total urban population) Population Percent of Total Urban Population City 2010 2015 2020 2025 2010 2015 2020 2025 Jakarta 9,210 9,709 10,256 10,850 8.90% 8.70% 8.40% 8.10% Surabaya 2,509 2,576 2,738 2,923 2.40% 2.30% 2.20% 2.20% Bandung 2,412 2,568 2,739 2,925 2.30% 2.30% 2.20% 2.20% Medan 2,131 2,266 2,419 2,586 2.10% 2.00% 2.00% 1.90% Semarang 1,296 1,334 1,424 1,528 1.30% 1.20% 1.20% 1.10% Makassar 1,294 1,409 1,512 1,621 1.30% 1.30% 1.20% 1.20% Other Urban 84,108 92,367 101,169 110,986 81.70% 82.30% 82.80% 83.20% Total Urban 102,960 112,229 122,257 133,419 100% 100% 100% 100% Source: UN DESA, World Population Prospects, 2009 Improved connectivity is The experience of successful countries suggests that inclusive development is about key to promoting reshaping economic geography without fighting economic concentration. From a policy inclusive development by perspective, it makes sense to connect lagging regions to growing large and medium reshaping economic centers to increase demand and to attract industries which are less dependent on geography without fighting economic agglomeration and localization economies such as agriculture, agro-processing, and concentration labor-intensive manufacturing. If there is efficient connectivity, industries with mature product cycle outputs are likely to be attracted to areas with low labor and other input costs. In some areas, agglomeration of activities may eventually develop, leading to the creation of their own (smaller) growth poles in the lagging regions. THE WORLD BANK | BANK DUNIA September 2010 52 Indonesia Economic Quarterly Looking forward The economies of small Output trends between Table 13: Regional GDP distribution by major urban and medium sized urban 1993-2006 also agglomerations, other areas and rural areas, 1993-2006 areas in Indonesia have demonstrate evidence (percent of non-oil and gas GDP) grown faster than the of the spread of larger cities production through a Metro Agglomeration 1993 1999 2003 2006 decentralized Jakarta (Jabodetabek) 25.1 26.1 26.3 25.5 concentration of Surabaya (Gerbangkertosusila) 6.9 6.7 6.8 7.1 economic activity Medan (Mebidangro) 2.5 2.1 2.8 3 Bandung (Bandung Raya) 1.7 1.5 2 2.2 (Table 13). While Semarang (Kedungsepur) 1.5 1.5 2 1.9 Jakarta and Java are Denpasar (Sarbagita) 0.9 0.7 0.6 0.6 still dominant in terms Kota 2.3 2.6 2.3 2.5 of economic output, Balikpapan/Samarinda/Bontang Kota Manado/Bitung 0.3 0.3 0.3 0.3 economic and Makassar (Maminasata) 1.1 1.9 1.9 0.9 demographic trends Other Urban 8.7 8.9 10.3 11.3 points toward a larger Rural 48.8 47.7 44.6 44.8 urban footprint for Total 100 100 100 100 Indonesia. Sources: World Bank staff calculations and BPS Many industrial activities, such as manufacturing and food and beverage production, have begun to decentralize. While the share of non-oil and gas gross regional product as a percentage of the national total for Jakarta, Balikpapan and Manado remained constant, there were slight increases in Surabaya, significant increases in the shares of Medan, Bandung and Semarang, and declines for Denpasar and Makassar. However, the most significant trend in the table is the substantial increase in the share of output accounted for by small and medium sized urban areas. The economies of these cities have grown faster than the large agglomerations and their share have consequently increased from 8.7 to 11.3 percent from 1993-2006. Also notable is the downward trend in rural area economic output, with the share falling from 48.8 to 44.8 percent over the same period. This pattern is consistent with trends in other countries. There are challenges to However, there are challenges that impede the country's economic competitiveness. Indonesia's Considering Indonesia's complex archipelagic geography, regional planning and competitiveness development is challenging. Providing efficient connectivity and accessibility for logistics, including inadequate commercial and demographic mobility requires the formulation of extensive systems of intra- and inter- island transportation networks intra- and inter-island transportation networks. However, in Indonesia's large metropolitan regions, mobility is impeded by rapid motorization and inadequate investments in transportation systems, especially in mass transit. This is also the case with inter-city and inter-island transportation networks, even on Java. A range of factors have led to this situation, including poor planning and implementation, and inadequate investment in transportation systems. ... as well as other Furthermore, other infrastructure services are also inadequate. Many of Indonesia's large infrastructure services cities have inadequate water supply and distribution networks, and extremely limited such as water, waste, wastewater treatment. Power generation and transmission systems are constraining electricity, due to poor economic development in many areas. Many cities and rural areas face significant planning and under- investment exposure to natural hazard risks. These limitations adversely affect economic competitiveness and place undue burdens on the poor and disadvantaged groups. The Government's under investment in infrastructure is an overarching constraint. Currently, the national and local governments devote less than four percent of gross regional domestic product to infrastructure--about half of what is necessary to promote accelerated economic growth. ...while complex urban Urban land markets are complex in Indonesia. This makes it very difficult to assemble land markets hamper the land for industrial estates and commercial centers, constraining the ability to foster the formation of industrial formation of industrial clusters. The absence of clusters impedes growth, because clusters localization economies are weak or non-existent. THE WORLD BANK | BANK DUNIA September 2010 53 Indonesia Economic Quarterly Looking forward Lack of planning, Finally, large and medium cities suffer from inefficient spatial structure. While both the implementation and inter- central and local governments prepare spatial plans, they do not fully implement them by governmental investing in required infrastructure or enforce implementation of plans though zoning and coordination contribute to development controls. In cases where cities straddle districts or provinces, there is a lack inefficient spatial structure of medium and of inter-governmental coordination, despite the adoption of new laws on coordinated large cities spatial planning and management. Accessibility, The forces of economic geography point to increased urbanization going forward. The infrastructure, industrial challenge for the Government is to put in place regional and urban economic development clusters, spatial structure policies to enhance economic performance and promote inclusive development. This and intergovernmental covers infrastructure issues such as enhancing accessibility within and between urban coordination are key to promoting regions and improving urban and regional infrastructure systems (especially competitiveness and transportation, power, water supply and distribution, wastewater collection and treatment). economic development It also relates to policies and institutions which can foster the formation of industrial clusters and economic activity centers and improve the efficiency of urban spatial structure and land markets. Finally, given the cross-cutting nature of regional and urban development issues across different levels of government, effective policies will require enhanced intergovernmental coordination. THE WORLD BANK | BANK DUNIA September 2010 54 APPENDIX: SNAPSHOT OF THE INDONESIAN ECONOMY Figure 1:GDP growth Figure 2: Contributions to GDP(E) (percent growth) (year-on-year, seasonally adjusted) Per cent Per cent Per cent Per cent 4 8 4 4 Year on year (RHS) 3 6 2 2 QoQ seas. adjust 2 (LHS) Average 4 (LHS)* 0 0 1 2 -2 -2 Jun-07 Mar-08 Dec-08 Sep-09 Jun-10 0 0 Discrepancy Net Exports Investment Jun-03 Mar-05 Dec-06 Sep-08 Jun-10 Government Private cons GDP *Average QoQ growth between Q1 2000 ­ Q2 2010. Source: BPS via CEIC and World Bank Sources: BPS, World Bank seasonal adjustment Figure 3: Contributions to GDP(P) Figure 4: Motor cycle and motor vehicle sales (year-on-year, seasonally adjusted) (levels) Percentage point Percentage point '000 '000 0.6 0.6 800 80 Other Motor cycles Q1 Q2 Retail trade (LHS) 600 60 0.4 0.4 Mining and cons 400 40 0.2 0.2 200 Motor vehicles 20 Agric. (RHS) 0.0 0.0 Manufacturing 0 0 comm. and transport Jul-06 Jul-07 Jul-08 Jul-09 Jul-10 -0.2 -0.2 Source: BPS via CEIC Source: CEIC Figure 5: Consumer indicators Figure 6: Industrial partial indicators (index levels) (year-on-year growth) Index Index Per cent, yoy Per cent, yoy 120 270 30 30 BI Retail BI Consumer Industrial sales (RHS) Survey Index Production (LHS) 20 20 230 100 10 10 190 0 0 80 -10 -10 150 Cement Electricity -20 -20 60 110 -30 -30 Aug-06 Aug-07 Aug-08 Aug-09 Aug-10 May-07 Feb-08 Nov-08 Aug-09 May-10 Source: BI via CEIC Source: CEIC 55 Indonesia Economic Quarterly Looking forward Figure 7: Real trade flows Figure 8: Balance of Payments (quarter-on-quarter growth) (USD billions) Per cent Per cent USD bn USD bn 10 10 7 7 Balance Current Account 4 4 0 0 Exports 0 0 Errors -10 -10 -4 -4 Imports Financial & Capital Account -20 -20 -7 -7 Jun-08 Dec-08 Jun-09 Dec-09 Jun-10 Jun-07 Jun-08 Jun-09 Jun-10 Source: CEIC Source: BPS and World Bank Figure 9: Trade balance Figure 10: International reserves (values, USD billions) (USD billions) USD bn USD bn USD bn USD bn 6 15 90 90 Exports (RHS) 4 10 80 80 2 5 Reserves 70 70 0 0 60 60 -2 -5 50 50 -4 Trade -10 Imports Balance (LHS) (RHS) -6 -15 40 40 Jul-07 Jul-08 Jul-09 Jul-10 Aug-07 Aug-08 Aug-09 Aug-10 Source: BPS and World Bank Source: BPS Figure 11: Terms of trade and monthly export and import chained Fisher-Price indices Figure 12: Inflation (index) (month-on-month & year-on-year) Index 2000 = 100 Index 2000 = 100 Per cent Per cent 300 300 5 25 250 250 Food Inflation 4 (RHS) 20 200 200 Poverty Terms of 3 15 Basket Trade Headline Inflation (RHS) 150 150 Inflation 2 10 (RHS) 100 Chained export 100 Core Inflation price Inflation MoM (RHS) 1 5 50 50 (LHS) Chained import price 0 0 0 0 May-04 May-06 May-08 May-10 Aug-08 Feb-09 Aug-09 Feb-10 Aug-10 Source: BPS and World Bank Source: BI and BPS THE WORLD BANK | BANK DUNIA September 2010 56 Indonesia Economic Quarterly Looking forward Figure 13: Monthly breakdown of CPI Figure 14: Inflation amongst neighboring countries (monthly percentage point contributions) (year-on-year, August 2010) Per cent Per cent Per cent Per cent 1.8 1.8 Volatile Administered Indonesia Core Headline Inflation Phillipines 1.2 1.2 China 0.6 0.6 Thailand Korea 0.0 0.0 Malaysia Japan -0.6 -0.6 Aug-08 Feb-09 Aug-09 Feb-10 Aug-10 -2 0 2 4 6 8 Sources: BPS, World Bank *August is latest data point Sources: National statistical agencies via CEIC and BPS Figure 15: Sub-national government bank deposits since Figure 16: Poverty, employment, and unemployment rates decentralization (percent) (IDR trillions) IDR trillions IDR trillions Per cent Per cent 25 40 120 120 Formal Poverty rate (LHS) 20 employment 32 Districts rate (RHS) 90 90 Open unemployment Provinces rate (LHS) 15 24 60 60 10 16 30 30 5 8 0 0 0 0 Mar-01 Mar-04 Mar-07 Mar-10 2006 2007 2008 2009 2010 Sources: BI Source: BPS, Sakernas and World Bank Figure 17: Regional equity indices Figure 18: Broad USD Index and Rupiah spot (daily, rebased index) (daily, index and levels) Index Jan08=100 Index Jan08=100 Index IDR per USD 125 125 120 8500 IDR/USD JCI (RHS) 100 100 110 9500 75 SGX 75 100 10500 SET Dollar Index (LHS) 50 50 90 11500 Shanghai BSE IDR Appreciation 25 25 80 12500 Sep-08 Mar-09 Sep-09 Mar-10 Sep-10 Sep-08 Mar-09 Sep-09 Mar-10 Sep-10 Sources: World Bank and CEIC Sources: World Bank and CEIC THE WORLD BANK | BANK DUNIA September 2010 57 Indonesia Economic Quarterly Looking forward Figure 19: 5 Year local currency bond yields Figure 20: Sovereign USD bond EMBI spreads (daily, percent) (daily, basis points) Per cent Per cent Basis points Basis points 24 24 1200 400 Indonesian EMBI 1000 USD bond spreads 300 18 18 (LHS) 800 200 12 12 600 100 Indonesia Philippines 400 0 6 6 Thailand Malaysia 200 Indo sreads less -100 Global EMBI average (RHS) 0 United States 0 0 -200 Sep-08 Mar-09 Sep-09 Mar-10 Sep-10 Sep 08 Mar 09 Sep 09 Mar 10 Sep 10 Sources: World Bank and CEIC Sources: World Bank and CEIC Figure 21: International commercial bank lending Figure 22: Banking sector financial indicators (monthly, index) (monthly, percent) Index Jan08=100 Index Jan08=100 Percent Percent 170 170 100 10 Loan to Deposit Ratio (LHS) 80 8 150 150 Indonesia India 60 6 130 130 Non-Performing Loans Malaysia (RHS) 40 4 Return on Assets Ratio Singapore (RHS) 110 110 Thailand 20 2 Capital Adequacy Ratio USA (LHS) 90 90 0 0 Aug-08 Feb-09 Aug-09 Feb-10 Aug-10 Jan-06 Nov-06 Oct-07 Aug-08 Jul-09 Jun-10 Sources: World Bank and CEIC Sources: World Bank and BI THE WORLD BANK | BANK DUNIA September 2010 58 Indonesia Economic Quarterly Looking forward Table 14: Budget outcomes and estimates (IDR trillion) 2008 2009 2010 (p) 2010 (p) 2011 (p) Semester I WB Proposed Actual Actual report estimates* budget A. State revenues and grants 981.6 848.8 994.7 1,020.1 1,086.4 1. Tax revenues 658.7 619.9 738.9 747.5 839.5 o/w natural resources 327.5 317.6 360.9 370.0 414.5 - Oil and gas 77.0 50.0 55.3 63.8 54.2 - Non oil and gas 250.5 267.5 305.6 306.2 360.3 2. Non tax revenues 320.6 227.2 253.9 272.7 243.1 o/w natural resources 224.5 139.0 164.7 170.6 158.2 i. Oil and gas 211.6 125.8 151.5 155.3 145.3 ii. Non oil and gas 12.8 12.8 13.2 15.3 12.9 B. Expenditures 985.7 937.4 1,089.8 1,079.3 1,202.0 1. Central government 693.4 628.8 742.4 736.4 823.6 2. Transfers to the regions 292.4 308.6 347.4 342.9 378.4 C. Primary balance 84.3 5.2 5.7 35.9 0.7 D. SURPLUS / DEFICIT (4.1) (88.6) (95.1) (59.2) (115.7) (per cent of GDP) (0.1) (1.6) (1.5) (0.9) (1.7) Note: *World Bank revenue forecasts are based on a different methodology to the Government to derive projections for nominal GDP (see Part C of the June 2010 IEQ for a full discussion) Source: MoF and World Bank estimates. Table 15: Balance of Payments (USD billion) 2009 2010 2007 2008 2009 Q1 Q2 Q3 Q4 Q1 Q2 Balance of Paym ents 12.7 -1.9 12.5 4.0 1.1 3.5 4.0 6.6 5.4 Per cent of GDP 3 -0.4 2.3 3.5 0.8 2.4 2.6 4.1 3.1 Current Account 10.5 .1 10.7 2.5 2.5 2.2 3.6 2.1 1.8 Per cent of GDP 2.4 0.0 2.0 2.2 1.9 1.5 2.4 1.3 1.1 Trade Balance 20.9 9.9 21.0 4.1 5.1 5.0 6.8 4.8 5.3 Net Inome & Current Transfers -10.4 -9.8 -10.3 -1.6 -2.6 -2.8 -3.2 -2.8 -3.5 Capital & Financial Accounts 3.6 -1.8 3.5 1.6 -1.8 2.5 1.3 4.3 3.3 Per cent of GDP 0.8 -0.4 0.6 1.4 -1.4 1.7 0.8 2.6 1.9 Direct Investment 2.3 3.4 1.9 .5 .4 .5 .6 1.7 1.2 Portfolio Investment 5.6 1.8 10.3 2.0 1.9 3.0 3.5 6.2 1.1 Other Investment -4.8 -7.3 -8.8 -.8 -4.1 -1.0 -2.9 -3.6 1.0 Errors & Om m issions -1.4 -.2 -1.8 -.1 .4 -1.1 -.9 .3 .3 Foreign Reserves* 56.9 51.6 66.1 54.8 57.6 62.3 66.1 71.8 76.3 Note: * Reserves at end-period Source: BI and BPS THE WORLD BANK | BANK DUNIA September 2010 59