F INDIA DEVELOPMENT UP DATE Financing Double-Digit Growth June 2016 India Country Management Unit Director: Onno Ruhl Macroeconomics and Fiscal Management Global Practice Manager: Shubham Chaudhuri Comments to: Volker Treichel vtreichel@worldbank.org Frederico Gil Sander fgilsander@worldbank.org The World Bank, New Delhi Office 70 Lodhi Estate New Delhi 110 003, India +91 (0) 11 4147 9301 www.worldbank.org/in Acknowledgements This edition of the India Development Update was prepared by Frederico Gil Sander (task team leader), Luisa Zanforlin and Anuradha Ray (financial sector); and Saurabh Shome, Smriti Seth and Jaba Misra (macroeconomics and fiscal management), under the overall guidance of Onno Ruhl, Niraj Verma, Shubham Chaudhuri, and Volker Treichel. Contributions from P.S. Srinivas (financial sector); Farah Zahir (governance); Ben Eijbergen and Dominic Patella (transport); Rinku Murgai, Urmila Chatterjee, and Yogeshwar Bharat (poverty) are gratefully acknowledged. The team also wishes to thank Paul Cashin (IMF), Mariano Cortes, Allen Dennis, Poonam Gupta, Markus Kitzmuller, Demetrios Papathanasiou, Hans Paris (IFC), Thomas Richardson (IMF) and Sebastian Saez for helpful comments, suggestions, and inputs. Jenny diBiase provided expert editing. Sudip Mozumder, Patsy D’Cruz, and Nandita Roy provided excellent assistance in external relations, web production and cover design, and Sapna John provided outstanding administrative and overall support. Photo credits: Rajarshi Mitra/Flickr (cover, Madhya Pradesh); Bernard Gagnon/Wikipedia (Mallick Ghat Flower Market, Kolkata); Simone D. McCourtie/World Bank (Red Tractor and Yellow Ledger). The findings, interpretations, and conclusions expressed in this report do not necessarily reflect the views of the Executive Directors of the World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. The report is based on information current as of June 3, 2016. i Abbreviations AP Andhra Pradesh ARCs Asset Reconstruction Companies AT&C Aggregate Technical and Commercial BBB Bank Boards Bureau BE Budget estimates BH Bihar BIC Bank Investment Company BOB Bank of Baroda BOI Bank of India CDR Corporate Debt Restructuring CERSAI Central Registry of Securitisation Asset Reconstruction and Security Interest of India CG Chhattisgarh CII Confederation of Indian Industry CIT Corporate Income Tax CNG Compressed Natural Gas CPC Central Pay Commission CPI Consumer Price Index CRAR Capital to risk-weighted assets ratio CRILC Central Repository of Information of Large Credits CRR Cash Reserve Ratio CSO Central Statistics Office CSS Centrally Sponsored Schemes DBT Direct Benefits Transfer DCAs Debtor-Creditor Agreements DDUGJY Deendayal Upadhyaya Gram Jyoti Yojana DICGC Deposit Insurance and Credit Guarantee Corporation DIPP Department of Industrial Policy and Promotion DISCOM Distribution Company DRTs Debt Recovery Tribunals DTLs Demand and Time Liabilities EMs Emerging Markets FC Finance Commission FDI Foreign direct investment FIIs Foreign Institutional Investors FPI Foreign Portfolio Investor FRA Financial Resolution Authority FRBM Fiscal Responsibility and Budget Management FSAP Financial Sector Assessment Program FY Fiscal year GDP Gross Domestic Product GE Grant effect GJ Gujarat GMS Gold Monetization Scheme GNFS Goods and Non-Factor Services GOI Government of India GSDP Gross State Domestic Product GST Goods and Services Tax GVA Gross Value-Added HP Himachal Pradesh HR Haryana IBC Insolvency and Bankruptcy Code IEBR Internal and Extra Budgetary Resources IFC International Finance Corporation IIP Index of industrial production IMD India Meteorological Department IMF International Monetary Fund INR Indian Rupee JH Jharkhand ii JLF Joint Lenders Forum KA Karnataka KL Kerala KPI key performance indicators LAF Liquidity Adjustment Facility LNG Liquified Natural Gas MGNREGA Mahatma Gandhi National Rural Employment Guarantee Act MH Maharashtra MoF Ministry of Finance MP Madhya Pradesh MSPs Minimum Support Prices MW Megawatt MZ Mizoram NEER Nominal Effective Exchange Rate NPA Non-Performing Asset OD Odisha ODE Overall devolution effect OROP One-Rank-One-Pension PFCE Private final consumption expenditure PIT Personal income tax PMJDY Pradhan Mantri Jan Dhan Yojana PPI Private Participation in Infrastructure PPP Public Private Partnership PSBs Public Sector Banks QFI Qualified Foreign Investors RBI Reserve Bank of India RCAP Review of the capital framework RE Revised Estimates REER Real Effective Exchange Rate RJ Rajasthan ROA Return on Assets ROE Return on Equity SARFAESI Securitization and Reconstruction Financial Assets and Enforcement of Security Interests SBI State Bank of India SC Scheduled Caste SCB Scheduled commercial banks SDLs State Development Loans SDR Strategic Debt Restructuring SE Share Effect SEBI Securities and Exchange Board of India SECC Socio-Economic and Caste Census SGB Sovereign Gold Bonds SLR Statutory Liquidity Ratio SOEs State-owned Enterprises ST Scheduled Tribe TG Telengana TN Tamil Nadu TPC Total Project Cost UDAY Ujwal DISCOM Assurance Yojna UK Uttarakhand UP Uttar Pradesh US United States USD United States Dollar UTs Union Territories VAT Value-Added Tax WB West Bengal WDI World Development Indicators WPI Wholesale price index WTO World Trade Organization iii Table of Contents Executive Summary .............................................................................................................................................................................. 1 The Indian Economy in Pictures ........................................................................................................................................................ 2 Financing Double-Digit Growth in Pictures .................................................................................................................................... 4 A. Recent Economic Developments and Outlook ............................................................................ 5 India’s economy expanded even as key growth engines stalled ...................................................................... 5 Domestic demand drives growth as exports continue to plummet ................................................................................... 7 Urban households drove consumption growth, as agriculture lagged .............................................................................. 9 Higher public spending could not avoid a deceleration in investments as private capex struggled ........................... 12 Prospects for activating stalled engines suggest a favorable outlook with significant risks .......................... 14 Agriculture and private consumption lift growth in the near term, followed by private investments ....................... 15 Exports are expected to recover, supporting growth and keeping the current account in check .............................. 17 Risks to the outlook are significant, but buffers mitigate against external shocks ........................................................ 17 B. Special Issue Notes ................................................................................................................... 21 1. India Fiscal Policy Update ......................................................................................................................... 21 Execution of the Union Budget in FY16 was largely as planned..................................................................................... 21 What has been the impact of the 14th Finance Commission on states? ........................................................................ 22 The Union FY17 budget: a strong signal of the importance of macroeconomic stability ........................................... 29 In FY17, modest consolidation at the level of the general government (states + center) ........................................... 31 Annexes ..................................................................................................................................................................................... 34 2. The Rural-Urban Divide in Private Consumption ..................................................................................... 37 A disproportionate share of consumption takes place in urban areas ............................................................................ 37 The rural economy has (partly) diversified from agriculture ............................................................................................ 38 For most rural families, incomes – and consumption – are still linked to agriculture.................................................. 38 Prospects of a normal monsoon should therefore lead to a boost in growth ............................................................... 39 C. Financing Double-Digit Growth ................................................................................................ 41 i. India will need a robust financial sector to finance double-digit growth .................................................. 41 Orderly development of the financial sector and economic growth go hand-in-hand ................................................ 41 India’s financial sector has built up important strengths ................................................................................................... 42 A number of challenges need to be addressed to ensure the financial sector can support growth............................ 45 Addressing the challenges will require both short- and long-run interventions ............................................................ 50 ii. A transformation toward a market-oriented sector is underway and can be accelerated .......................... 51 The participation of foreign and private banks has gradually increased, boosting competition ................................. 51 Government mandates are gradually being phased out, which may help banks’ profitability .................................... 53 Non-banking institutions and capital markets have started to develop .......................................................................... 56 Returns on savings instruments are increasingly market-determined.............................................................................. 58 iii. Corporate and bank balance sheets need to be repaired and strengthened .............................................. 59 Bank and corporate balance sheets are intertwined ........................................................................................................... 59 Governance reforms need to accompany fresh equity for banks .................................................................................... 61 Addressing underlying corporate stress is important to closing the pipeline of future NPAs .................................... 71 iv. Summary of policy options to ensure the financial sector remains a pillar of growth ............................... 77 Structural transformation of the financial Sector ............................................................................................................... 77 Dealing with rising NPAs ....................................................................................................................................................... 77 References ............................................................................................................................................................................................ 78 iv BOXES Box 1: Divergence in input and output price indices .................................................................................................................... 13 Box 2: Revision in State GSDPs ....................................................................................................................................................... 28 Box 3: State Control of Banks – Some International Examples ................................................................................................. 63 Box 4: Governance of Bank Boards in India: Bank Investment Company and Bank Board Bureau ................................... 64 Box 5: India’s New Insolvency and Bankruptcy Code ................................................................................................................. 67 Box 6: Impact of infrastructure loans on SBI ................................................................................................................................ 74 FIGURES Figure 1. Steady contributions from household consumption in FY16 supported GDP growth ........................................... 6 Figure 2. The economy expanded in the fourth quarter but GVA growth has been lower than GDP growth .................... 6 Figure 3. Industrial growth picked up in the second half of the FY, led by manufacturing ..................................................... 6 Figure 4. Three key divergences: domestic vs. external, urban vs. rural, and public vs. private capex ................................... 6 Figure 5. Lower oil prices were not the only drivers of the decline in exports as “core” exports also contracted ............... 7 Figure 6. Travel and IT remain bright spots, but weak merchandise trade spilled over to freight exports ........................... 7 Figure 7. The recent decline in India’s exports has been mostly due to weakness in demand conditions… ........................ 8 Figure 8. … and in the case of services market share has increased............................................................................................. 8 Figure 9. Machinery and agricultural exports declined the most in FY16 ................................................................................... 8 Figure 10. Shipments to the Middle East and Asia posted the largest contractions .................................................................. 8 Figure 11. Lower oil imports explain almost all of the decline in nominal imports, implying volumes increased ............... 9 Figure 12. Growth in services imports also decelerated, but remained positive overall............................................................ 9 Figure 13. Manufacturing led the acceleration of the industrial sector ...................................................................................... 10 Figure 14. The momentum in IP growth was broken in November because of the floods in Chennai .............................. 10 Figure 15. Until recently, sales of passenger vehicles were accelerating… ................................................................................ 11 Figure 16. …while purchases of two-wheelers lagged .................................................................................................................. 11 Figure 17. Urban inflation declined more…................................................................................................................................... 11 Figure 18. …as housing and food inflation abated........................................................................................................................ 11 Figure 19. After spiking in late FY15, the real policy rate started to come down in FY16..................................................... 12 Figure 20. Growth in personal loans has been robust .................................................................................................................. 12 Figure 21. The private sector accounts for nearly ¾ of GFCF ................................................................................................... 13 Figure 22. Capacity utilization has declined .................................................................................................................................... 13 Figure 23. CPI and WPI have diverged… ...................................................................................................................................... 13 Figure 24. … as lower oil prices did not reach consumers........................................................................................................... 13 Figure 25. Consensus forecasts expect a sharper acceleration ..................................................................................................... 14 Figure 26. Growth is expected to gradually accelerate to 7.8 percent in the medium-term.................................................... 14 Figure 27. The output gap remains negative................................................................................................................................... 15 Figure 28. GVA growth will be lifted by a recovery in agriculture…......................................................................................... 16 Figure 29. …while on the expenditure side the growth drivers gradually shift from consumption to investment ............ 16 Figure 30. Trade is expected to recover modestly along with the global economy.................................................................. 17 Figure 31. Higher capital goods imports and oil prices lead to a widening of the current account from FY18 ................. 17 Figure 32. Foreign exchange reserves have been stable ................................................................................................................ 19 Figure 33. The REER has appreciated over the past two fiscal years ........................................................................................ 19 Figure 34. Revenue receipts exceeded budget targets despite under-performance of direct taxes ........................................ 22 Figure 35. Net devolution was unambiguously positive ............................................................................................................... 25 Figure 36. Higher devolution was the main contributor to funding for additional spending in FY16 ................................. 26 Figure 37. States spent additional resources on health and education… ................................................................................... 27 Figure 38. … as well as on infrastructure (capital outlays) ........................................................................................................... 27 Figure 39. Average contribution of states to total GDP growth during FY13-FY15.............................................................. 28 Figure 40. …Average growth during FY13-FY15 ......................................................................................................................... 28 Figure 41. Planned consolidation is achieved through higher divestments and spectrum sales ............................................ 30 Figure 42. The deficit of the general government has been on a gradually declining path since 2012 ................................. 32 Figure 43. In FY17, states plan to stick to FRBM limits, excluding the impact of UDAY .................................................... 32 Figure 44. The combined debt stock has been stable ................................................................................................................... 33 Figure 45. Debt ratios vary significantly across states ................................................................................................................... 33 v Figure 46. Rural-urban consumption gaps have increased ........................................................................................................... 37 Figure 47. Farm activities comprise a smaller number of jobs .................................................................................................... 37 Figure 48. …and a declining share of income in rural India ........................................................................................................ 38 Figure 49. The rural-urban wage gap has narrowed ...................................................................................................................... 38 Figure 50. Real wages for rural men have decelerated or downright declined .......................................................................... 39 Figure 51. MGNREGA person days generated in FY16 grew, reflecting the stress in rural areas........................................ 39 Figure 52. Consumption growth has accelerated but remains below levels of mid-2014 ....................................................... 39 Figure 53. Pent-up consumption of durable goods in rural areas may be substantial ............................................................. 40 Figure 54. Financial development and economic growth are correlated in most countries ................................................... 42 Figure 55. Credit growth surged following the early 1990s reforms, supporting growth ....................................................... 42 Figure 56. India’s credit-to-GDP ratio is in line with most of its emerging market peers ...................................................... 42 Figure 57. Financial access of firms in India is in line with comparator countries................................................................... 43 Figure 58. India lags on the number of bank branches ................................................................................................................ 44 Figure 59. … and ATMs .................................................................................................................................................................... 44 Figure 60. Banks’ z-scores are above most peer countries ........................................................................................................... 44 Figure 61. Capital levels exceed regulatory requirements…......................................................................................................... 44 Figure 62. Nominal credit growth has decelerated ........................................................................................................................ 45 Figure 63. … but the trend is more stable when considering credit growth in real terms ...................................................... 45 Figure 64. The slowdown in credit growth was concentrated on PSBs ..................................................................................... 46 Figure 65. The deceleration in credit growth was concentrated in loan to industrial businesses… ...................................... 46 Figure 66. …as personal loan growth remained buoyant ............................................................................................................. 46 Figure 67. Non-performing assets have reached 6.2 percent, driven by higher NPAs in public sector banks.................... 47 Figure 68.Taking a long-term view, NPAs were higher in the early 2000s and came down................................................... 47 Figure 69. Restructured loans have climbed ................................................................................................................................... 47 Figure 70. Sub-standard and doubtful advances climbed ............................................................................................................. 47 Figure 71. Some of the largest PSBs have NPAs approaching 10 percent ................................................................................ 48 Figure 72. Stressed assets are concentrated in larger banks ......................................................................................................... 48 Figure 73. Until recently, net profits of PSBs had been declining .............................................................................................. 48 Figure 74. Lending rates declined less than the policy rate .......................................................................................................... 49 Figure 75. Since the rate cut cycle started, profit levels have been maintained and in some cases increased ...................... 49 Figure 76. Private banks have increased their presence ................................................................................................................ 51 Figure 77. Top 10 banks control about 60% of the market ......................................................................................................... 51 Figure 78. Interest rates show relatively low dispersion especially among the market-dominant PSBs ............................... 52 Figure 79. A longer-term view of lending rates against the T-bill rate ....................................................................................... 52 Figure 80. Interest rate spreads in India are high compared to other emerging economies ................................................... 53 Figure 81. Affordability and availability of financial services lag peers ...................................................................................... 53 Figure 82. The SLR has been progressively reduced but remains relatively high ..................................................................... 54 Figure 83. About 25 percent of bank assets are absorbed by investments in government securities ................................... 54 Figure 84. Credit to government and SOEs is above most peers ............................................................................................... 54 Figure 85. In 2012 and 2013, real interest rates on government securities had been negative ............................................... 54 Figure 86. Government mandates one factor constraining profitability despite wide interest spreads…............................ 55 Figure 87. …along with increasing NPAs in Public Sector Banks.............................................................................................. 55 Figure 88. Commercial banks dominate the financial sector ....................................................................................................... 56 Figure 89. The share of commercial banks is in line with peers but higher than in developed countries ............................ 56 Figure 90. India’s stock market capitalization is higher than China’s… .................................................................................... 57 Figure 91. But bond markets are much smaller than in China and advanced economies ....................................................... 57 Figure 92. Foreign participation in Indian bond markets is small compared to most other economies .............................. 58 Figure 93. Infrastructure lending surged… ..................................................................................................................................... 59 Figure 94. …with large contributions from the power and telecom sectors ............................................................................. 59 Figure 95. Infrastructure loans are not a large share in total advances, but a large share of those are distressed ............... 60 Figure 96. As NPAs mounted, infrastructure was a key contributor to the credit growth decline ....................................... 60 Figure 97. The exposure of SCBs to the power and roads sectors is large and correlated ..................................................... 61 Figure 98. Some banks require additional capital to meet Basel III requirements ................................................................... 62 Figure 99. PSBs have made lower provisions compared to private and foreign banks ........................................................... 62 Figure 100. The time to resolve insolvencies is long… ................................................................................................................ 66 vi Figure 101. …similar to the time to enforce a contract ................................................................................................................ 66 Figure 102. India could consider reducing its single-borrower limit in line with peers ........................................................... 68 Figure 103. Buoyant credit growth to the power sector declined as the sector was under stress… ..................................... 72 Figure 104. …largely because of persistently high technical losses ............................................................................................ 72 Figure 105. Breakdown of stress in the portfolio of national highways contracts (2014) ....................................................... 73 Figure 106. SBI concentrates a large share of roads and transport assets ................................................................................. 74 Figure 107. Roads and ports are the most stressed part of SBI’s infrastructure portfolio ...................................................... 75 Figure 108. SBI funded exposure to highways and other infrastructure sectors ...................................................................... 75 Figure 109. Sizing up programs under stress .................................................................................................................................. 75 TABLES Table 1: Reforms are moving to the states...................................................................................................................................... 19 Table 2: Key macroeconomic statistics and forecasts ................................................................................................................... 20 Table 3: FY16 Budget Execution ..................................................................................................................................................... 21 Table 4: Revenue performance ......................................................................................................................................................... 22 Table 5: Transfers were higher than budgeted as a percent of GDP ......................................................................................... 24 Table 6: All states report higher net transfers following the 14th FC.......................................................................................... 25 Table 7: Most states used additional transfers to spend more on health, education, and infrastructure .............................. 27 Table 8: Tax measures in Budget FY17 ........................................................................................................................................... 31 Table 9: Changes in health, education, and capital expenditures in FY17 ................................................................................. 32 Table 10: India’s instruments for fiscal transfers to states ........................................................................................................... 35 Table 11: A rural consumer class is emerging ................................................................................................................................ 40 Table 12: Banks have recovered about ¼ of claims through SARFAESI ................................................................................. 66 Table 13: The power and transport sectors pose significant risks .............................................................................................. 71 Table 14: Options for relieving stress in the portfolio of active contracts ................................................................................ 76 vii Executive Summary RECENT ECONOMIC DEVELOPMENTS excess capacity domestically and globally as well as AND OUTLOOK strained corporate and bank balance sheets dampened private capital expenditures, which account for ¾ of India’s economy expanded at a faster pace even as total investments. The eventual recovery of private several growth engines stalled: agriculture faced a second investments―critical for a sustained pickup in economic consecutive drought year, rural households were under growth―will be aided by a crowding-in from stress, private investments flat-lined, and exports infrastructure investment and ongoing reforms, such as plummeted. Despite these drags, the working the recently-passed Bankruptcy Code, which will engines―demand from urban households and public improve the business environment. investments―propelled the economy to a higher growth path. GDP growth accelerated to 7.6 percent y/y in Urban households drove the robust acceleration of FY16 from an average of 6.5 percent during FY13-FY15. consumption growth. Key drivers have been the Gross value-added (GVA) growth also accelerated, led expansion in the manufacturing and services sectors, by a pickup in industrial activities and a reliable which creates non-agricultural jobs, and greater contribution from services. purchasing power due to declining inflation and lower interest rates. Inflation abated primarily because of To remain on this growth path and sustain the 7.6 slower food price increases, but urban households in percent growth rate into FY17, India will need to restart particular benefited from decelerating housing inflation. dormant growth drivers, while ensuring the working Lower inflation and prudent fiscal policy have allowed engines do not run out of fuel. The dissipation of the the RBI to cut interest rates, and the resulting credit large boost from historically low oil prices in the past growth to the household sector went disproportionally year will make this a challenging task, but prospects of a to the more financially integrated urban dwellers. normal monsoon, which can reactivate agriculture and the rural economy, may help. A stabilization of global Notwithstanding greater diversification of income trade that lifts exports would provide further support. sources to non-farm activities, rural households remain largely dependent on agriculture and suffered in the past In FY16 domestic demand had to compensate for the two years from the drought-induced weakness in the drag in exports, which contracted for five consecutive agriculture sector. A rebound in agriculture on the quarters due to the slowdown in global trade. Imports expectation of a normal monsoon in 2016 will boost the also declined, but by less than exports, and net trade rural economy, which along with moderating inflationary subtracted from GDP growth. From the point of view expectations will support a broad-based acceleration in of the current account, export weakness was consumption. overshadowed by the large positive terms-of-trade shock delivered by plummeting oil prices. The trade and There are four key risks to the growth outlook. First, current account (CA) deficits narrowed, and in recent rains may not take place at the right places and times for quarters rising Foreign Direct Investment has more than optimal production; second even under a favorable fully financed the CA. Other capital account items were monsoon, the consumption impulse in the rural more volatile, and foreign exchange reserves were stable economy could be dampened by rural household at about 9 months of imports, a level that provides indebtedness following two years of hardship; third, adequate buffer against external volatility. given the waning oil dividend (and possibly rising oil prices) and high targets for asset sales, the government The oil bonanza most directly benefited the government, may find it challenging to achieve fiscal consolidation which for the first time in several years exceeded its while avoiding a negative impact of higher taxes or lower revenue collection targets and used the resources to capital expenditures on growth; fourth, export demand contain the fiscal deficit, transfer more resources to may continue to deteriorate given elevated uncertainties states, and spend more on infrastructure. The central about the global economy. Adequate levels of reserves government ramped up capital spending, its efforts and a favorable current account position mitigate amplified by state governments that had additional external risks, but domestic financial risks may rise as resources from larger fiscal devolution. In contrast, banks move to classify non-performing assets (NPAs) based on stricter criteria. 1 « INDIA DEVELOPMENT UPDATE JUNE 2016 The Indian Economy in Pictures Key growth engines stalled in FY16… … but economic growth accelerated, supported by resilient domestic drivers… Index, 2010=100, four-quarter moving average Change from the previous year, percent 12.5 130 125 10.0 120 7.5 115 5.0 110 2.5 105 100 0.0 95 -2.5 Private Investment Index Exports 90 Pvt Consumption Gov Consumption -5.0 Agriculture Gross Cap Formation Net Exports 85 Discrepancy GDP -7.5 …especially urban consumption, as reflected by higher Lower exports have followed global trade movements, and sales of passenger cars. better prospects for the latter should help growth Decomposition of nominal USD y/y export growth (line), darker bars Sales of passenger cars, change from the previous year, percent are percentage point difference of country growth minus global growth 20 60 Contrib growth India's Share (goods) 50 15 Contrib growth world goods exports 40 30 Growth in goods exports 10 20 5 10 0 0 -10 -20 -5 -30 -10 Normal monsoons could yield a second ‘terms-of-trade Meeting revenue targets will be challenging but crucial shock,’ boosting consumption of rural households that to achieve both fiscal consolidation and continued still predominantly rely on agriculture growth in public investments Principal source of household income, percent of households Structure of revenues and expenditures, percent of GDP 15.0 Other expenditure Non-Farm 13 Casual 11 3.5 3.9 10.0 0.6 1.0 Capex, direct & Regular/Other 15 1.5 grants 11 5.0 7.0 Net Borrowing Non-Farm Self 16 (Fiscal Deficit, 16 0.0 GOI) -2.6 -2.7 Spectrum sales & Farm Casual 21 divestments 27 -5.0 -10.4 Other Revenue Farm 34 -10.0 36 -15.0 Tax Revenue 0 10 20 30 40 FY16 FY17 2012 2005 INDIA DEVELOPMENT UPDATE JUNE 2016 » 2 FINANCING DOUBLE-DIGIT GROWTH Reversing the trend of rising NPAs is an immediate India’s ambitions to accelerate GDP growth to double- challenge, and addressing it will require a two-pronged digits will require a strong financial sector to allocate approach to corporate and banking balance sheets. savings towards the most productive investment Specifically, banks will require additional capital to opportunities. A well-functioning financial sector boosts ensure they are able to both meet Basel III requirements productivity growth, while guarding against volatility. and resume credit (and balance sheet) expansion to finance growth. The timely recognition of expected In many respects, India’s financial sector is performing losses is essential to assess the extent of fresh resources well, especially considering how it has evolved from an needed, and recapitalization must go hand-in-hand with essentially state-controlled system in less than three strengthening banks’ governance. Banks also need more decades. Financial access is comparable to peers; banks tools for restructuring or selling distressed assets. have capital levels in excess of current regulatory requirements; regulations have been strengthened, and In addition, as NPAs are concentrated in the corporate overall credit growth in real terms has been resilient. On sector―primarily metals, transport, road infrastructure, the other hand, concerns have arisen about growing non- and power―, addressing corporate balance sheet stress performing assets (NPAs) and declining credit growth, will be critical to preempt the emergence of new NPAs. particularly in public sector banks (PSBs), as well as the muted transmission of monetary policy to the real In the power sector, an important effort is the Ujwal economy, where lending rates have been sticky and DISCOM Assurance Yojna (UDAY) scheme, which interest rate spreads have increased. provides a financial bailout to state-owned electricity distribution companies (DISCOMs), and is in some ways Two sets of reforms are required to ensure that India’s akin to recapitalizing the banks holding DISCOM debt. financial sector can be a reliable pillar of future economic To avoid moral hazard, UDAY includes measures and growth. First, the government can accelerate the ongoing targets to reduce the losses of the DISCOMs that led to structural transformation of the sector into a more distress in the first place. Subject to the implementation market-oriented and competitive one. Second, it will of the measures and adherence to the targets, UDAY will need to address the NPA challenge proactively, not only reduce the prospects of future NPAs in the sector. through the recapitalization of PSBs, but also through reforms aimed at strengthening governance of both In the case of infrastructure, a more proactive approach commercial banks and corporate sectors that have may be considered to resolve current Public Private generated the largest share of NPAs. Partnership (PPP) projects under stress, especially in the roads sector. Moving toward a new PPP model for the A number of current transitions in the financial sector roads sector will prevent renewed accumulation of reflect its long-term transformation toward market- NPAs in the future, while ensuring the effective delivery oriented institutions, instruments, and practices. First, of much-needed infrastructure. the system has been transitioning from one that was almost entirely state-owned in the early 1990s toward Considering the strengths of the financial sector, RBI’s one with greater participation of private banks and strong capacity as a regulator, the earlier experience in generally more competition. Second, government the 2000s of bringing down even higher levels of NPAs, mandates aimed at ensuring stable financing for key and the improving fiscal position of the public sector as sectors of the economy (the government itself, through the ultimate backstop, the prospects of a crisis in the near the statutory liquidity ratio, and strategic economic term appear substantially diminished, even if NPAs rise sectors, through priority sector lending requirements) further from the reclassification of assets under stress. have been gradually relaxed as financial markets have developed to serve those sectors. Third, non-banking Nevertheless, accelerating structural, long-term reforms financial institutions and capital markets are nascent, but that are already underway and addressing the issue of expanding. Fourth, remuneration of savings instruments, rising stress in corporate and bank balance sheets are which have traditionally included gold and small savings both imperative and urgent for two reasons. First, certificates, increasingly reflect market rates, facilitating immediate action is needed to revert trends that, if left financial deepening, credit intermediation and monetary unchecked, may eventually lead to a crisis―a fact that is policy transmission. Accelerating these transitions by, well recognized by the Indian authorities. Second, if the for example, adopting more commercially oriented economy is to move permanently to a higher growth practices in PSBs, is likely to increase competition and path, the pace of reforms in the financial sector also the efficiency of the sector. needs to pick up to support productivity growth. 3 « INDIA DEVELOPMENT UPDATE JUNE 2016 Financing Double-Digit Growth in Pictures Credit and GDP growth usually go hand-in-hand Nominal credit growth declined partly owing to lower Left axis: Real GDP; Right axis: Real Credit Growth; change from inflation, as real credit growth has been stable the previous year; percent Year-on-year growth in bank credit, percent 10 25 20 9 18 8 GDP Growth 20 16 7 14 6 15 12 5 10 4 10 8 3 6 Credit 2 Growth 5 4 Sep-12 Dec-12 Sep-13 Dec-13 Sep-14 Dec-14 Sep-15 Dec-15 Jun-12 Jun-13 Jun-14 Jun-15 Mar-13 Mar-14 Mar-15 (RHS) 1 0 0 1963 1970 1977 1984 1991 1998 2005 2012 Nominal Deflated by GDP deflator The decline in credit growth and increases in NPAs have Nearly half of restructured standard advances are in three been concentrated in public sector banks infrastructure sectors Non-performing assets as a share of total assets Percent All SCBs 50 7 Public Sector Banks (76% of total) 6.20 45 6 Private & Foreign Banks (24% of total) 40 35 14.6 5 5.10 30 4 25 3 20 2.41 15 29.3 2 10 3.2 3.8 1 5 9.2 5.0 0 0 Sector's share in Share of sector in Sector's share in total advances Total NPAs - restructured aggregate level standard advances Power Transport Telecom Other Lending rates have declined less than deposit rates The spread between lending and deposit rates is higher in India than many of its peers Lines: Index (Sept ‘14=100), LHS; Bars: change, percentage points Spread between borrowing and deposit rates, percentage points CHN IND MEX 105 0.0 8 RUS ZAF -0.2 100 7 -0.26 -0.4 95 6 -0.6 Policy Rate -0.8 5 90 Deposit Rate -1.0 4 Lending rates (Public) 85 -1.15 -1.2 3 Lending rates (Private) -1.25 80 -1.4 Sep-14 Dec-14 Mar-15 Jun-15 Sep-15 Dec-15 Change 2 (Sept 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 '14-Dec '15) INDIA DEVELOPMENT UPDATE JUNE 2016 » 4 A. Recent Economic Developments and Outlook India’s economy expanded even as key growth engines stalled India’s economic Notwithstanding the slump in global trade, a second year of sub-par monsoons and weakness expansion in private investment, real Gross Domestic Product (GDP) grew by 7.6 percent year-on-year accelerated in the (y/y) in FY16.1,2 This represents an acceleration from the 7.2 percent recorded in FY15 and an face of strong average growth of 6.1 percent in FY13-FY14. In the fourth quarter of FY16, GDP grew by a headwinds robust 7.9 percent y/y, jumping 10.1 percent from the previous quarter at seasonally adjusted annualized rate (saar) (Figure 1 and Figure 2).3 Private consumption and government-driven investments – both partly fueled by the favorable terms-of-trade shock of sharply lower oil prices in the period – were the main growth drivers. However, significant uncertainties about the demand-side drivers of the economy remain, as a large portion of growth in every quarter of FY16 has been attributed to statistical discrepancies (light blue bars in Figure 1). 4 Both GVA and GDP In recent quarters, Gross Value-Added (GVA) has grown more slowly than GDP, although the grew faster, but the momentum has been similar: GVA growth also accelerated both in the fourth quarter (Figure 2 pace of acceleration and Figure 3) and for the year as a whole (FY16: 7.2 percent; FY15: 7.1 percent, y/y). While diverged services were as usual the largest contributor to GVA growth, the acceleration in industrial activity was noteworthy, accounting for nearly one-third of overall growth (one-quarter in FY15). Manufacturing picked up, while the performance of the electricity and mining sectors appeared to improve, possibly reflecting a more conducive policy environment in these sectors. 1 “FY” refers to the Indian fiscal year, the twelve-month period ending March 31st of the year in question. 2 Gross domestic product is valued at market (purchaser’s) prices, whereas gross value -added is measured at basic (producer) prices, excluding product taxes and subsidies. 3 Data are seasonally adjusted using the X13 procedure in JDemetra+. 4 The IMF (2016, p. SA10) notes “weaknesses in estimating taxes less subsidies on products in constant prices” as possible sources of discrepancies in GDP expenditure estimates compared to GVA production estimates, which are deemed a more reliable indicator of economic growth. 5 « INDIA DEVELOPMENT UPDATE JUNE 2016 Figure 1. Steady contributions from household Figure 2. The economy expanded in the fourth quarter but consumption in FY16 supported GDP growth GVA growth has been lower than GDP growth GDP adjusted for inflation; change from the previous year (line), percent; GDP and GVA adjusted for inflation and seasonal fluctuations; change and contributions to y/y growth (bars), percentage points from the previous quarter, annualized, percent 12.5 15.0 10.0 8.3 GDP GVA 7.7 7.8 7.9 7.5 7.5 7.6 7.2 12.5 6.7 6.4 6.6 6.7 7.5 5.8 5.0 5.4 4.6 10.0 5.0 2.5 7.5 0.0 5.0 -2.5 2.5 Pvt Consumption Gov Consumption -5.0 Gross Cap Formation Net Exports 0.0 Discrepancy GDP 7/1/2012 1/1/2013 4/1/2013 7/1/2013 1/1/2014 4/1/2014 7/1/2014 1/1/2015 4/1/2015 7/1/2015 1/1/2016 10/1/2012 10/1/2013 10/1/2014 10/1/2015 -7.5 Source: Haver, CSO, and World Bank staff calculations Source: CSO, and World Bank staff calculations Note: Dotted lines correspond to four-quarter moving averages Uneven India’s sources of growth have been few, with many segments of the economy facing challenges. performance of the External demand has been tepid and GDP growth has relied exclusively on domestic demand, economy’s main which itself has been running on just a few engines. Private investment stagnated and likely growth engines and contracted in the fourth quarter; the modest 3.9 percent y/y expansion in overall investments is statistical likely due to the surge in government capital expenditures. Similarly, the rural economy, over 40 discrepancies percent of which depends on agriculture, reeled under weather-induced stress. Consumption increase the from urban households (who rely more on non-agricultural sectors) had to be driving most of uncertainty about the robust expansion in household consumption. Figure 4 illustrates these performance gaps. underlying trends and drivers Contributing to uncertainty about underlying economic drivers, statistical discrepancies, as reported in the national accounts, explain an unusual share of growth in FY16 (2.4 pp of FY16 growth compared to 0.1 pp in FY15). Moreover, declining commodity prices have led to a sustained decline in wholesale prices (in contrast to consumer prices; see Box 1) and as a result, nominal GDP and GVA both decelerated (the latter sharply, from 10.5 to 7.0 percent). Figure 3. Industrial growth picked up in the second half of Figure 4. Three key divergences: domestic vs. external, the FY, led by manufacturing urban vs. rural, and public vs. private capex GVA adjusted for inflation; change from the previous year (line), percent; and contributions to y/y growth (bars), percentage points Index, 2010 = 100. 12.5 170.0 Private Investment Index Gov't Capex 10.0 Exports 8.1 7.1 7.4 7.4 7.2 7.3 6.9 7.4 Final Consumption + Fixed Investment 6.7 6.3 6.7 6.2 150.0 Agriculture 7.5 5.4 Services & Manufacturing 4.9 4.5 4.9 5.0 2.5 130.0 0.0 110.0 -2.5 Agriculture Manufacturing -5.0 Other Industry Skilled Svcs (non-Gov't) Other Svcs GVA 90.0 -7.5 Source: Haver, CSO, and World Bank staff calculations Source: CEIC, CSO, and World Bank staff calculations INDIA DEVELOPMENT UPDATE JUNE 2016 » 6 Domestic demand drives growth as exports continue to plummet Only domestic Domestic demand (consumption and capital formation) accounted for 110 percent of y/y demand contributed growth in real GDP excluding discrepancies in FY16, a higher share than in FY15 (97 percent) to GDP growth and well above the FY13-FY14 average of 66 percent. Final consumption expenditure grew by 6.6 percent y/y, accounting for 84 percent of overall growth excluding discrepancies (FY15: 67 percent). Fixed investments grew by 3.9 percent y/y compared to 4.9 percent in FY15. Both import and export volumes declined in FY16 (-5.2 and -2.8 percent y/y, respectively), although the export contraction moderated in the fourth quarter (-1.9 percent y/y). Net trade made a negative contribution to growth (excluding discrepancies) of -0.5 percentage points (FY15: +0.2 percentage points). The export engine Nominal exports of goods and services in USD terms contracted by 11.8 percent y/y in FY16 stalled further… (Figure 5). Lower merchandise exports (-15.8 percent y/y) account for most of this decline, and exports of petroleum products in particular explain nearly half of the decline in total exports, but non-oil, non-gold exports also declined substantially (-9.9 percent y/y). Services exports also underperformed (-3.9 percent y/y). BOP data for the first nine months of FY16 (when services exports contracted by 1.1 percent y/y) suggest that a deterioration of exports of traditional services, especially freight, which is linked to the weak performance of goods exports, was the main driver of poor performance (Figure 6).Of note, at least through December 2015, IT exports continued to grow. Figure 5. Lower oil prices were not the only drivers of the Figure 6. Travel and IT remain bright spots, but weak decline in exports as “core” exports also contracted merchandise trade spilled over to freight exports Decomposition of export growth in USD, y/y, percent Decomposition of services export growth in USD, y/y, percent 4.0 7.5 0.9 2.0 4.8 1.3 1.2 5.0 0.0 2.2 -1.6 -2.0 2.5 -5.7 2.2 -4.0 1.6 Services 0.0 -6.0 -2.1 -8.0 Non-oil, non-gold goods -4.5 -2.5 -1.1 -10.0 Gold & Jewelry -1.3 -5.0 -12.0 Minerals, incl. petroleum 9M FY15 9M FY16 -11.8 -14.0 Freight Travel Other traditional FY15 FY16 Financial IT-related Other modern Source: Haver and World Bank staff calculations Source: Haver and World Bank staff calculations Note: Service exports combine monthly and quarterly BOP data …largely due to A decomposition of export growth by changes in global demand and changes in India’s weak global competitiveness (as proxied by market shares in global trade) reveals that the most recent demand conditions contraction has been due to weak demand (note the dominance of the darker bars in Figure 7 and Figure 8). India’s market share in services trade has been growing, while the market share in goods has declined somewhat given the share of commodities in exports (39 percent, including jewelry). 7 « INDIA DEVELOPMENT UPDATE JUNE 2016 Figure 7. The recent decline in India’s exports has been Figure 8. … and in the case of services market share has mostly due to weakness in demand conditions… increased. Decomposition of nominal USD y/y export growth (line), light bars are Decomposition of nominal USD y/y export growth (line), light bars are percentage point difference of country growth minus global growth percentage point difference of country growth minus global growth 60 40 Contrib growth India's Share (goods) Contrib growth India's Share (svcs) 50 35 Contrib growth world goods exports 30 Contrib growth world svcs exports 40 25 30 Growth in goods exports Growth in service exports 20 20 15 10 10 5 0 0 -10 -5 -20 -10 -30 -15 Source: WTO and World Bank staff calculations Source: WTO and World Bank staff calculations Merchandise As shown in Figure 5, the sharp decline in merchandise exports was largely the result of a 45.3 exports declined on percent fall in the value of exports of petroleum products (19 percent of the total basket in lower prices for FY15), which mirrored the decline in oil prices in the period (44.4 percent). Jewelry exports petroleum products contracted 17 percent on lower shipments to the Middle East, but exports of gold bullion picked and subdued up and the “gold and jewelry” category (13 percent of the basket) was little changed. The demand for capital contraction in ‘core’ (non-oil-non-gold) exports was also large and broad-based (Figure 9). goods, especially Agricultural exports suffered owing to poor output and lower global prices, contracting by 17.9 from emerging percent y/y in FY16, worse than the 5.9 percent contraction recorded in the previous year, when economies monsoons were also deficient. Manufacturing exports (especially of high-tech products such as machines and vehicles) declined 8.2 percent. Demand from all destinations fell in FY16 (Figure 10) but the slowdown in growth in emerging markets is clearly reflected in the export data. In particular, shipments to the Middle East and Asia were the most depressed. Figure 9. Machinery and agricultural exports declined the Figure 10. Shipments to the Middle East and Asia posted most in FY16 the largest contractions Decomposition of non-oil, non-gold export growth by type, y/y, percent Decomposition of total export growth by destination, y/y, percent 5.0 5.0 1.5 1.0 2.7 2.4 0.0 -0.6 -1.4 0.0 -1.2 -1.1 -2.9 -3.2 -5.0 -3.6 -5.0 -3.0 Other Manufacturing (9%) -2.5 -10.0 China & HK Other Asia Chemicals, plastics, base metals (29%) -2.7 -2.5 Shoes, Textiles, Apparel (21%) Middle East Other large EMs -10.0 -1.5 -15.0 Other EM USA Hi Tech Manufacturing (25%) Non-oil Non- Gold (Core) EU Other DM -15.8 Agriculture (incl. processed) (16%) Exports, -9.9 -15.0 -20.0 FY15 FY16 FY15 FY16 Source: Haver and World Bank staff calculations Source: Haver and World Bank staff calculations Note: Figures in parenthesis correspond to the share in core exports. Note: “Other large EMs” include 6 large emerging economies INDIA DEVELOPMENT UPDATE JUNE 2016 » 8 Simultaneously, Imports of goods and services (in USD) declined by 12.5 percent y/y in FY16, driven by lower imports also merchandise imports (FY16: -14.9 percent y/y vs. FY15: -1.0 percent; Figure 11). Similar to declined rapidly exports, the decline in oil prices drove imports sharply lower. Oil imports (34 percent of merchandise imports during FY15) fell by 40.1 percent y/y in FY16. Core imports (excluding oil and gold), declined by a more modest 3.7 percent. The decline in core imports may be linked to the weakness in exports through lower imports of intermediate goods, as capital goods imports were nearly flat. Service import growth decelerated to 1.6 percent y/y in FY16 (FY15: +3.3 percent y/y). BOP data for the first nine months of FY16 suggest this was largely because of a contraction in traditional services import growth (Figure 12) that is also partially associated to export weakness as it includes imports of shipping and storage services. Figure 11. Lower oil imports explain almost all of the Figure 12. Growth in services imports also decelerated, decline in nominal imports, implying volumes increased but remained positive overall Decomposition of merchandise import growth by type, y/y, percent Decomposition of service import growth by type, y/y, percent Petroleum Agriculture (incl. processed) 10.0 10.0 Gold Capital Goods Other goods Services 8.0 1.0 0.5 6.0 5.0 3.8 2.0 4.0 2.1 1.2 0.2 0.0 1.0 -0.4 2.0 4.8 3.3 1.4 2.5 -4.9 0.0 -1.2 -5.0 -10.2 -2.9 -2.0 -4.0 -10.0 -6.0 -1.3 9M FY15 9M FY16 -15.0 -12.5 Freight Travel Other traditional FY15 FY16 Financial IT-related Other modern Source: CEIC and World Bank staff calculations Source: CEIC and World Bank staff calculations The fall in imports The current account deficit narrowed to 1.4 percent of GDP during the first three quarters of limited the impact FY16 from 1.7 percent in the same period the previous year. This was largely because of the of weak exports on significant terms-of-trade shock arising from a lower oil import bill, which contributed 1.6 growth as well as percentage points of GDP to the improvement in the current account deficit. This offset the the current account deterioration of 1.2 percentage points of GDP to the non-oil trade balance, mostly (1.0 pp) coming from the weakness in the non-oil merchandise trade balance. Remittances have been relatively steady in the past, but also declined sharply in Q3 FY16 (-31 percent y/y) as lower oil prices dampened flows from Gulf Countries.5 This translated into a decline of -0.2 percent of GDP (Q1-Q3 FY16), partly offset by a small increase in other income and transfers (+0.1 pp). Urban households drove consumption growth, as agriculture lagged Consumption Government consumption in the national accounts expanded by a modest 2.2 percent y/y in growth was resilient FY16 compared to 12.8 percent in FY15 as the general government deficit, excluding capital in the face of rural transactions, expanded only slightly to 7.2 percent of GDP in FY16 from 7.0 percent in FY15. stress and lower Meanwhile, capital expenditures increased (see Special Issue Note 1), compressing government government current consumption. In contrast, private consumption accelerated to 7.4 percent y/y in FY16 from 6.2 spending percent in FY15. This pickup is to some extent puzzling given that over two-thirds of Indian households are in rural areas, which suffered a second year of sub-par monsoons in FY16. Since most rural households depend directly and indirectly on agriculture for their income (see Special Issue Note 2), households from urban areas had to be the main drivers of consumption growth. 5 Remittances to India, the world’s largest remittance recipient, decreased by 2.1 percent in 2015 to USD 68.9 billion, the first decline since 2009. 9 « INDIA DEVELOPMENT UPDATE JUNE 2016 The agriculture Growth in agricultural output remained muted at 1.2 percent in FY16 from an average of 2.8 sector was stagnant percent during FY13-FY14. The main reason for the weak performance was deficient rain in because of 2015 for the second consecutive year, with south-west monsoons 14 percent below normal. challenging water Moreover, declining water tables in North India and deteriorating soil health in some areas and soil conditions because of overuse of subsidized urea fertilizer also contributed to weakness in the agricultural sector.6 Since 2014 was also a drought year and conditions in 2015 improved slightly, agricultural production growth accelerated compared to the 0.2 percent contraction observed in FY15.7 Food grain production during FY16 increased marginally by 0.5 percent to 253.2 million tons – almost entirely because of higher wheat production as the winter crop performed better owing to some unseasonal rains that improved soil moisture at the time of sowing. In comparison, India had produced 265 million tons of grains in FY13. In contrast, Industrial activities accounted for more than 32 percent of overall real GVA growth in FY16, manufacturing compared to 26 percent in FY15. Industrial growth accelerated to 7.4 percent in FY16 from 5.9 growth revived, and percent in FY15 (Figure 13). Within industry, manufacturing – particularly of basic and services remained a consumer goods, which accounted for more than 100 percent of the IIP growth in the period reliable growth (Figure 14) – gained strength, despite a slowdown in capital goods production in the last quarter, driver in line with the contraction in fixed investments. Manufacturing output expanded faster than the overall economy for five consecutive quarters, at an average of 8.8 percent y/y.8 Momentum in construction activity was volatile despite the increase in highway capital expenditures by the central government (Q4: +1.0 percent saar following +9.6 percent in Q3 and -8.2 percent in Q2). Services accounted for about two-thirds of overall growth but decelerated from 10.3 percent y/y registered in FY15 to 8.9 percent y/y in FY16 (3.5 percent q/q saar in Q4 vs. 10.1 percent in Q3). Modern services (financial, real estate and professional services) grew more rapidly compared to traditional services, in part due to the slowdown in trade as discussed earlier. Figure 13. Manufacturing led the acceleration of the Figure 14. The momentum in IP growth was broken in industrial sector November because of the floods in Chennai Decomposition of industrial growth, y/y, percent Decomposition of the growth in the IIP, y/y, percent 10.0 Mining Manufacturing Construction 8.0 Utilities Industry 9.0 6.0 7.0 4.0 2.0 5.0 0.0 3.0 -2.0 Basic Goods Capital Goods -4.0 1.0 Intermediate Goods -6.0 Consumer Goods IP -1.0 -8.0 -3.0 Source: CEIC and World Bank staff calculations Source: CEIC and World Bank staff calculations Growth in Growth in manufacturing and services is likely to have created jobs in those sectors, which along manufacturing and with declining inflation and lower interest rates underpinned consumption growth of urban 6 See for example Economic Survey 2015-16, Chapter 9: “underpricing urea, relative to other fertilizers, especially [potassium and phosphate], encourages overuse, which has resulted in significant environmental externalities, including depleted soil quality.” (p. 131). 7 In addition, the share of horticulture and vegetables has been increasing, which makes agriculture less exposed to monsoons. 8 There has been a divergence between the IIP and manufacturing value-added linked to a wider gap between producer and consumer prices that increased the profitability (and value-added) in manufacturing firms. Box 1 discusses further the divergence between producer and consumer prices. INDIA DEVELOPMENT UPDATE JUNE 2016 » 10 services linked to households, which was the main engine of India’s growth in FY16. Proxies for urban and rural higher urban consumption support this hypothesis. During FY16, passenger vehicle sales grew by 7.3 percent consumption y/y (Figure 15), passenger traffic at airports increased by 17.6 percent y/y, and production of consumer durables displayed very robust growth with an average of 11.1 percent y/y. In comparison, sale of two-wheelers grew by just 2.7 percent (Figure 16), and sales of tractors contracted by 12.1 percent y/y during Apr-Feb FY16. Figure 15. Until recently, sales of passenger vehicles were Figure 16. …while purchases of two-wheelers lagged accelerating… Sales of passenger vehicles, change from the previous year (3mma), percent Sales of two-wheelers, change from the previous year (3mma), percent 20 25.0 15 20.0 10 15.0 5 10.0 0 5.0 -5 0.0 -10 -5.0 Source: CEIC and World Bank staff calculations Source: CEIC and World Bank staff calculations Urban households Urban inflation, as measured by the consumer price index, decelerated faster than the all-India have benefited from index from 5.7 percent y/y during FY15 to 4.1 percent during FY16 (Figure 17). In contrast, higher purchasing rural inflation decelerated from 6.2 percent to 5.6 percent y/y during the same period. While power as inflation the slowdown in inflation everywhere was primarily because of more restrained growth in food declined prices, in urban areas, lower increases in clothing and housing prices also contributed (Figure 18). In addition, urban food inflation slowed to 4.6 percent y/y in FY16, 1.8 p.p. lower than last fiscal year. Rural food inflation decelerated by a smaller 1.1 p.p. (to 5.4 percent y/y) during the same period. Although the direct impact of lower oil prices on consumer prices has been limited by higher excise duties imposed by the government (fuel and light prices increased by 2.7 percent in FY16), the indirect effect on food inflation (and urban household purchasing power) cannot be discounted. Figure 17. Urban inflation declined more… Figure 18. …as housing and food inflation abated Consumer Price Index, change from the previous year, percent Decomposition of Urban CPI changes y/y, percent 14.0 11.0 Other Housing 12.0 Fuel & Light 9.0 Food, Beverages, Tobacco, etc. 10.0 Urban 7.0 8.0 5.0 6.0 3.0 4.0 Rural Urban 2.0 1.0 0.0 -1.0 Apr-12 Oct-12 Apr-13 Oct-13 Apr-14 Oct-14 Apr-15 Oct-15 Apr-16 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16 Jul-12 Jul-13 Jul-14 Jul-15 Source: CEIC and World Bank staff calculations Source: CEIC and World Bank staff calculations 11 « INDIA DEVELOPMENT UPDATE JUNE 2016 Lower inflation and Between April 2015 and April 2016 the Reserve Bank of India (RBI) reduced the policy rate by prudent fiscal policy 100 basis points from 7.5 percent in April 2015 to 6.5 percent in April 2016 (Figure 19). has allowed the RBI Although the pass-through from the policy rates to the lending rate has been incomplete, as to cut interest rates; discussed in Part C, credit growth to households accelerated – especially personal loans, benefits of higher including for consumer durables, vehicles, and housing (Figure 20). Lower interest rates and credit growth go credit growth to households likely benefited a larger share of urban households relative to rural disproportionally to ones. According to the World Bank’s FINDEX database of financial inclusion, in 2014 only 2.4 urban households percent of rural adults had a credit card, and only 1.7 percent used it in the past year. In contrast, for all-India, 4.2 percent of adults had a credit card and 3.4 percent used it in the past year. Figure 19. After spiking in late FY15, the real policy rate Figure 20. Growth in personal loans has been robust started to come down in FY16 Policy rate, nominal and adjusted for inflation Credit growth by type y/y, percent 9 20 8 18 7 16 14 6 12 5 10 4 8 3 Non-food credit Personal loans 6 2 4 1 2 0 0 -1 Source: CEIC, RBI, and World Bank staff calculations Source: CEIC, RBI, and World Bank staff calculations Higher public spending could not avoid a deceleration in investments as private capex struggled Making good use of In FY16, the government used higher revenues from excises on fuel and savings from a lower the oil bonanza, subsidy bill to implement an increase in public capital expenditure, especially on roads, railways, both the union and and irrigation. Capital spending by the central government grew by 20.9 percent y/y in FY16, state governments increasing to 1.8 percent of GDP from 1.6 percent of GDP in FY15. The largest contribution gave a large boost to to the increase in capital spending was from the Roads, Transport, and Highways Ministry, capital where capital spending increased by 83 percent during FY16, followed by the ministries of rural expenditures… and urban development and railways. At the beginning of FY15, more than 70 road construction projects were languishing, equivalent to 8,300 kilometers. By February 2016, however, according to the budget announcements, nearly 85 percent of these projects were back on track. Meanwhile, a sample of 20 states accounting for close to 90 percent of spending shows an increase in capital expenditure from 1.8 percent of GDP in FY15 to 2.1 percent in FY16, amplifying the center’s infrastructure push (see Special Note 1 for additional details). The rapid pace of growth moderated in the second half of the fiscal year, consistent with a front-loading of public investments, but despite the Q4 slowdown the execution of the capital budget of the center improved to 98.5 percent, compared to an average of 83.4 percent in FY13-FY15. …but fixed Notwithstanding the Government’s efforts, the share of the public sector in overall investments investment growth is relatively small at 26 percent (Figure 21) and weakness in the private sector drove growth in slowed, as public fixed capital formation to slow to 3.9 percent in FY16 from an average 4.4 percent in FY13- investments were FY15. This decline was also a function of the loss of momentum of public investments in the not enough to offset fourth quarter of FY16, which led to a contraction of overall investment (-1.9 percent y/y) for slower private the first time in seven quarters. investments… INDIA DEVELOPMENT UPDATE JUNE 2016 » 12 …caused by excess Despite the acceleration in overall economic activity, including the manufacturing sector, capacity evidence suggests that the Indian economy continues to operate below potential, and domestically and manufacturing firms still have idle installed capacity. According to quarterly surveys conducted globally, the trade by RBI, capacity utilization improved by 1.5 percent from the previous quarter, but still stood slowdown, and a at 72.5 percent in Q3 FY16 (Figure 22). Excess capacity domestically is partly a reflection of stressed banking excess capacity globally. The decline in growth in China and weakness of recovery in advanced sector economies has been greater than anticipated when manufacturing facilities were built in the recent past. Investments in the manufacturing sector are further constrained by weak external demand. Rising non-performing assets in the banking sector, which weakened monetary policy transmission and dampened credit growth (see Part C), have also hampered private investments. Figure 21. The private sector accounts for nearly ¾ of Figure 22. Capacity utilization has declined GFCF Shares of total GFCF by type (as of FY14) Capacity utilization rate, percent 80 78 12.3 76 35.6 74 72.5 13.8 72 70 68 66 Q1 2011-12 Q2 2011-12 Q3 2011-12 Q4 2011-12 Q1 2012-13 Q2 2012-13 Q3 2012-13 Q4 2012-13 Q1 2013-14 Q2 2103-14 Q3 2013-14 Q4 2013-14 Q1 2014-15 Q2 2014-15 Q3 2014-15 Q4 2014-15 Q1 2015-16 Q2 2015-16 Q3 2015-16 38.3 PSUs General Government Private Corporates Households Source: CSO Source: CEIC, RBI, and World Bank staff calculations Box 1: Divergence in input and output price indices Wholesale and consumer inflation diverged in FY16. The wholesale price index or WPI, which serves as a proxy for producer prices in India and is used in the calculation of the GDP and GVA deflators, was until April in a state of deflation for 17 consecutive months. In FY16, WPI registered an average deflation of 2.5 percent y/y. In contrast, the consumer price index (CPI), which measures retail inflation, has been volatile during FY16 after declining in FY15, and averaged 4.9 percent y/y for the year (Figure 23). The widening of the gap between the two indices from January 2015 is at first puzzling. Figure 23. CPI and WPI have diverged… Figure 24. … as lower oil prices did not reach consumers Year-on-year change, percent Year-on-year change, percent 10 CPI: Fuel and Light 8 20.0 WPI: Fuel and Power Crude oil, Brent 10.0 6 0.0 4 -10.0 2 -20.0 0 -30.0 -2 -40.0 CPI WPI -4 -50.0 -6 -60.0 Source: CEIC, RBI, and World Bank staff calculations Source: CEIC, RBI, and World Bank staff calculations 13 « INDIA DEVELOPMENT UPDATE JUNE 2016 A steep decline in commodity prices appears to have largely underpinned wholesale deflation. Most of the fall in wholesale prices was driven by the decline of prices of mineral oils (FY16: -18.0 percent). Mineral oils are mostly petroleum and petroleum products, the prices of which declined by 44.0 percent in the same period. However, lower wholesale prices were not passed through to consumers as the government hiked taxes and retailers increased margins. Figure 24 shows that pass-through from oil prices to the WPI (fuel and power) was less than 30 percent. Since the WPI includes excise taxes on petroleum, part of the difference can be attributed to the union raising excise taxes on fuel, while part is attributable to lower price declines of other components (the WPI for high speed diesel prices declined by 16 percent, while coal prices were flat, for example). However, at the same time that the fuel and power WPI declined by an average of 11.5 percent, the fuel and light CPI rose 5.4 percent y/y, and the transportation and communication CPI (which includes fuels) was flat during the period. The gap between the energy-related categories of CPI and WPI reflects the different composition of the categories, but also a combination of higher state-level taxes (VAT) on petroleum and products, as well as higher margins by retailers. In the case of energy, for example, lower wholesale inflation in petroleum and coal prices was not passed through to consumers (retail energy prices were up by 5.4 percent y/y), suggesting that some of the benefits from lower fuel prices was used for balance sheet repair in those corporates. Prospects for activating stalled engines suggest a favorable outlook with significant risks A boost from a The challenge for the Indian economy is to activate the stalled engines – agricultural growth and normal monsoon is rural demand; trade; and private investment – while ensuring the active engines do not run out expected to mitigate of fuel. Prospects in that regard are favorable. A rebound in agriculture on the expectation of a the waning impulse normal monsoon in 2016, along with lower interest rates, should offset the waning tailwinds of from oil prices, with lower oil prices and support broad-based consumption growth in FY17. Investment growth is private investments expected to remain moderate and still largely supported by the public sector. Finally, assuming taking over as the India’s share of global trade remains stable and global trade resumes modest growth, exports main growth driver can make a larger contribution to growth. in the longer term Overall, a modest acceleration in economic momentum (as measured by GVA) is projected in FY17 (Figure 25). This momentum is likely to be sustained in the longer term (FY18 and beyond) assuming private investments recover, supported by the crowding-in effect from the earlier public investment push, a better investment climate, and less leveraged corporate and financial balance sheets. Real GDP growth (market prices) is maintained at 7.6 percent in FY17, and accelerates modestly in FY18 and FY19 (Figure 26 and Table 2). Figure 25. Consensus forecasts expect a sharper Figure 26. Growth is expected to gradually accelerate to acceleration 7.8 percent in the medium-term Change from the previous year, percent Change from the previous year, percent 8.4 GDP MP GVA BP 8.2 9.0 8.0 7.7 7.8 7.6 7.6 7.8 7.7 8.0 7.2 7.6 6.6 7.6 7.0 7.7 7.8 7.2 7.4 5.6 7.1 7.4 6.0 6.3 7.2 5.0 5.4 7.0 4.0 6.8 3.0 6.6 6.4 2.0 FY17 FY18 1.0 Consensus: Low World Bank 0.0 Consensus: Median Consensus: High 2012-13 2013-14 2014-15 2015-16 2016-17 2017-18 2018-19 Source: World Bank staff calculations and projections Source: World Bank staff calculations and projections INDIA DEVELOPMENT UPDATE JUNE 2016 » 14 Potential GDP While data constraints make it difficult to calculate India’s potential GDP with precision, we growth stable; estimate potential GDP growth to converge to about 7.5 percent in the medium term (Figure acceleration would 27), assuming a meaningful and sustainable pickup in investments as in the baseline scenario. require additional The economy currently has a small negative output gap, which is likely to close by FY19; a investments and simultaneous increase in potential output (and supply capacity) will prevent imbalances from productivity growth emerging. Achieving a higher potential would require productivity enhancements, a larger pickup in investment, and an increase in women’s participation in the labor force as India’s labor force is short of its potential given the large gender gap in economic participation. 9 Figure 27. The output gap remains negative Lines: Year-on-Year growth of GDP and potential GDP, percent; Bars: output gap, percentage points 8.0 4.0 7.0 3.5 6.0 3.0 5.0 2.5 4.0 2.0 Output gap (RHS) 3.0 1.5 GDP growth 2.0 Growth in Potential GDP 1.0 1.0 0.5 0.0 0.0 -1.0 -0.5 -2.0 -1.0 2011-12 2012-13 2013-14 2014-15 2015-16 2016-17 2017-18 2018-19 Source: World Bank staff calculations and projections Risks to the outlook There are four key risks to the growth outlook: first, even if forecasts of a normal monsoon are significant materialize, rains may not take place at the right places and times for optimal production; second, the consumption impulse of a good monsoon will depend on the extent of rural deleveraging that may be required as rural households may have become indebted; third, the waning oil dividend will make it challenging for the government to achieve fiscal consolidation and simultaneously avoid a negative impact on growth; and fourth, export demand may not improve given elevated uncertainties about the global economy. Risks from external financial volatility are lower, mitigated by high levels of reserves and a favorable current account position, but domestic risks may rise as banks move to classify NPAs based on stricter criteria. Agriculture and private consumption lift growth in the near term, followed by private investments A normalization in Based on the forecast of an above normal monsoon in 2016 owing to a fading El Niño10 as well agricultural activity as a strong base effect, agricultural growth is expected to accelerate to 3.5 percent during FY17 can be a second and return to its long-run average growth of 2.7 percent in the following years (Figure 28). The ‘terms-of-trade direct effect of the normalization of agriculture would be between 0.3 and 0.5 percentage points; shock,’ making up the indirect effect on other sectors from non-agricultural consumption, assuming conservatively for the dissipating a 0.75 pp pickup in non-agricultural consumption growth, could add a similar amount, largely impulse of lower oil offsetting less help from oil prices. prices 9 Potential GDP estimates are based on a standard production function with labor and capital shares held at two-thirds and one third, respectively. Capital stock estimates are based on the old (base-year: 2004-05) national accounts series – and have been converted to the new series using the relationship between the flow of capital formation in the new and old series. Labor force estimates are from the UN population projections. 10 The Indian Meteorological Department (IMD) forecasts rains in the 2016 monsoon to be 106 percent of the long-period average with less than 5 percent probability of below-normal rains; private sector forecaster Skymet makes a similar forecast. 15 « INDIA DEVELOPMENT UPDATE JUNE 2016 A revival of agricultural activity, along with stable industrial and services growth, will drive a modest acceleration in overall GVA growth to 7.4 percent in FY17 from 7.2 percent in FY16, before climbing further as manufacturing gains further from an eventual revival of the private investment cycle and in response to the government’s efforts to improve the business climate. Improvements in Private consumption is expected to accelerate from 7.4 percent in FY16 to 7.8 percent in FY17 agriculture and rural supported by (i) a revival in consumption demand from rural households following the pickup in demand will lead to agricultural activity (0.3 pp); (ii) continued credit growth to households (0.3 pp); and (iii) moderating a broad-based inflationary expectations of households – the three-month ahead inflation expectation remains high acceleration in at 8.1 percent in March 2016, but is down from 10.5 percent in September 2015 (according to the consumption in the RBI household inflation expectations survey) and should decline further toward the RBI’s inflation near term target of 5 percent by January 2017 (0.1pp). The disappearing tailwind oil prices and uncertainty about the timing and extent of civil service pay revisions partly offset these gains (-0.3 pp). According to budget estimates of FY17, public consumption is expected to be maintained, and consolidation is expected to emanate from the one-time sale of public assets, neutralizing most of the negative fiscal impulse of consolidation. Figure 28. GVA growth will be lifted by a recovery in Figure 29. …while on the expenditure side the growth agriculture… drivers gradually shift from consumption to investment Change from the previous year, percent Change from the previous year, percent 12.0 10.0 9.0 10.3 9.0 8.3 10.0 9.3 9.3 7.8 8.9 8.8 8.0 7.4 7.5 8.1 7.2 7.8 6.8 7.4 7.3 7.0 6.2 6.5 8.0 7.0 5.9 6.0 5.3 6.0 4.9 4.9 4.7 5.0 5.0 4.2 3.6 3.4 4.0 3.4 4.0 2.8 2.7 2.7 3.0 2.0 1.5 1.2 2.0 1.0 0.0 0.0 -0.2 2012-13 2013-14 2014-15 2015-16 2016-17 2017-18 2018-19 -2.0 Agriculture Industry Services 2012-13 2013-14 2014-15 2015-16 2016-17 2017-18 2018-19 Gross Fixed Investment Private Consumption Source: CEIC, World Bank staff calculations and projections Source: CEIC, World Bank staff calculations and projections In the medium- The performance of investments in FY16 was supported by a concerted increase in public capital term, the growth expenditure, which is budgeted to be sustained at 2.7 percent of GDP in FY17.11 Private momentum is investments are expected to start picking up in late FY17 as balance sheets are repaired, but are expected to be only expected to grow meaningfully over the medium-term (FY18-FY19). Private investment sustained by growth is expected to eventually respond to steps taken to address the problem of leveraged crowding-in private corporate and financial balance sheets; and as the crowding-in effect of the recent push in investments infrastructure spending materializes. According to Bahal et al. (2015), private investments in India respond to public capital expenditures in four-to-twelve quarters. Moreover, the Government’s initiatives such as “Make in India” and “Startup India” should also help accelerate investments. As a result, overall investments are expected to accelerate from 3.9 percent in FY16 to 9.0 percent in FY19 (Figure 29). 11 This measure of capital expenditure includes the center’s direct capital expenditure and current transfers to states for capital spending. INDIA DEVELOPMENT UPDATE JUNE 2016 » 16 Exports are expected to recover, supporting growth and keeping the current account in check Export growth is Assuming India’s share of global exports is maintained, a recovery in global trade is expected to likely to recover in take growth of exports of goods and services modestly higher to 3.5 percent in FY17, consistent conjunction with with the latest World Bank projections of the growth in global trade volumes of 3.1 percent global trade (Figure 30).12 In later years, export growth is projected to accelerate in line with global trade and modest gains in India’s market share (as agricultural exports may resume), but remain well below the levels registered during the boom years before 2012. The sustainability of the export recovery is contingent on an improved external environment, as well as the relief in supply-side bottlenecks that would allow India to increase its export market share. Although exports of services remain a bright spot and are expected to resume double-digit growth in the medium term, merchandise exports, similar to manufacturing output, are likely to expand meaningfully only once investments and productivity-enhancing reforms to labor, land, and capital markets are realized. Figure 30. Trade is expected to recover modestly along Figure 31. Higher capital goods imports and oil prices with the global economy lead to a widening of the current account from FY18 Trade growth, y/y, percent Current account balance, percent of GDP 0.0 10.0 8.0 -1.0 6.0 -1.3 -1.4 -1.4 -1.5 -1.6 4.0 -2.0 -1.7 2.0 -3.0 0.0 2012-13 2013-14 2014-15 2015-16 2016-17 2017-18 2018-19 -2.0 -4.0 -4.0 -6.0 -5.0 -4.8 -8.0 -6.0 Exports, GNFS Imports, GNFS 2012-13 2013-14 2014-15 2015-16 2016-17 2017-18 2018-19 -10.0 Source: CEIC, World Bank staff calculations and projections Source: CEIC, World Bank staff calculations and projections Imports will pick up Import growth is expected to accelerate gradually from a contraction of 2.8 percent in FY16 to in response to a +2.0 percent in FY17 and further to +6.3 percent by FY19. The initial recovery in imports is recovery in expected on account of higher private consumption and a modest recovery in exports, which will investment-related lead to demand for imported consumption and intermediate goods, respectively. In outer years, demand, leading to a imports are likely to grow owing to higher demand for capital goods (which currently account for modest widening of approximately 13 percent of total merchandise imports) from revived investment as well as higher the current account demand for intermediate inputs in response to improved manufacturing activity and exports. Accordingly, the current account deficit is expected to widen gradually from 1.4 percent of GDP in FY16 to 1.6 percent of GDP by FY19 (Figure 31). Foreign direct investment (FDI), which reached record highs in FY16, is likely to remain robust. As the government continues to open additional sectors of the economy, FDI is expected to be sufficient to finance the current account deficit in the projection period. Risks to the outlook are significant, but buffers mitigate against external shocks Downside risks to Risks of deceleration of the active engines (public investment and urban consumption) are the outlook are substantial. The public sector will need to manage current expenditures and revenue collection 12 Global trade forecasts from Global Economic Prospects June 2016. 17 « INDIA DEVELOPMENT UPDATE JUNE 2016 substantial, but are carefully to deliver on its further increase in capital expenditures and deficit target. Higher oil balanced in the prices are a significant risk. Consumers reaped some of the benefits of lower oil prices, and short run by the those benefits will at best dissipate; a marked increase in oil prices would lead to a negative possible upside impulse assuming the government retains current excise levels. Increases in inflation may halt from a favorable the RBI’s rate cut cycle. Without the benefit of the growth in oil-related excises (and possible monsoon and reductions if oil prices increase) the government will find it challenging to meet its deficit target improvements in without further cuts in expenditure – and usually capital expenditure is the first to be cut. the global environment Risks of the stalled engines not restarting are also large. While the current forecasts suggest a normal monsoon, the India Meteorological Department (IMD) forecasts have been higher than the realized rainfall in recent years, and cannot be precise whether rain will be distributed temporally and geographically to ensure growth in agricultural production. In addition, soil and water table problems are likely to persist. The construction sector, which provides significant employment in rural areas, has not performed well and will depend on higher investment growth to recover meaningfully. Finally, farmers may have become indebted to moneylenders to cope with the two years of weak monsoons, so even if the weather forecasts materialize, the consumption impulse may be muted as rural households deleverage. Downside risks to the global economy – and export growth – are also substantial given geopolitical risks, monetary policy normalization in the United States, and slowdown in large emerging economies linked to India such as China and those in the Middle East. The most As discussed in detail in Part C, large corporates are overleveraged and NPAs are elevated, thus significant near- restricting financing for private investment. Measures to address the debt overhang and public and medium-term sector bank’s recapitalization, including ones underway such as the “Indradhanush” program risks stem from the for the financial sector, will be required to achieve the pickup in investments assumed in the banking sector and baseline scenario. Private investment growth continues to face several impediments in the form its ability to finance of excess global capacity, regulatory and policy challenges, in addition to corporate debt the required overhang. A negative contribution from private investment cannot be ruled out if corporate and investment pickup financial sector balance sheet stresses are not alleviated. Longer term, in the absence of investments and resulting expansion of production capacity, not only may faster growth not materialize, but imbalances could build up in the medium term. In the long term, the A sustained and rapid expansion of India’s physical, human, and “institutional” capital is crucial greatest risk to for realizing (and possibly exceeding) the baseline growth projections. Steadfast policy India’s economic implementation is critical to accelerate investments in infrastructure and in people – through expansion is the improved service delivery in health, sanitation, and education – as well as in building lack of sustained “institutional capital,” namely the policies and institutions that enable private investments – e.g. policy reforms to GST, land acquisition, and insolvency. The credibility of public finances is also an important build up the part of institutional capital, and in this regard, the government’s reaffirming its fiscal deficit nation’s assets target of 3.5 percent for FY17 was a positive development that now needs to be implemented. Reform As noted in the previous Development Update and Special Note 1, the majority of public implementation expenditures are now undertaken at the state rather than the union level. States have also (and related risks) increasingly been taking the lead on structural reforms, with some such as Gujarat, Karnataka, has increasingly Rajasthan, Maharashtra and recently Madhya Pradesh moving faster than the center in reforming moved to the state labor or land laws. This model of “cooperative and competitive federalism” has served China level relatively well as many reforms can be more efficiently implemented at the state level, and states can learn from each other’s experiences. The government has supported these efforts, for example by creating a scorecard on reforms to improve the ease of doing business in the states (Table 1). It will be important that regulations on paper translate into a better business environment on the ground and that lagging states receive adequate support to ensure that this model accelerates growth without exacerbating inequalities among states. INDIA DEVELOPMENT UPDATE JUNE 2016 » 18 Table 1: Reforms are moving to the states Implementation of 98-point action plan to improve the business environment by state, score Source: DIPP, World Bank, KPMG, CII, Federation of Indian Chambers of Commerce & Industry Adequate buffers While volatile portfolio investments witnessed a pullout from the country (Q1-Q3 FY16: -0.2 minimize risks from percent of GDP; Q1-Q3 FY15: +1.9), FDI benefitted from new investments in computer external shocks hardware and software, services, and trading businesses. FDI flows grew by 24.8 percent y/y and amounted to 1.8 percent of GDP in Q1-Q3 FY16 (Q1-Q3 FY15: 1.5 percent). Supported by higher FDI inflows and a narrowing of the current account deficit (1.4 percent of GDP in Q1-Q3 FY16), foreign exchange reserves increased to USD 350.4 bn or 8.9 months of imports as of December 2015 (Figure 32).13 Figure 32. Foreign exchange reserves have been stable Figure 33. The REER has appreciated over the past two fiscal years Foreign exchange reserves, USD billion Real Effective Exchange Rate Index (2010=100) 400 140 356 350 350 360 342 350 316 314 321 130 294 304 300 120 250 110 100 200 90 150 80 100 70 50 60 0 01-2013 03-2013 05-2013 07-2013 09-2013 11-2013 01-2014 03-2014 05-2014 07-2014 09-2014 11-2014 01-2015 03-2015 05-2015 07-2015 09-2015 11-2015 01-2016 03-2016 Brazil China India Indonesia Russia Turkey Source: CEIC, RBI, and World Bank staff calculations Source: Bank for International Settlements 13 The stability of reserves in the past year despite the narrowing current account and higher FDI flows is partly due to a reversal of portfolio flows, outflows on external commercial borrowings by Indian corporates as they de-leveraged, a moderation in Non-Resident India (NRI) deposits given liabilities coming due, and a decline in short-term trade credit owing to lower imports. 19 « INDIA DEVELOPMENT UPDATE JUNE 2016 The flexible The period of March-December of 2015 was characterized by relative dollar strength as the exchange rate acts Federal Reserve raised interest rates in December, the exchange rate depreciated by a relatively as an additional modest 6.7 percent vis-à-vis the US dollar (the NEER—Nominal Effective Exchange rate— buffer depreciated by 3.4 percent). The real effective exchange rate (REER) appreciated by 0.4 percent in the same period, and depreciated by 1.4 percent in FY16 (Figure 33) on nominal depreciation and a stabilization of the inflation rate since December. This followed a 12 percent appreciation the previous fiscal year that was in contrast with depreciation in many other large emerging economies (except China), and bears watching as further appreciation may affect the competitiveness of the economy. Table 2: Key macroeconomic statistics and forecasts BASELINE 2012-13 2013-14 2014-15 2015-16 2016-17 2017-18 2018-19 GDP, market prices y/y percent 5.6 6.6 7.2 7.6 7.6 7.7 7.8 Private Consumption y/y percent 5.3 6.8 6.2 7.4 7.8 7.5 7.2 Government Consumption y/y percent 0.5 0.4 12.8 2.2 7.2 7.1 6.9 Gross Fixed Investment y/y percent 4.9 3.4 4.9 3.9 6.5 8.3 9.0 Exports, GNFS y/y percent 6.7 7.8 1.7 -5.2 3.5 5.0 6.9 Imports, GNFS y/y percent 6.0 -8.2 0.8 -2.8 2.0 4.8 6.3 GVA, basic prices y/y percent 5.4 6.3 7.1 7.2 7.4 7.7 7.8 Agriculture y/y percent 1.5 4.2 -0.2 1.2 3.4 2.7 2.7 Industry y/y percent 3.6 5.0 5.9 7.4 7.0 7.3 7.5 Services y/y percent 8.1 7.8 10.3 8.9 8.8 9.3 9.3 percent of Current account balance GDP -4.8 -1.7 -1.3 -1.4 -1.4 -1.5 -1.6 percent of Total Revenues GDP 19.7 19.5 19.6 21.7 21.9 21.7 21.6 percent of Total Expenditures GDP 26.6 26.2 26.6 28.5 28.3 27.6 27.1 percent of Fiscal Balance GDP -6.9 -6.7 -7.0 -6.8 -6.4 -5.8 -5.5 percent of General government debt GDP 66.6 64.5 65.2 65.8 63.8 61.8 60.5 Source: CEIC, Haver, Ministry of Finance, RBI, World Bank calculations and projections INDIA DEVELOPMENT UPDATE JUNE 2016 » 20 B. Special Issue Notes 1. India Fiscal Policy Update Execution of the Union Budget in FY16 was largely as planned Fiscal consolidation The government achieved its FY16 deficit target of 3.9 percent of GDP. Notwithstanding a targets were shortfall in capital receipts because targets for privatization receipts were missed, total receipts achieved, as higher (in nominal terms) were slightly higher than budgeted for the first time since 2011-12. Because revenue receipts and revised estimates of nominal GDP for FY16 came in 3.8 percent lower than the budget forecast, some expenditure revenues were 0.6 percentage points of GDP higher than budgeted. On the expenditure side, restraint offset revenue expenditures exceeded budgeted amounts while capital expenditures were marginally shortfalls in lower in nominal terms even as they expanded as a share of GDP. The excess revenue privatization expenditures were due to larger than budgeted expenditures on plan revenue expenditures receipts (possibly grants to states – see below) and on grants for creation of capital assets by sub-national entities. The contraction in capital expenditures was marginal – 1.5 percent of the budgeted amount – compared to an average of 12.8 percent slippage experienced between 2011-12 and 2014-15. Table 3 summarizes the execution of the union’s FY16 budget. Table 3: FY16 Budget Execution Percent of GDP 2015–16 (BE) 2015–16 (RE) Difference % contribution Total receipts 12.6 13.2 0.56 Revenue receipts 8.1 8.9 0.80 142.2 Capital receipts 4.5 4.3 -0.24 -42.2 Total expenditure 12.6 13.2 0.56 Revenue expenditure 10.9 11.4 0.52 92.7 Capital expenditure 1.7 1.8 0.04 7.3 Revenue expenditure (excl. grants for capital assets) 10.1 10.4 0.33 59.0 Capital expenditure (incl. grants for capital assets) 2.5 2.7 0.23 41.0 Source: Ministry of Finance and World Bank staff calculations 21 « INDIA DEVELOPMENT UPDATE JUNE 2016 Although tax In percentage terms, revenue receipts in 2015-16 were 5.7 percent higher than budgeted, in revenues increased, contrast to the earlier trend of an average slippage by 5.6 percent over budgeted values between it was a jump in 2011-12 and 2014-15 (Table 4). Non-tax revenues (one-fifth of total revenue receipts) exceeded non-tax revenues budgeted amounts by 16.6 percent, largely owing to higher dividends and proceeds from that led revenue spectrum auctions, and were the main drivers of revenue growth (Figure 34). Direct tax revenue receipts to exceed fell short of the target, but a 23 percent increase in excise taxes (vs. FY16 budget), largely because nominal budgeted of higher taxes on petroleum products, made up the difference, and tax revenues were 3.0 amounts for the first time since 2011-12 percent above budget targets. Table 4: Revenue performance Deviation between budget and actual (nominal values), percent 2011–12 2012–13 2013–14 2014–15 2015–16 Revenue receipts -4.9 -6.0 -3.9 -7.4 5.7 Tax revenue (net to center) -5.2 -3.8 -7.7 -7.5 3.0 Non-tax revenue -3.0 -16.6 15.5 -6.9 16.6 Source: Ministry of Finance and World Bank staff calculations Figure 34. Revenue receipts exceeded budget targets despite under-performance of direct taxes Contribution of tax and non-tax revenues to the change between budget and actual revenue receipts, percentage points 8 Non-Tax Revenue 6 Tax Revenue (net to centre) Revenue Receipts 3.2 4 2.5 2 2.4 0 -4.4 -3.1 -6.5 -6.2 -2 -4 -6 -0.5 -2.9 -8 -1.2 -10 2011-12 2012-13 2013-14 2014-15 2015-16 Source: Ministry of Finance and World Bank staff calculations What has been the impact of the 14th Finance Commission on states? The 14th Finance Specifically, the 14th FC recommended an increase of the states’ share in the central divisible Commission (FC) pool of tax revenues from 32 percent to 42 percent. Moreover, the share of transfers earmarked recommended a for specific purposes (“tied” transfers, largely linked to centrally sponsored schemes – CSS) was significant increase to be reduced to provide greater autonomy for states to make decisions about spending. 14 in the discretionary spending of states There were at least  To what extent would the forecast tax revenues – and by implication higher three sources of transfers out of those revenues – materialize? Greater dependence of state uncertainty in the budgets on tax devolution relative to grants-in-aid introduced additional revenue implementation of uncertainty for the states. While grants-in-aid are specific amounts that had been 14 See the October 2015 India Development Update for details on the recommendations of the 14th Finance Commission. INDIA DEVELOPMENT UPDATE JUNE 2016 » 22 the 14th FC (largely) recommended by the FC, devolution receipts are dependent on the center’s recommendations ability to collect taxes, with shortfalls passed down to states.  How would individual states fare, considering some states’ shares of the divisible pool had been reduced? The 14th FC also recommended a change in the formula for distribution of devolved resources to compensate states with higher rates of in-migration and other demographic changes, forest cover, and poor fiscal capacity (ability to raise its own-revenues). Some states saw their shares increase (e.g. Chhattisgarh, Madhya Pradesh), while others (e.g. Bihar, Rajasthan) saw their shares reduced. There was concern in those states that the smaller “share of the pie” could offset the larger pie size.  What would happen to the grants tied to the Centrally Sponsored Schemes (CSS)? As part of implementing the 14th FC, the Union Government reviewed its contributions to CSS. Lower contributions from the center in the form of tied grants (relative to the 13th FC period) would require states to increase their own share, thus reducing fiscal space. The rationalization of the CSS was completed in November 2015, and therefore at the time of state budget preparation there was uncertainty about the amount the states would have to cover out of other sources. Data constraints This section aims to provide a preliminary assessment of how these uncertainties were resolved prevent answering in the fiscal year by analyzing the budgets of 20 states that account for over 90 percent of India’s these questions with GDP. However, there are important data constraints that prevent drawing definitive precision, but some conclusions from this analysis. Data from the union and the states, especially regarding grants- directions can be in-aid, are not easy to reconcile and interpret, even more so as the institutions, procedures, and inferred. classification systems for fiscal transfers between the center and the states have been in flux for the past few years. As a result, lags may exist between states in the adoption of recording and classification of financial flows. While recording tax devolution is straightforward, there may be variations between the center and the states and among states in classifying and recording grants. According to the Union budget, aggregate transfers to States were as promised – even in nominal terms Devolution to states Based on revised estimates published with the FY17 Union budget, the center transferred an proceeded largely as additional 1 percent of GDP to states in devolved taxes (as budgeted), although this represented budgeted a smaller nominal amount than budgeted given the realization of a lower nominal GDP base. While nominal tax revenues largely met budget projections, this was due to increases in indirect taxes accrued from cesses that are not part of the divisible pool of resources for devolution. On the other hand, Grants were reduced by 0.3 percent of GDP compared to 0.5 percent of GDP in the budget. the reduction in Tied grants declined by 0.5 percent of GDP (vs. -0.6% expected) while untied grants increased grants was smaller by 0.2 percent of GDP (vs. +0.1% expected). One observer (HSBC) claimed that “several state than expected ministries accustomed to ‘tied’ funding suddenly realized that money was no longer coming in automatically. Halfway through the year they successfully lobbied for additional funds, resulting in higher overall transfers to states.” The higher amount of grants (vs. budget) offset lower (nominal) tax devolution. Overall, transfers to the states increased by 0.7 percent of GDP in FY16 compared to the budget estimate of a net increase of 0.5 percent described earlier. Table 5 summarizes the changes. 23 « INDIA DEVELOPMENT UPDATE JUNE 2016 Table 5: Transfers were higher than budgeted as a percent of GDP Percent of GDP 2014-2015 2015-16 2015-16 2016-17 Actuals BE RE BE Total transfers to states 5.5 6.0 6.2 6.2 Tied Grants to states 2.2 1.6 1.7 1.7 Untied Grants to states 0.6 0.7 0.8 0.7 Devolution of States' share in taxes 2.7 3.7 3.7 3.8 Source: Ministry of Finance and World Bank staff calculations Notes: Untied grants include non-plan grants from the Ministry of Finance (MoF); tied grants are calculated as a residual. According to states’ budgets, transfers increased everywhere but data discrepancies prevent a firm conclusion An analysis of the To estimate the extent to which individual states gained or lost from the implementation of the gains and losses 14th FC recommendations, we analyzed the published budgets from 20 states and performed a following the simple exercise as follows:15, 16 implementation of the 14th FC 1. Overall devolution effect (ODE). First, we isolated the impact of the overall increase recommendations in resources that came from the higher share of the divisible tax pool going to states must take into (increase from 32 to 42 percent). account changes in 2. Share effect (SE). Second, we isolated the impact of the change in the relative shares tax devolution as of individual states in overall devolution (the size of the piece of the devolution pie that well as the changes each state gets) resulting from the new state shares recommended by the 14th FC. in grants 3. Grant effect (GE). One potential source of losses relative to the 13th FC period is that states may need to raise their contributions to CSS. If higher devolution compensates for lower contributions from the center to CSS, the states will not in fact have additional resources. A thorough analysis of this question would require considering each scheme separately and analyzing the contributions of the state and center to it.17 We take a simplified approach and assume that central grants would have remained constant as a share of Gross State Domestic Product (GSDP) in a counterfactual “no 14th FC” scenario. This implies that expenditures previously covered by grants from the center under the 13th FC would be growing at the same rate as GSDP. 4. The sum of ODE, SE, and GE gives the net impact for each state. Wide variation in An analysis of states for which data is available suggests all states gained following the gains among states implementation of the 14th FC recommendations, but the extent of gains varied significantly.18 Table 6 and Figure 35 show the initial result of this analysis for the 20 states for which data was available from budget documents, accounting for over 92 percent of the tax devolution. A few observations may be made:  The grant effect is positive for most states and positive overall. This cannot be reconciled with figures from the Union budget that suggest grants have declined by 0.3 percentage points of GSDP. Since these 20 states include over 92 percent of tax devolution, additional analysis is required to understand the discrepancy between the data sources, and conclusions about whether states received higher net transfers are highly tentative.  The net effect appears positive for all states, varying from 0.5 percent of GSDP in Punjab to 5.3 percent of GSDP in Mizoram. 15 It is important to keep in mind the different data sources used in this section. Each sub-section of the analysis is internally consistent, but there are some important discrepancies between grants as reported in the Union budget and grants-in-aid as reported in state-level budgets. 16 See Annex II for the formulas used to calculate each effect. 17 See, for example, Chakraborty (2016). 18 The accounts for the state of Andhra Pradesh were separated from Telangana in June, 2014. Therefore, changes between FY15 and FY16 may not be meaningful. Devolution estimates adjust the 10-month figures by 20%. INDIA DEVELOPMENT UPDATE JUNE 2016 » 24  There was wide variation of gains as reported by the states. Maharashtra, Gujarat, and Tamil Nadu had small gains (0.6, 0.6, and 0.7 percent of GSDP respectively) while gains were 3.1 percent of GDP in Uttar Pradesh, 2.9 percent in Odisha, and 2.1 percent in West Bengal. It is noteworthy that Bihar, where the SE and GE effects are negative, still comes up ahead (1.3% of GSDP) because of ODE. In Odisha, the significant increase in grants-in-aid leads the state to be a large gainer.  Special category states, and those with large forest cover gained on a net basis. Himachal Pradesh and Mizoram posted the largest gains. Chhattisgarh, which had the second largest increase in share, and Jharkhand, another major gainer in terms of relative share (both owing to the inclusion of forest cover as a criteria), also appear as large beneficiaries on a net basis. In other words, for those states, a higher share of the larger devolution pie also translated to high net gains as a share of GSDP. Table 6: All states report higher net transfers following the 14th FC Changes between FY15 and FY16, percentage points of GSDP ODE SE Devolution GE Net Transfers AP 0.94 -0.17 0.77 -0.71 0.06 PJ 0.54 0.25 0.78 -0.31 0.47 MH 0.36 0.09 0.45 0.15 0.60 GJ 0.42 0.08 0.50 0.11 0.61 KA 0.86 0.28 1.13 -0.40 0.73 TN 0.80 -0.37 0.43 0.31 0.74 KL 0.51 0.21 0.72 0.11 0.84 RJ 1.18 -0.26 0.92 -0.04 0.88 HR 0.28 0.04 0.32 0.57 0.89 UK 0.77 -0.14 0.63 0.47 1.10 BH 2.44 -0.95 1.48 -0.18 1.30 TG 0.54 0.06 0.60 0.77 1.37 MP 1.24 0.59 1.83 -0.03 1.81 WB 1.05 0.18 1.23 0.83 2.07 OD 2.10 -0.20 1.90 1.03 2.93 UP 2.72 -0.50 2.22 0.89 3.11 CG 1.25 1.46 2.72 0.91 3.62 JH 2.00 1.01 3.01 1.51 4.52 HP 1.04 -0.11 0.93 3.95 4.88 MZ 1.70 7.08 8.78 -3.44 5.34 Source: State budgets and World Bank staff calculations Figure 35. Net devolution was unambiguously positive Changes between FY15 and FY16, percent of GSDP 6 ODE SE GE Net Transfers 4.9 5 4.5 3.6 4 2.9 3.1 3 1.3 2.1 1.8 2 1.4 1.1 0.7 0.7 0.8 0.9 0.9 1 0.5 0.6 0.6 0.1 0 -1 -2 AP PJ MH GJ KA TN KL RJ HR UK BH TG MP WB OD UP CG JH HP Source: State budgets and World Bank staff calculations 25 « INDIA DEVELOPMENT UPDATE JUNE 2016 Higher devolution If states used the additional funds transferred from the center to reduce their own revenue from the center collections, or to decrease their fiscal deficit (reduce borrowing), thus offsetting increased contributed to resources, state-level expenditures may not have increased. However, as Figure 36 shows, all increased states increased spending in FY16. expenditures by the  Tax revenue effort was generally maintained. Concerns about states replacing own- states tax revenues with revenues from the center did not pan out – at least in FY16. Eleven out of 16 states in our sample increased own-tax revenues in FY16 as a share of GSDP. In only one state – Madhya Pradesh – the decline in own-revenues was greater than the gains from increased central transfers.  There was no clear pattern on state deficits. Excluding the direct impact of Ujjwal DISCOM Assurance Yojna (UDAY), only 7 out of 16 states in our sample reduced their fiscal deficit in FY16; including UDAY, 5 out of 16 states reduced the fiscal deficit. Figure 36. Higher devolution was the main contributor to funding for additional spending in FY16 Changes between FY15 and FY16, percent of GSDP 6 Net Transfers Own-Revenues 4.6 4.7 5 4.2 4.3 4.3 Deficit/Borrowing Resources/Expenditures 3.7 4 3 2.1 1.9 2 1.5 1.0 1.2 0.7 0.8 1 0.3 0.3 0.3 0 -1 -2 -3 KL GJ UK WB KA MH MP HR RJ TN UP CG OD TG HP BH Source: State budgets and World Bank staff calculations Note: Figures for Rajasthan, Haryana, and UP exclude the direct impact of UDAY About half of the Figure 37, Figure 38, and Table 7 contrast the additional resources that states had in FY16 with additional resources the additional expenditures in two specific areas: capital outlays, as a proxy for infrastructure in FY16 (coming spending; and health and education, as proxies for human capital investments. In both cases, from changes in there is a clear positive correlation, with states with greater resources spending more in both central transfers, areas. On average, an additional 1 percentage point in additional resources was converted into own-revenues, and 0.3 percentage points in higher capital outlays, and 0.2 percentage points in health and education deficits) was used to spending, with the remainder used for other revenue expenditures, including transfers to local increase health, administrative bodies. education, and  Health and education expenditures increased almost everywhere in FY16. infrastructure Combined health and education expenditures increased in 13 of the 14 states for which expenditures data was available, with education expenditures generally increasing more than health, likely on account of implementation of the Right to Education Act and as states allocated additional amounts to cover lower contributions from the center for CSS. On average, states increased health and education expenditures by 0.4 percent of GSDP. Uttar Pradesh was a large spender on health and education, channeling slightly over one-third of its significant additional resources in those sectors.  Capital expenditures (excluding UDAY) also increased in many states. Thirteen out of 16 states for which data was available increased capital outlays (Uttarakhand, Gujarat, and Uttar Pradesh were the exceptions), with 5 increasing expenditures in infrastructure by more than 1 percent of GSDP. Across these 16 states, capital outlays increased by 0.3 percent of (all-India) GDP, suggesting a substantial jump compared to INDIA DEVELOPMENT UPDATE JUNE 2016 » 26 the 0.2 percent increase in capital expenditures recorded by the center. In Uttar Pradesh, capital outlays excluding UDAY contracted by more than 1 percent of GSDP, although the extent to which FY15 capital outlays also included equity support to DISCOMs is not clear.  Rajasthan and Kerala stand out as spending the equivalent of over 70 percent of additional resources on health, education, and infrastructure. Some large states – namely Madhya Pradesh and Haryana – have also spent the equivalent of more than 50 percent of their additional resources on health, education, and infrastructure.  The balance of revenue expenditures includes devolution to local levels of governments. Revenue expenditures outside of health and education include transfers to local governments, which are expected to have increased according to the recommendations from the 14th FC. Figure 37. States spent additional resources on health and Figure 38. … as well as on infrastructure (capital outlays) education… Change between FY15 and FY16, percentage points of GSDP Change between FY15 and FY16, percentage points of GSDP 1.4 2.0 1.2 y = 0.1888x - 0.0126 Change in Health & Edu Expenditures R² = 0.4684 1.5 1.0 Change in Capital Outlays 1.0 0.8 0.5 y = 0.2x + 0.0244 0.6 R² = 0.2919 0.4 0.0 0.0 2.0 4.0 6.0 8.0 0.2 -0.5 0.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 -1.0 -0.2 -1.5 -0.4 Change in Resources Change in Resources Source: State budgets and World Bank estimates Source: State budgets and World Bank estimates Table 7: Most states used additional transfers to spend more on health, education, and infrastructure Percentage points of GSDP Resources/ Capital Other Rev. Expenditures Outlays Health Education Health & Ed Exp. KL 0.31 0.24 0.10 -0.09 0.01 0.06 GJ 0.33 -0.11 0.13 0.11 0.23 0.21 UK 0.34 -0.45 0.08 0.14 0.22 0.57 WB 0.67 0.54 0.00 -0.25 -0.26 0.39 KA 0.84 0.01 0.07 0.09 0.16 0.67 MH 0.97 0.36 … … … … MP 1.15 0.55 -0.03 0.06 0.02 0.57 HR 1.46 0.53 0.08 0.19 0.27 0.65 RJ 1.85 1.04 0.35 0.13 0.48 0.33 TN 2.10 … 0.09 0.29 0.38 UP 3.68 -1.32 0.28 0.95 1.24 3.77 CG 4.25 1.31 0.32 0.76 1.07 1.87 OD 4.28 1.43 0.04 0.39 0.44 2.42 TG 4.33 1.11 0.25 -0.23 0.02 3.19 HP 4.65 0.55 0.34 0.90 1.24 2.86 BH 4.66 0.95 … … … … JH 6.81 1.76 … … … … Source: State budgets and World Bank staff calculations Note: Figures for Rajasthan, Uttar Pradesh, and Haryana exclude the direct impact of UDAY 27 « INDIA DEVELOPMENT UPDATE JUNE 2016 Box 2: Revision in State GSDPs In early 2015, the Ministry of Statistics released a new GDP series revising the base year from 2004/05 to 2011/12 incorporating new data as a result of which contribution of industry, particularly manufacturing, to total output increased significantly in the new series while that of services declined. Following the national revision, the base year for states’ GDP (GSDP) series was revised in 2016 from 2004/05 to 2011/12. As a result, the top five states that contributed the most to all-India GDP growth were Maharashtra, Tamil Nadu, Gujarat, Uttar Pradesh, and Karnataka which replaced West Bengal from the previous series (Figure 39). As manufacturing had a greater contribution to GDP under the new series, average growth during FY13-FY15 was negatively affected in Puducherry (-5.3 p.p.), Meghalaya (-4.6 p.p.), Assam (-3.2 p.p.), Madhya Pradesh (-2.6 p.p.), and Uttarakhand (-1.1 p.p.) – as these states saw lower contributions from manufacturing to overall growth as compared to other states, on an average between FY11 and FY15. However, considering the contribution of manufacturing was the second highest in Uttarakhand, the changes under the new series can be attributed to factors other than merely the relative level of manufacturing activity in the state. Simultaneously, on account of their manufacturing sectors’ contribution, growth in some states increased: Gujarat (2.1 p.p.), Tamil Nadu (1.9 p.p.), Arunachal Pradesh (1.5 p.p.), Telangana (1.2 p.p.), and Nagaland (0.8 p.p.). Tamil Nadu, the second largest contributor to overall growth of the economy, benefitted from the revisions (between FY13 and FY15 average growth under old series: 6.0 percent; under the new series: 7.9 percent). 19 The states that have unambiguously fared worst following the revisions are Assam, Meghalaya, Uttarakhand, Delhi, and Puducherry. Under the old series, Assam, Meghalaya, and Uttarakhand, with an average of 6.4, 7.6, and 8.4 percent from FY13 to FY15, were well above the average all-India growth of 5.5 percent in the same period. However, under the new series, the average growths decelerate to 3.1, 3.0, and 7.4 percent. Bihar, one of the most populous states, also saw a dip in growth following the revision: average growth for FY13 to FY15 under the new series: 8.8 percent; under the old series: 9.8 percent (Figure 40) Along with change in trends of growth, the new series also shows divergence in the growth pattern of the GSDP deflator between the new and old series for almost two-thirds of the states. In particular, all states that had lower growth under the revised series experienced the divergence—the deflator for Puducherry decelerates under the new series but remained volatile under the old one during FY13 and FY15, for Delhi the deflator decelerated under the old series but was volatile under the new one for the same time period. Figure 39. Average contribution of states to total GDP Figure 40. …Average growth during FY13-FY15 growth during FY13-FY15 Percent Year-on-year change, percent Maharashtra Tamil Nadu Gujarat Bihar Delhi Gujarat 1.2 Karnataka Uttar Pradesh Delhi Maharashtra Karnataka Andhra Pradesh 12 Kerala Tamil Nadu Rajasthan Rajasthan Kerala Bihar Uttar Pradesh All India Andhra Pradesh 1.0 10 0.8 8 0.6 6 0.4 4 0.2 2 0.0 0 04-05 base 11-12 base 04-05 base 11-12 base Source: CEIC, RBI, and World Bank staff calculations Source: CEIC, RBI, and World Bank staff calculations 19 Contributions are calculated as an average of contributions from FY13 to FY15 under the new series. INDIA DEVELOPMENT UPDATE JUNE 2016 » 28 The Union FY17 budget: a strong signal of the importance of macroeconomic stability Three key thrusts of the budget 1. Macroeconomic The main message the budget sent was that the government values macroeconomic stability and stability a priority, as intends to preserve it; hence, it has chosen to adhere to the fiscal consolidation path set in the deficit target previous budget despite calls for relaxation. Authorities achieved the fiscal deficit target of 3.9 maintained percent in FY16 despite a lower nominal GDP,20 and confirmed their commitment to remaining on the path to achieve a deficit of 3.5 percent of GDP in FY17 despite calls for flexibility because of the headwinds the economy is facing. There had been substantial uncertainty about the target deficit for FY17, with many analysts projecting 3.7 or 3.8 percent of GDP. To address the issue of meeting fiscal deficit targets in the future, a committee has been announced to take a fresh look at the Fiscal Responsibility and Budget Management (FRBM) act, especially the fiscal targets, and examine whether a range rather than a point target is most appropriate. 2. Concern for the The second key message the budget sent was that the government is sensitive to the plight of the stressed rural rural economy, with some emphasis on productivity over populism. With the rural sector reeling economy after two successive bouts of weak monsoons and upcoming elections in key states such as Uttar Pradesh, the budget increased spending on rural sectors by 31 percent. The strategy seems to be to boost rural consumption and enhance safety nets by enhancing the job guarantee scheme in the short term, while improving real incomes through productivity improvements in the medium term, without resorting to overt populism through higher minimum support prices (MSPs) or other subsidies. For rural income support, record commitments were announced in the flagship rural job guarantee scheme MGNREGA—amounting to INR 38,500 crores (approximately USD 5.6 billion). This allocation at 0.26 percent of GDP is the highest among all the 28 Centrally Sponsored Schemes. To provide safety nets in the rural agricultural sector, a crop insurance scheme – Pradhan Mantri Fasal Bima Yojna – has been put in place. To increase productivity in the agricultural sector, an allocation of INR 54,212 crores (INR 47,912 crores excluding IEBR21) or approximately 0.36 percent of GDP has been made. This is a sharp increase of over 84 percent in the budgeted amount from the revised estimates of FY16. 3. Infrastructure The third key point was that the infrastructure push would continue. Although the allocation for push to continue capital expenditure declined as a share of total expenditure, the thrust on public investments has been maintained. Spending in infrastructure sectors, including both current and capital expenditures, is expected to increase significantly by 16 percent. Highways alone were allocated INR 2,21,246 crores (USD 52 bn). The budget also proposed five new ultra-mega power projects of 4,000 MW each, besides aiming to corporatize public sector ports to improve efficiency. Moreover, the decline in capital expenditures by the center is more than offset by an increase in grants for the creation of capital assets (largely grants to states, which are classified as revenue expenditure, but are used to build infrastructure by the states). And some questions about the details of consolidation Fiscal consolidation Figure 41 summarizes the changes in the budget. relies on higher  On the expenditure side, few details were provided on the implementation of the one- asset sales, leading rank-one-pension (OROP) and the recommendations of the 7th Central Pay Commission to risks to the fiscal (CPC). Pension spending was increased by 0.1 percent of GDP on account of OROP, path. while other non-plan revenue expenditures (excluding pensions, interest, subsidies, and transfers to states) were flat as a percent of GDP, suggesting either a staggered implementation of the CPC recommendations (some analysts argue that pay will be increased while allowance increases will be postponed) or risks that actual expenditures 20 Advance estimates (AE) of GDP in FY16 are INR 135,671,920 million, lower relative to a budgeted estimate of INR 141,089,450 million at the time of presenting the budget in February, 2015. 21 IEBR – Internal and Extra Budgetary Resources such as direct borrowings of Public Sector Undertakings. 29 « INDIA DEVELOPMENT UPDATE JUNE 2016 will exceed the budget (with attendant risks of fiscal slippage or a contraction in capital expenditure). The increase in civil servant pay is lower than the estimates by investment banking analysts of as much as 0.8 percent of GDP.  The subsidy bill is budgeted to decline from 1.9 percent of GDP in the current fiscal year to 1.7 percent in FY17, almost entirely because of the further decline in oil prices from the FY16 average given that no new subsidy reforms were announced. Two key announcements were made, however: pilots for using DBT for fertilizer, and legislation that will provide the statutory basis for using Aadhaar for targeting subsidies.  Other than the 0.1 percent net reduction from subsidies and pensions, grants for creating capital assets (+0.1 percent) and other plan expenditures on the revenue account (+0.1 percent) also increased, and overall revenue expenditure is budgeted to increase by 0.1 percent of GDP. Since capital expenditures by the center are expected to decrease by 0.2 percent of GDP, overall expenditures remain flat as a percent of GDP.22  On the revenue side, y/y growth in overall tax revenues is budgeted at 11.7 percent (10.8 percent of GDP or 0.1 percent of GDP higher than FY16). However, the additional tax revenue is transferred to states, and net tax revenues to the center remain flat at 7.0 percent of GDP. While overall reasonable, estimates for personal income tax growth (18%, based on the announced amnesty and higher surcharge on upper incomes) and excise taxes (12% given average excises should be higher in FY17 following several increases of the excise on petroleum in FY16) have a higher risk of not materializing.  With net tax revenues and expenditures flat, fiscal consolidation comes from higher non- tax revenues, specifically divestments and auctions of cellular spectrum, which combined are budgeted to increase by 0.4 percent of GDP from FY16. Using the more standard definition of fiscal deficit excluding divestments from revenue, consolidation is less marked, with the center’s deficit declining from 4.2 to 3.9 percent of GDP. On the one hand, achieving fiscal consolidation through asset sales avoids a negative impact on growth; on the other, given the history of slippages in divestments and the ambitious target on spectrum sales, there is a high risk that such revenues will fall short. This would result in either fiscal slippage, or a combination of expenditure cuts and revenue increases that would lead to a negative fiscal impulse. Figure 41. Planned consolidation is achieved through higher divestments and spectrum sales Changes between FY16 and FY17, percent of GSDP 15.0 Net Borrowing Net Borrowing (Fiscal Deficit, GOI), (Fiscal Deficit, GOI), 10.0 3.9 Divestments, 0.2 3.5 Divestments, 0.4 Other non-tax Other non-tax Spectrum Sales, Spectrum Sales, revenue, 1.5 revenue, 1.5 0.4 0.7 5.0 Tax Revenue (net to Tax Revenue (net to center), 7.0 center), 7.0 0.0 Revenue Revenue Expenditure, excl. Expenditure, excl. -5.0 grants to create grants to create capital assets, capital assets, (10.4) (10.4) Grants for creation of Grants for creation of -10.0 capital asseets, capital asseets, Capital expenditure, Capital expenditure, (1.1) (1.0) (1.5) (1.7) -15.0 FY16 FY17 Source: Union budget and World Bank staff calculations Efforts have been Additional revenues will be raised by (i) raising the services tax to 15 percent by including a new made to make cess, (ii) raising specific taxes on tobacco products by 10 to 15 percent, (iii) introducing new taxes 22 Some figures do not add because of rounding. INDIA DEVELOPMENT UPDATE JUNE 2016 » 30 income taxes more on automobiles, and (iv) increasing taxes on high earners with targeted tax relief provided to small progressive while taxpayers. Base erosion resulting from an increase in tax slab limits was avoided. Table 8 improving the base summarizes the key changes in tax policies. Table 8: Tax measures in Budget FY17 Direct taxes Increase in Rate/Amount Additional dividend tax on gross dividends on entities earning dividend income of more than 10 lakhs per year 10% Income tax surcharge increased from 12 to 15 percent on individuals earning incomes of more than 1 crore 3% Indirect taxes Additional tax deduction at source on luxury cars valued above 10 lakhs and cash transactions exceeding 2 lakhs of goods and services 1% Security transaction tax on “options” increased from 0.017 to 0.05% 0.04% Krishi Kalyan Cess on all taxable services 0.50% 1% on petrol, CNG, LNG, 2.5% on diesel, Infrastructure cess on cars 4% on higher engine capacity/SUVs Clean environment cess hiked from INR 200 to INR 400/ton 200/ton Excise duties hiked on jewelry if not claiming input tax credit 1% Excise duties hiked on clothes over INR 1000 if not claiming input tax credit 2% Excise duties hiked on tobacco products except bidi 10 to 15% In FY17, modest consolidation at the level of the general government (states + center) The general government deficit is expected to decline modestly in FY17 The general The fiscal deficit of the general government had progressively declined to 4.0 percent of GDP by government deficit 2008, before picking up sharply in 2009 as the state, especially the union government, stepped in has declined from a to cushion the impact of the global financial crisis on the economy. Since then, the deficit has peak of nearly 10 been declining, and the General Government gross fiscal deficit stood at 6.5 percent of GDP in percent in 2010 FY16 (GoI definition, excluding UDAY)23, 24 and is expected to be reduced to 6.4 percent in FY17 (Figure 42). In the post crisis period (2009 onward) the states have accounted for approximately 32 percent of the general government fiscal deficit, compared to 40.4 percent between 2000 and 2008, implying a relatively prudent fiscal stance on the part of the states, reflecting their implementation of state Fiscal Responsibility and Budget Management acts. In FY17 most states In presenting their budgets for FY 2017, a majority of the states have targeted deficits close to 3 aim to stay within percent of GSDP to stay within their FRBM limits and maximize available fiscal space (Figure the 3 percent fiscal 43). An analysis of the budget documents of 19 large states shows that all states except Telangana, deficit threshold Odisha, Madhya Pradesh, and Himachal Pradesh have committed to confining their fiscal deficits while maximizing below the 3 percent threshold.25 Most states ran deficits below 3 percent in FY16 and planned to the available fiscal increase their deficit slightly in FY17 to close to 3 percent. The exceptions are the states entering space the UDAY scheme, as well as Bihar, which ran a deficit of 5.9 percent of GSDP in FY16 and now aims to bring it to less than 3.0 percent in FY17. 23 The GoI definition includes capital receipts as revenues. The World Bank and the IMF exclude non-recurrent capital receipts from deficit calculations, which yield general government deficit of 7.2 percent for FY16 compared to 7.0 percent in FY15. 24 The Central Government announced the UDAY scheme in November 2015 for the financial turnaround of the state power distribution companies (DISCOMs). According to the rules of the UDAY scheme, the takeover of DISCOM debt (essentially a capital transfer) is not included when computing the state fiscal deficits for FY16 and FY17. See Part C for additional details on the program. 25 The 14th Finance Commission allowed a relaxation of the 3 percent fiscal deficit limit, by an additional 0.5 percent conditional upon meeting certain targets—0.25 percent if debt to GSDP ratio was less than or equal to 25 percent of GSDP in the preceding year and an additional 0.25 percent if interest payments were less than or equal to 10 percent of revenue receipts in the previous year. Madhya Pradesh, Telangana, and Odisha satisfy both conditions. 31 « INDIA DEVELOPMENT UPDATE JUNE 2016 Figure 42. The deficit of the general government has been Figure 43. In FY17, states plan to stick to FRBM limits, on a gradually declining path since 2012 excluding the impact of UDAY Percent of GDP, excluding UDAY Percent of GDP 12 10.0 9.6 9.3 9.0 FY16 FY17 10 9.2 9.3 8.3 8.3 8.0 FY16xU FY17xU 7.8 8 7.2 6.9 6.6 6.9 6.8 6.9 7.0 6.5 6.4 5.4 6.0 5.6 6 4.0 5.0 4.0 4.3 4 3.8 4.0 3.5 3.5 3.2 3.0 3.0 2.9 3.0 2.9 2.9 3.0 3.0 3.0 2 3.0 2.5 2.1 2.2 2.0 1.6 0 2001 2003 2005 2007 2009 2011 2013 2015 2017 1.0 0.1 0.0 Central Fiscal Deficit State Fiscal Deficit BH MZOD KA TG HP UKMHAP CG GJ JH TN KL MP RJ HR UP Gross Fiscal Deficit Source: RBI and World Bank staff calculations Source: RBI and World Bank staff calculations Note: RJ, HR, and UP, with and without UDAY debt takeover Most states have Resources available to the states in FY17 are expected to be lower than in FY16 as a share of budgeted in FY17 to GDP. Nonetheless, for the 18 states for which expenditure data on health and education are increase capital available from their respective state budgets, approximately one-third have increased their expenditures while allocation toward health and education (Table 9). States have budgeted on average 3.4 percent health and of GSDP on education, marginally lower than in FY16. Notably, while Madhya Pradesh education budgeted to increase its allocation on education by 0.6 percent of GSDP in FY17, Bihar cut expenditures are education expenditure by 1.1 percent of GSDP. On health, states have budgeted to cut more likely to have expenditures on average by 0.1 percent of GDP compared to 2016. Most of the low-income been rationalized, as states such as Bihar, Uttar Pradesh, Chhattisgarh, and Madhya Pradesh have increased health states adjust to the allocations, however. Karnataka appears to have reduced their health and education new sharing expenditures drastically, but this is mostly due to the surge in GSDP estimates for the state patterns for CSS following the methodological correction at the central level (see Box 2 on changes in the state- level GSDP series). Capital expenditures have been increased by 11 of the 17 states for which data is available. The states that have increased capital expenditures substantially include Telangana, Madhya Pradesh, and UP. Table 9: Changes in health, education, and capital expenditures in FY17 Changes between 2016 and 2017 State Fiscal Deficit Own Tax Revenues Devolution Grants Health Education Capex AP 0.2 0.3 0.0 1.0 0.1 0.2 0.2 BH -3.0 0.0 0.0 1.6 0.4 -1.1 0.0 CG 0.2 -0.8 0.2 -0.2 0.1 0.0 0.3 GJ 0.1 0.1 0.0 0.0 -0.1 -0.2 0.1 HP -0.5 -0.2 -0.2 -1.1 -0.1 -0.3 -0.5 HR 1.7 -0.3 -0.1 -0.3 0.0 0.0 -0.7 JH -0.1 -0.2 -0.2 -0.3 -0.1 -0.1 0.2 KA -0.7 -3.7 -1.0 -0.6 -0.4 -1.1 -0.7 KL 0.0 0.3 0.0 0.2 0.0 0.2 0.4 MH -0.3 -0.1 0.0 -0.1 -0.1 -0.1 0.0 MP 0.0 0.3 -0.4 0.0 0.1 0.6 1.4 MZ -0.9 -0.1 0.2 -3.7 -1.0 -1.5 1.1 OD 0.8 -0.5 -0.2 -0.4 0.0 0.1 -0.3 RJ -0.6 0.2 0.0 0.0 0.0 0.0 -0.7 TG 0.6 0.7 -0.1 0.0 0.2 -0.1 1.7 TN 0.0 -0.2 0.0 -0.2 0.0 -0.1 … UK 0.4 0.8 0.0 0.9 0.0 0.3 0.3 UP 0.0 0.3 0.0 0.1 0.1 -0.3 0.9 Source: State budgets and World Bank staff calculations INDIA DEVELOPMENT UPDATE JUNE 2016 » 32 Debt ratios have been stable Debt ratios have Most of the debt of both the union and the states is raised domestically. The share of debt raised been stable, and domestically increased from 88.2 percent to 95.2 percent between 2001 and 2015. The combined debt remains debt stock of the center and the states has been declining since 2004 and stabilized at close to overwhelmingly 67 percent of GDP since FY13 (Figure 44). Debt ratios of individual states vary significantly domestic. from 33 percent of GSDP in West Bengal to 15 percent in Chhattisgarh (Figure 45). Variability in the yields of state development loans (SDLs) is limited, and has little relation with debt levels. Figure 44. The combined debt stock has been stable Figure 45. Debt ratios vary significantly across states Percent of GDP LHS: Percent of GDP (FY15 RE); RHS: Yields, percent 100 40.0 9.0 35.5 90 8.8 86 88 88 85 35.0 80 81 80 8.6 78 72 71 70 69 30.0 70 70 68 8.4 60 25.0 8.2 50 20.0 8.0 40 15.2 15.0 7.8 30 7.6 20 10.0 7.4 10 5.0 7.2 0 0.0 7.0 MH TN MP AS KK AP PJ BH RJ KR CH JH HY UT UP WB GJ OD GO Center States Total Debt to GSDP ratio Yields on SDLs (RHS) Source: MOF and World Bank staff calculations Source: RBI 33 « INDIA DEVELOPMENT UPDATE JUNE 2016 Annexes Annex I – Key Fiscal Indicators 2014–2015 2015–-16 2016–17 Actuals RE BE Non-plan expenditure 9.6 9.6 9.5 On revenue account 8.9 8.9 8.8 Subsidies 2.1 1.9 1.7 Pensions 0.7 0.7 0.8 Interest payments 3.2 3.3 3.3 Defense 1.1 1.1 1.1 Grants to states & UTs 0.6 0.8 0.8 On capital account 0.7 0.7 0.7 Plan expenditure 3.7 3.5 3.7 On revenue account 2.9 2.5 2.7 On capital account 0.8 1.0 1.0 Total expenditure 13.3 13.2 13.1 Revenue expenditure 11.7 11.4 11.5 Capital expenditure 1.6 1.8 1.6 Grants for creation of capital assets 1.0 1.0 1.1 Total transfers to states 5.5 6.2 6.2 Tied grants to states 2.2 1.7 1.7 Untied grants to states 0.6 0.8 0.7 Devolution of states' share in taxes 2.7 3.7 3.8 Revenue Receipts 8.8 8.9 9.1 Tax revenue (net to center) 7.2 7.0 7.0 Direct tax 5.6 5.5 5.6 … of which CIT 3.4 3.3 3.3 … of which PIT 2.1 2.2 2.3 Indirect tax 4.4 5.2 5.2 … of which excise 1.5 2.1 2.1 … of which services 1.3 1.5 1.5 0.0 0.0 0.0 Others, net Non-tax revenue 1.6 1.9 2.1 … of which spectrum auctions 0.2 0.4 0.7 Capital receipts 0.4 0.3 0.4 Recoveries of loans 0.1 0.1 0.1 Other receipts (divestments) 0.3 0.2 0.4 Net receipts 9.2 9.2 9.6 Source: Union budget and World Bank Staff calculations 2014–2015 2015–16 2016–17 Actuals RE BE Union fiscal deficit, GOI definition 4.1 3.9 3.5 1 Union fiscal deficit World Bank/IMF definition 4.4 4.2 3.9 2, 3 Deficit of the states 2.8 2.9 2.9 General government deficit, GOI definition 6.9 6.8 6.4 General government deficit, World Bank/IMF 7.2 7.0 6.8 definition Source: Union budget, state budgets, and World Bank Staff calculations Notes: 1Excludes proceeds from divestments from receipts. 2Source: analysis of 20 state budgets accounting for 90 percent of GDP. 3Excludes impact of UDAY bonds. INDIA DEVELOPMENT UPDATE JUNE 2016 » 34 Annex II – Evolving Fiscal Federalism India’s system of fiscal transfers includes both devolution of taxes from the central divisible pool, as well as a number of grants. Fiscal transfers between the center and the states are of two broad kinds – statutory (those recommended by the Finance Commissions) and non-statutory (those that are outside the purview of the Finance Commission). Table 10 summarizes the different types of transfers. Table 10: India’s instruments for fiscal transfers to states Finance Commission Transfers Other (Non-FC) Transfers Tax devolution Grants-in-aid Plan grants Non-plan grants Revenue deficit support Normal central assistance grants based on Gadgil-Mukerjee Formula (untied) Disaster relief grants Additional central assistance for specific scheme/purpose (tied) Grants to local bodies Special central assistance to north eastern and hilly states (untied) Sector-specific grants Special plan assistance State-specific grants Central plan schemes Centrally sponsored schemes (tied) Source: Authors Fiscal federalism is evolving in India along three broad themes. Three broad trends are discernable in recent times – first toward greater transparency, second toward rationalization and simplification of the transfers and finally a relative shift toward untied transfers.  Greater transparency. The practice of routing funds through two channels – one via the consolidated funds of the state and the other directly to implementing agencies has been discontinued. To enhance transparency, since FY15 funds for Centrally Sponsored Schemes (CSS) are being routed through the consolidated funds of the state.  Rationalization and simplification of transfers has been achieved by consolidating the number of CSSs. The Sub-Group of Chief Ministers on the rationalization of CSSs classifies the schemes into core and optional schemes. While the core (and so-called “core of the core”) schemes have been accorded high importance, among optional schemes states have been allowed not only to choose whether to join those schemes, but are also given the flexibility to allocate resources across optional schemes subject to the overall CSS envelope for the state. Rationalization of the schemes has been accompanied with a lower share of the center in the CSSs and standardization of sharing patterns across schemes. Furthermore, with the end of the 12th five-year plan in FY 2016–17, expenditures will no longer be classified as plan and non-plan with a corresponding simplification in the kinds of interstate transfer instruments.  The shift toward untied transfers noted above is reflected in the increased use of tax devolution as an instrument of fiscal transfers, relative to grants, to allow states greater autonomy of expenditure. 35 « INDIA DEVELOPMENT UPDATE JUNE 2016 Annex III. Calculations of the impact of the 14th Finance Commission  Overall devolution effect (ODE). First, we isolated the impact of the overall increase in resources that came from the higher share of the divisible tax pool going to states (increase from 32 to 42 percent). Let S i,t be the share of the divisible pool of state i in year t = FY15, FY16, and Dt the overall divisible tax pool in year t. The ODE is given by (Si,FY15DFY16)/GSDPi,FY16 – (Si,FY15DFY15)/GSDPi,FY15. Since DFY16 / DFY15 -1 = 49.8 percent and higher than GSDP growth in every state, this effect will be positive everywhere.  Share effect (SE). Second, we isolated the impact of the change in the relative shares of individual states in overall devolution (the size of the piece of the devolution pie that each state gets) resulting from the new state shares recommended by the 14th FC. Using the same notation as above, the SE is given by (Si,FY16DFY16)/GSDPi,FY16 – (Si,FY15DFY16)/GSDPi,FY16. In this case, it is possible that some states will gain, while others will lose given that Si,FY16 <> Si,FY15. Adding ODE and SE gives (Si,FY16DFY16)/GSDPi,FY16 – (Si,FY15DFY15)/GSDPi,FY15, the overall change in resources for state i.  Grant effect (GE). One potential source of losses relative to the 13th FC period is that states may need to raise their contributions toward CSS. If higher devolution must be used to compensate for lower contributions from the center toward CSS, the state will not in fact have additional resources. A thorough analysis of this question would require considering each scheme separately and analyzing the contributions of the state and center toward it. We take a simplified approach and assume that central grants would have remained constant as a share of GSDP in a counterfactual “no 14th FC” scenario. This implies that expenditures previously covered by grants from the center under the 13th FC would be growing at the same rate as GSDP. Given this assumption and letting Gi,t be the grants received by state i in year t, GE is given simply by Gi,FY16/ GSDPi,FY16 – Gi,FY15 / GSDPi,FY15. Therefore, the GE is negative if grants increased by less than GSDP.  The sum of ODE, SE, and GE gives the net impact of the change in fiscal transfers for each state. INDIA DEVELOPMENT UPDATE JUNE 2016 » 36 2. The Rural-Urban Divide in Private Consumption The resilience of On the one hand, as described in Part A, household consumption growth has remained a steady final consumption driver of GDP growth in FY16. On the other hand, (a) most of India’s population lives in rural expenditure in the areas; (b) most of the population in rural areas depends, directly and indirectly on agriculture; face of stress in and (c) the agriculture sector is depressed following two years of sub-par monsoons, declining rural areas is at first water tables, deteriorating soil health from overuse of urea, and limited increases in minimum glance puzzling support prices (MSPs). How to reconcile the two sets of facts? This note explores (i) inequality in consumption between rural and urban areas is high, such that the three potential contribution of urban consumption to overall consumption is greater than the share explanations of urban households suggests (ii) rural households are becoming progressively less dependent on agriculture for incomes, therefore the negative impact of weak agriculture is more muted than the high share of rural households suggests; and (iii) despite greater diversification of incomes, most rural households still depend directly and indirectly on agriculture, and private consumption has in fact felt some of the impact of rural stress A disproportionate share of consumption takes place in urban areas The relative Data from the National Sample Survey suggests that inequality in consumption between rural resilience of private and urban areas in India is high and has been increasing – at least between 1994 and 2005, and consumption in the between 2005 and 2012 (Figure 46). On average, urban consumption was 59 percent higher than face of rural stress rural consumption in 2011 compared to 49 percent in 2005. Therefore, the share of partly reflects consumption by urban households in overall expenditures is higher than the share of urban inequality between households in the population, and until recently, this gap had been increasing. This helps explain rural and urban why private consumption growth can be sustained by a minority of households. areas with respect to consumption Figure 46. Rural-urban consumption gaps have increased Figure 47. Farm activities comprise a smaller number of jobs Urban-rural consumption gap Employment share, percent of usual status 2004-05 2011-12 .8 .7 Non-Farm Casual 7 14 .6 6 .5 Non-Farm Regular 8 .4 14 .3 Non-Farm Self 14 .2 Farm Casual 25 .1 22 0 0 10 20 30 40 50 60 70 80 90 100 Farmers 47 Percentile 42 1993 2004 2011 0 10 20 30 40 50 Source: NSSO and World Bank staff calculations Source: NSSO and World Bank staff calculations 37 « INDIA DEVELOPMENT UPDATE JUNE 2016 The rural economy has (partly) diversified from agriculture Rural India is There has been a notable decline in the share of adults engaged in farming or as agricultural increasingly farm labor, dampening the impact of weak agricultural output on rural consumption. Non-farm diversifying away casual jobs have been the predominant source of new jobs, mostly in construction (Figure 47). from the farm With people taking up more non-farm casual farm jobs, the sources of household income have diversified. Meanwhile, the share of households whose main source of income is from farming or casual farm labor also has decreased as shown in Figure 48. Nonetheless, agriculture remains the principal source of household income for more than half of rural households, when both farming and farm labor are taken into account. Figure 48. …and a declining share of income in rural Figure 49. The rural-urban wage gap has narrowed India Principal source of household income, percent of households Wage gap 2012 2005 13 Non-Farm Casual 11 Regular/Other 15 11 Non-Farm Self 16 16 21 Farm Casual 27 34 Farm 36 0 10 20 30 40 Source: NSSO and World Bank staff calculations Source: NSSO and World Bank staff calculations In contrast to the There was an unprecedented rise in wages for casual labor (wages for the unskilled) between widening gap 2005 and 2012, especially in rural areas. Since over 80 percent of the wage employment in rural between rural and areas is for casual labor, growth was faster at the bottom end of the distribution than at the top. urban consumption, In urban areas, by contrast, casual labor accounted for less than 25 percent of wage employment. rural-urban wage Consequently, wage gaps for most of the population have been declining (Figure 49). High self- gaps had been employment reduces the significance of wages in explaining consumption patterns, however, as closing owing to wages capture only a part of the story. A total of 55 percent of rural workers and 40 percent of fast wage growth in urban workers are self-employed. In the absence of earnings information of the self-employed, rural areas. the understanding of rural-urban consumption gaps remains incomplete. For most rural families, incomes – and consumption – are still linked to agriculture Poor performance of In addition to lower production owing to the drought, percent increases in minimum support agriculture results in prices declined during the last two years from an average of 5.1 percent in 2013–14 to 2.1 percent direct effects to in 2015–16 for Kharif crops, as input and global commodity prices declined. Consistent with households that lower agricultural incomes, real wage growth of agricultural laborers turned negative for several depend directly on occupations for an average of four months in Apr–Nov 2015 (Figure 50). Even households not it, and also in directly dependent on agriculture were under stress. Construction sector activity (which negative spillovers accounts for 11 percent of the rural economy) displayed increased volatility, and wages in the to other economic sector also declined, while the growth of wages in other non-agricultural occupations slowed. activities in rural Rural stress was also visible from increased reliance on the rural employment guarantee scheme areas INDIA DEVELOPMENT UPDATE JUNE 2016 » 38 (MGNREGA), reflecting hidden demand for employment in rural areas. MGNREGA person days generated in FY16 grew by 38 percent compared to the previous year (Figure 51). Figure 50. Real wages for rural men have decelerated or Figure 51. MGNREGA person days generated in FY16 downright declined grew, reflecting the stress in rural areas Rural wages adjusted for inflation: male workers, y/y, percent Person-days generated under MGNREGA Carpenter Mason Construction 250 Ploughing Sowing 10 8 200 6 4 150 2 0 100 -2 -4 50 -6 0 FY 2012-2013 FY 2013-2014 FY 2014-2015 FY 2015-2016 Source: CEIC and World Bank staff calculations Source: CEIC and World Bank staff calculations The pace of Despite the clear signs of rural stress, as a result of more diversified sources of income, the consumption improvement in agricultural output in the fourth quarter of FY16, and the disproportionate suggests little weight of urban consumption, private consumption growth in FY16 was higher than during the impact from last normal monsoon in FY14 (7.4 percent and 6.8 percent, respectively). Nevertheless, the fastest agriculture pace of household consumption growth was between December 2013 and September 2014, which suggests possibly a small negative impact in overall consumption growth (Figure 52). Figure 52. Consumption growth has accelerated but remains below levels of mid-2014 Private final consumption expenditure growth, y/y, percent (solid line) and four-quarter moving average (dotted line) 12.0 10.0 8.0 8.3 7.4 6.0 4.0 2.0 0.0 Sep-12 Dec-12 Sep-13 Dec-13 Sep-14 Dec-14 Sep-15 Dec-15 Jun-12 Jun-13 Jun-14 Jun-15 Mar-13 Mar-14 Mar-15 Mar-16 -2.0 -4.0 Source: CSO and World Bank staff calculations Prospects of a normal monsoon should therefore lead to a boost in growth In parallel with Rural India can make a significant contribution to growth if agricultural output increases. Table more diversified 11 and Figure 53 show key statistics from the recently released Socio-Economic and Caste sources of income Census (SECC) that shed some light on the characteristics of rural households and their asset and early wage ownership. Income levels and job profiles in rural areas remain relatively low: less than 8 percent 39 « INDIA DEVELOPMENT UPDATE JUNE 2016 gains, the Socio- of the highest earners in rural households earn more than INR10,000 per month and more than Economic and 51 percent of the population is employed as manual casual labor. However, a growing Caste Census proportion of these rural households are owners of durable goods: more than 20 percent of suggests an rural households own motorized vehicles, more than 10 percent own a refrigerator, and two- emerging consumer thirds own a mobile phone. In addition, the declining dependence of rural households on class in rural India agriculture is visible in household participation in formal jobs: nearly 10 percent of the households have salaried jobs. Since 73 percent of the households continue residing in rural areas, this emerging consumer class presents great potential for increasing consumption demand. Figure 53. Pent-up consumption of durable goods in rural areas may be substantial Percent 80 70 All-India average 65.7 67.1 60 Typical state (median) 50 40 30 20 10.7 8.9 10 3.5 1.7 0 Households having Households Own a Households Owns kisan credit card with refrigerator mobile only the credit limit of Rs.50,000 and above Source: SECC and World Bank staff calculations Table 11: A rural consumer class is emerging Key statistics from the SECC for All India rural households, percent (unless otherwise noted) Total households (millions) 244 Households with one member with salaried job 9.70% Households in rural areas 73.4% With government job 5.00% Average household size (persons) 4.9 With public sector job 1.10% Women-headed households 12.8% Households with private sector job 3.60% Illiterates 35.7% With main source of income from: Disabled 1.5% Cultivation 30.1% SC households 18.5% Manual casual labor 51.1% ST households 11.0% Households with land 44.0% With rented-in dwellings 3.5% Households with no land 56.0% Houses with kuccha walls 44.5% Total un-irrigated land in acres 40.0 Households not consented to public list 27.8% Households with mech 3/4 wheeler agri equipment 4.1% Less than Rs. 5,000 74.5% Households with irrigation equipment 9.9% Households with motor. 2/3/4 wheelers or fishing Own Rs. 10,000 or more 8.3% a boat 20.7% Own a Refrigerator 11.0% Source: NSSO and World Bank staff calculations INDIA DEVELOPMENT UPDATE JUNE 2016 » 40 C. Financing Double-Digit Growth i. India will need a robust financial sector to finance double-digit growth Orderly development of the financial sector and economic growth go hand-in-hand India’s ambition to The link between financial development and economic growth has been hypothesized at least accelerate GDP since Schumpeter, who in 1911 argued that financial intermediaries influenced economic growth growth to double- through their role in allocating society’s savings toward specific investments (Ang, 2007). A well- digits will require a functioning financial sector allocates savings toward the highest-return investments, increasing strong financial the effectiveness of aggregate investments and overall productivity growth in the economy. A sector number of papers (King and Levine, 1993; Beck, Levine and Loayza, 2000; Calderon and Liu, 2003; Levine, 2005; Ang, 2007; Cecchetti and Kharroubi, 2012) confirm the link between finance and growth, and establish that development of the financial sector directly contributes to higher rates of economic growth. While there is some disagreement about the mechanisms, many authors find that financial development boosts growth by increasing private investment;26 and nearly all find that financial development (at least up to a point) raises productivity, 27 possibly by facilitating technology adoption (Cole et al., 2015). In a recent paper, Hasan et al. (2016), find that the efficiency of financial intermediaries is especially important for economic growth. On the other hand, instability caused by financial crises is harmful to growth (Stiglitz, 2000), highlighting the importance of orderly financial sector development. Figure 54 illustrates the relationship 26 King and Levine, 1993; Ang, 2007; and Calderon and Liu, 2003 find favorable evidence; Beck et al. do not. 27 The global financial crisis led to questions of whether there can be “too much” financial development such that after a point the effect on productivity and growth becomes negative. Cecchetti and Kharroubi (2012) study this question and find support for the “inverted-U” hypothesis. However, their work also confirms that at “low levels” of credit-to-GDP (which they estimate as below 100 percent of GDP vs. 50 percent in India), a larger financial system does go hand-in-hand with higher productivity growth. 41 « INDIA DEVELOPMENT UPDATE JUNE 2016 between credit growth and GDP growth for a few large emerging market countries; in all cases, the correlation is evident. Figure 54. Financial development and economic growth are correlated in most countries Left axis: Real GDP; Right axis: Real Credit Growth; change from the previous year; percent 10 Indonesia 50 10 Mexico 35 6 South Africa 16 40 30 14 8 5 8 30 25 12 6 4 20 20 10 6 15 3 8 4 10 4 10 2 6 2 0 5 1 4 -10 2 0 0 2 -20 0 1983 1990 1997 2004 2011 -5 0 -2 0 -30 -10 -1 -2 -4 -40 -2 -15 -2 -4 Source: WDI and World Bank staff calculations India’s financial sector has built up important strengths India’s financial Guided by a wide-ranging reform agenda reflecting the implementation of recommendations of sector has come a successive government-sponsored high-level committees, there have been steady improvements long way following in the legal, regulatory, and supervisory frameworks. These reforms supported significant liberalization more growth in bank credit to the private sector that accompanied the post-1992 growth spurt, such than two decades that India also shows a close correlation between credit and GDP growth (Figure 55). By 2008, ago, and has been a India’s ratio of domestic credit to the private sector by banks was on par with, or higher than key contributor to that of many major emerging markets (Brazil, Indonesia, Russia), even if lower than in Turkey, growth South Africa and especially China (Figure 56). Figure 55. Credit growth surged following the early 1990s Figure 56. India’s credit-to-GDP ratio is in line with most reforms, supporting growth of its emerging market peers Left axis: Real GDP; Right axis: Real Credit Growth; change from the previous year; percent Domestic credit to private sector by banks, percent of GDP 10 25 80 160 9 70 8 20 60 120 7 50 6 15 40 80 5 30 4 10 20 40 3 10 2 5 0 0 1 India Indonesia 0 0 Russian Federation South Africa 1963 1966 1969 1972 1975 1978 1981 1984 1987 1990 1993 1996 1999 2002 2005 2008 2011 2014 Turkey China (RHS) Brazil Source: WDI and World Bank staff calculations Source: WDI and World Bank staff calculations INDIA DEVELOPMENT UPDATE JUNE 2016 » 42 Financial access has been expanding Financial access has India has a relatively high share of firms’ working capital financed through the banking sector improved, but as compared to benchmark countries and a relatively high share of firms that use the banking remains somewhat sector to fund their operations (Figure 57). However, India can reach a higher share of firms below comparator using banks for financing working capital, such as Brazil or Turkey. Moreover, only 21 percent countries of firms report having a loan or line of credit, compared to nearly 60 percent in Brazil, 25 percent in China and 30 percent in South Africa. Figure 57. Financial access of firms in India is in line with comparator countries Percent 70 59.2 60 52.3 50 42.4 40.2 40 36.4 32.0 30.1 25.3 26.9 30 22.1 21.3 21.5 21.1 21.3 21.6 17.8 16 20 10.9 6.4 7.1 8.1 10 0 India (2014) Brazil (2009) China (2013) Mexico Turkey South Africa Russia (2010) (2014) (2007) (2012) Percent of firms using banks to finance working capital Proportion of working capital financed by banks (%) Percent of firms with a bank loan/line of credit Source: World Bank Enterprise Surveys The GOI and RBI According to World Bank’s Global Findex data (2014), 53.1 percent of adults above the age of have supported 15 in India owned formal bank accounts but only 6.4 percent borrowed from institutional and initiatives to expand formal sources. Substantively building on earlier initiatives, the Government in 2014 announced availability of the “Jan Dhan Yojana” (PMJDY) financial inclusion initiative, which gives access to all financial services to households to transactions accounts with a limited credit facility (overdraft) backed by a the general guarantee fund, as well as micro-insurance, pension, and other services, alongside an effort to population promote financial literacy. As of June 1, 2016, 220 million basic banking accounts were open under PMJDY with a cumulative balance of USD 5.8 billion. A series of related measures are also under development to promote greater use of electronic payments. Benefits of the PMJDY scheme, including higher penetration for direct-benefit transfer (101 million accounts are “Aadhaar seeded”), are yet to translate into credit offtake. Moreover, transforming access into use and ensuring that the initiatives are financially attractive from a bank’s business perspective (and therefore self-sustaining) remains a challenge. Rates of dormancy have been reduced, but remain high (26 percent of the accounts have zero-balance) and banks have pointed to significant transaction costs. RBI has issued a Among the new licensees were two universal, 11 small finance, and 10 payment banks. In number of new addition to electronic payments, initiatives to strengthen last mile delivery infrastructure and banking licenses financial literacy are also strongly supporting financial inclusion. While India lags comparator since 2014 aimed at countries in some financial services access indicators (number of branches and ATMs - Figure deepening financial 58 and Figure 59) technological changes are providing new delivery channels (e.g., mobile and inclusion, including branchless banking) that could prove more cost-efficient and help reach out unserved segment through digital of the population. platforms 43 « INDIA DEVELOPMENT UPDATE JUNE 2016 Figure 58. India lags on the number of bank branches Figure 59. … and ATMs Number of commercial bank branches per capita Number of ATMs per capita BRA IND MEX BRA CHN IND MEX RUS TUR ZAF RUS TUR ZAF 50 200 45 180 40 160 35 140 30 120 25 100 20 80 15 60 10 40 5 20 0 0 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Source: Finstats and World Bank staff calculations Source: Finstats and World Bank staff calculations Commercial banks have capital levels in excess of regulatory requirements At present, the Reported capital levels are in excess of regulatory minima, reflecting regulatory changes to bring commercial banking capitalization in line with Basel III requirements. As of September 2015, 71 out of 91 Scheduled sector appears to be Commercial Banks (SCBs) already meet the 11.5 percent Basel III capital requirement ahead of sound 2019 deadline. The capital to risk-weighted assets ratio (CRAR) of SCBs declined to 12.7 percent from 13.0 percent between March and September 2015 although the Tier 1 ratio increased to 6.5 percent during the same period. Liquidity coverage is ample, and compliance with Basel liquidity requirements is high. Indian banks also compare favorably in an alternative measure of financial stability, the z-score28, which compares buffers (capitalization and returns) with risk (volatility of returns) to measure a bank’s solvency risk (Figure 60). These indicators suggest the system as a whole has adequate buffers in case of economic shocks. Figure 60. Banks’ z-scores are above most peer countries Figure 61. Capital levels exceed regulatory requirements… z-score, 2011-2013 average CRAR, percent 40 Public Private Foreign 35 25% 34 35 30 30 20% 25 21 19 19 19 20 16 16 17 15% 15 13 14 13.1% 12.4% 11.5% 10 6 4 5 10% 5 3 0 5% 0% 2011 2013 2015 Source: Global Financial Development Database Source: Finstats and World Bank staff calculations 28The z-score is defined as z ≡ (k+µ)/σ, where k is equity capital as percent of assets, µ is return as percent of assets, and σ is standard deviation of return on assets as a proxy for return volatility. See Čihák et al. (2013) for additional details. INDIA DEVELOPMENT UPDATE JUNE 2016 » 44 However, capitalization levels have been declining, in particular for public sector banks (PSBs), where leverage has also been increasing as loan growth has been financed through a reallocation of assets from investments (Figure 61). Moreover, the full impact of capital charges associated with the phasing out of regulatory forbearance in the classification and provisioning of NPAs (discussed below) has yet to work itself out. A number of challenges need to be addressed to ensure the financial sector can support growth Credit growth to business has slowed, especially from public sector banks Overall credit Growth in bank credit reached a low of 8.8 percent (y/y) in October 2015 from 11.5 percent a growth has been year earlier and a high of 28.0 percent in December 2010 (Figure 62). Credit growth showed decelerating since signs of recovery since October, however, reaching 12.1 percent in February 2016. Declining 2011, partly as a inflation, which reduces firms’ demand for working capital credit, appears to be one of the result of the decline drivers of declining credit growth. As Figure 63 indicates, real credit growth was more stable in inflation when measured in real terms deflated by the GDP deflator or has even increased in the past year if deflated by Wholesale Price Index (WPI). Figure 62. Nominal credit growth has decelerated Figure 63. … but the trend is more stable when considering credit growth in real terms Change from the previous year, percent Change from the previous year, percent 20 20 WPI GDP deflator 18 18 16 16 14 14 12 12 10 10 8 8 6 6 4 4 2 2 0 0 Aug-12 Oct-12 Dec-12 Apr-13 Aug-13 Oct-13 Dec-13 Apr-14 Aug-14 Oct-14 Dec-14 Apr-15 Aug-15 Oct-15 Dec-15 Jun-12 Jun-13 Jun-14 Jun-15 Feb-13 Feb-14 Feb-15 Dec-12 Dec-13 Dec-14 Dec-15 Sep-12 Sep-13 Sep-14 Sep-15 Jun-12 Jun-13 Jun-14 Jun-15 Mar-13 Mar-14 Mar-15 Mar-16 Source: RBI and World Bank staff calculations Source: RBI and World Bank staff calculations Public sector banks Public sector banks include the nationalized banks and the State Bank of India (SBI) and its account for the bulk affiliates (51 percent and 22 percent of outstanding credits as of June 2016). Credit growth of of the deceleration the nationalized banks declined from 14.1 percent (y/y) in September 2013 to 4.6 percent in in credit growth September 2015, before recovering to 5.1 percent in December (Figure 64), while in the SBI and affiliates, credit growth declined from 18.7 percent to 6.8 percent in the same period but recovered strongly to 9.8 percent in December. The trends are starkly different in private banks (22 percent of credits outstanding), where credit growth has accelerated from 13.2 percent in September 2013 to 20.3 percent in September 2015 and 24.9 percent in December 2015. Credit growth in foreign banks (5 percent of credits outstanding) has been more stable, declining from 15.7 percent in September 2013 to 11.0 percent in September 2013 and 11.9 percent in December 2015. 45 « INDIA DEVELOPMENT UPDATE JUNE 2016 Figure 64. The slowdown in credit growth was concentrated on PSBs Change from the previous year, percent 35 30 25 20 15 10 5 0 -5 SBI and its Associates Nationalised Banks -10 Foreign Banks -15 Private Sector Banks All SCBs -20 Source: RBI and World Bank Staff Calculations The deceleration With lower non-performing assets (NPAs) compared to the business sector and declining has been inflation boosting demand for durable goods, banks have been relatively comfortable in lending concentrated in for retail credit and personal loans, which have been consistently expanding above 15 percent corporate credit, as y/y (Figure 65). Housing credit (53 percent of personal loans) has expanded quickly as policy consumer credit has become supportive (fiscal through tax incentives, and regulatory through favorable risk growth, including weights). As of December 2015, total housing loan assets (including non-bank financial for housing, institutions) amounted to INR 11.9 trillion, translating into 18 percent annualized growth. One remained steady of the drivers of this growth has been affordable housing, including in Tier II and III towns. Growth in assets of smaller housing finance corporations (HFCs) corroborate this development. On the other hand, credit to service-sector and industrial businesses has declined, accounting for much of the slowdown in overall credit growth (Figure 66). Credit to industry stabilized below 6 percent, while growth in advances to service-sector businesses picked up only recently, to reach close to 10 percent in early 2016. Within the industrial sector, credit to large companies appears to have been recovering since May 2015, in contrast to medium-size enterprises. Figure 65. The deceleration in credit growth was Figure 66. …as personal loan growth remained buoyant concentrated in loan to industrial businesses… Change from the previous year, percent Change from the previous year, percent Agriculture & Allied Activities Personal loans Agriculture Industry (Micro & Small, Medium and Large) Personal Loans Industry Services (non-financial) Services Total 25 20 18 20 16 14 15 12 10 10 8 6 5 4 2 0 0 Jun-12 Dec-12 Jun-13 Dec-13 Jun-14 Dec-14 Jun-15 Dec-15 Source: Finstats and World Bank staff calculations Source: Finstats and World Bank staff calculations INDIA DEVELOPMENT UPDATE JUNE 2016 » 46 Asset quality of PSBs has deteriorated, weighing on credit growth The large and In parallel with the developments in credit growth, non-performing assets (NPAs) appear increasing particularly concentrated on balance sheets of Public Sector Banks (PSBs). NPAs stood at 6.2 accumulation of percent of total loans in PSBs as compared to 2.2 percent for private banks (Figure 67). Higher non-performing and NPA levels in PSBs reflect a faster growth rate of non-performing assets in recent years: NPAs stressed assets in have increased by an average rate of 40 percent over the last four years in PSBs compared to 35 PSBs hinders their percent for all SCBs. It is worth noting, as can be seen in Figure 68, that the level of NPAs was capacity to provide even higher in the past, coming down from 16 percent in 1998 to just over 2 percent before the fresh loans global financial crisis as the authorities took adequate measures to correct the problem. Figure 67. Non-performing assets have reached 6.2 Figure 68.Taking a long-term view, NPAs were higher in percent, driven by higher NPAs in public sector banks the early 2000s and came down Non-performing assets as a share of gross advances, percent Non-performing assets as a share of gross advances, percent 7 18 PSBs Private Banks Foreign Banks 6.20 6 16 5.10 5 14 4 12 10 3 2.41 8 2 All SCBs Public Sector Banks (76% of total) 6 1 Private & Foreign Banks (24% of total) 4 0 2 0 1998 2000 2002 2004 2006 2008 2010 2012 2014 Source: RBI and World Bank staff calculations Source: RBI and World Bank staff calculations NPAs are likely to Banks have been amending initial terms of stressed loans, availing themselves of initiatives to rise further as there alleviate the burden of corporate distress on the banking sector. Public banks hold the largest has also been a share of restructured advances (Figure 69 and Figure 70). This category includes sub-standard large increase in and doubtful advances. Stressed assets (non-performing and restructured advances) of PSBs restructured loans were 17.7 percent of total loans as of December 2015 compared to 11.0 percent in March 2013 and 4.6 percent for private banks, reflecting a particularly large share of restructured loans in PSBs (7.8 percent). Given the tendency to restructured loans to become nonperforming, it is likely that the asset deterioration trend for public banks will continue. Figure 69. Restructured loans have climbed Figure 70. Sub-standard and doubtful advances climbed Restructured loans as a share of total advances, percent Breakdown of stressed assets of PSBs, INR million 9 8.5 Sub-Standard Advances 1,800 60,000 Foreign Banks Private Sector Banks State owned banks Doubtful Advances 8 7.3 1,600 Loss Advances 7.0 Standard Advances 50,000 7 1,400 6 1,200 40,000 5 1,000 30,000 4 3.6 3.7 800 2.9 3 600 20,000 1.8 400 2 10,000 0.8 200 1 0 0 0 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2008 2009 2010 2011 2012 2013 2014 2015 Source: RBI and World Bank staff calculations Source: RBI and World Bank staff calculations 47 « INDIA DEVELOPMENT UPDATE JUNE 2016 Stressed assets are Stressed assets are concentrated in a relatively small number of mostly large banks, but a few small concentrated in banks also appear to have high rates of stressed assets. Banks with ratios of stressed assets above large banks the SCB average of 11.3 percent are large PSBs including Bank of Baroda (BOB) and Bank of India (BOI; Figure 71). Meanwhile, India’s largest bank (SBI) has a below-average ratio. Most banks – in both number and share of gross advances – also have ratios below average. However, a few smaller banks appear to have stressed ratios of more than 12 percent (Figure 72), and these may require additional attention. Figure 71. Some of the largest PSBs have NPAs Figure 72. Stressed assets are concentrated in larger approaching 10 percent banks Percent Distribution of SCBs based on stressed advances ratio (September 2015) 16 14 0.1 0.3 12 4.0 4.5 10 1.2 8 3.5 6 9.7 9.2 4 2 5.1 0 SBI BOB BOI Gross NPLs Restructured Loans SDR Source: JP Morgan Source: RBI Financial Stability Report 2015 As the level of Provisioning needs are expected to accelerate in the near future if the asset deterioration trends stressed assets continue, and as restructured assets migrate to classification categories that require higher increases, so has the provisioning rates. This trend will be particularly notable as banks phase in the regulatory change necessary provision- of June 2015 whereby restructured loans are no longer exempt from provisioning requirements. ing requirement, The profits of public sector banks have been on a declining trend since 2011 (Figure 73) in line creating a drag on with deteriorating asset quality. As resources will have to be set aside to cover for such losses and banks’ profits and maintain sound capitalization levels, in the absence of fresh capital infusions this will limit lending significantly the lending capacity of the most affected banks, and thereby future credit growth. Figure 73. Until recently, net profits of PSBs had been declining Ratio of net profits to total assets, percent FOREIGN BANKS 2.5 State Owned Banks PRIVATE SECTOR BANKS ALL SCHEDULED COMMERCIAL BANKS 2 1.5 1 0.5 0 2008 2009 2010 2011 2012 2013 2014 2015 Source: RBI and World Bank staff calculations INDIA DEVELOPMENT UPDATE JUNE 2016 » 48 The transmission of policy rates to lending rates has been muted, keeping finance costs up Lending rates The RBI has now cut its policy interest rate by 150 basis points (bps) in the past 16 months, remain sticky, but lending rates remain sticky, dampening the impact of monetary policy on the real economy. dampening the Despite the cumulative 125 bps reduction in the policy rate between January and December impact of monetary 2015, the median base lending rates of banks have fallen by only 63 bps during that period while policy on the real deposit rates have declined by 115 bps (Figure 74). While the gap between the policy rate and economy lending rates widened by 10 bps, the gap between the deposit rate and lending rates widened by as much as 70 bps to 6.1 percent in public sector banks. Figure 74. Lending rates declined less than the policy rate Figure 75. Since the rate cut cycle started, profit levels have been maintained and in some cases increased Interest rate index, September 2014=100 Change, percentage points Interest rate trends Profit ratio 14.00 4 105 0 12.00 3.5 -0.19 -0.2 3 100 10.00 -0.26 2.5 -0.4 8.00 -0.40 1.70 2 95 6.00 1.351.49 1.49 -0.6 1.5 -0.63 4.00 0.59 0.63 1 -0.8 0.41 0.35 90 Policy Rate 2.00 0.5 Deposit Rate -1 0.00 0 Base Rate 85 Mar-2014 Dec-2014 Mar-2015 Jun-2015 Lending rates (Public) -1.2 Profit ratio FB Lending rates (Private) -1.15 Profit ratio PB Lending Rates (Foreign) -1.25 Profit ratio Nationalized 80 -1.4 profit ratio STATE BANK OF INDIA & ITS ASSOCIATES Sep-14 Dec-14 Mar-15 Jun-15 Sep-15 Dec-15 Change STATE BANK OF INDIA AND ITS ASSOCIATES Nationalized (Sept FB '14-Dec PB '15) RBI rate Source: CEIC, RBI, and World Bank staff calculations Source: CEIC and World Bank staff calculations Note: “Lending rates” are the simple average of the upper points in the Note: Interest rates by bank group and profit rate by bank group median ranges of interest rates for term loans, demand loans, and cash credit, at which at least 60 percent of business has been contracted. Some of the causes Two related mechanisms have been suggested to explain why banks have been reluctant to behind the weak lower the deposit rate aggressively – thus capping their willingness to reduce lending rates:29 transmission of First, schemes such as small savings deposits offer rates that are not tied to market rates, and if monetary policy are deposit rates were reduced depositors would withdraw from banks to move to those schemes. linked to banking Second, system liquidity, which is controlled through other monetary policy instruments such sector liquidity as the Liquidity Adjustment Facility (LAF), may be tighter despite lower policy rates. In this conditions case, banks would be similarly unwilling to reduce deposit rates aggressively. To these explanations, one may add that households in India can substitute deposits for gold, especially in a context where inflation expectations have been slower to decline compared to actual inflation. Explanations that focus on deposit rates however cannot account for the widening gap between deposit and lending rates. High levels of NPAs Two channels connect corporate and bank balance sheet stress to the limited transmission of and stress in policy to lending rates. First, the higher rate of default experienced by the banks by itself would corporate balance be associated with higher interest rates that reflect the higher risk of default in the overall loan sheets are likely portfolio. This is in fact part of the vicious cycle that policy easing is supposed to address: higher linked to widening interest costs weigh on the debt servicing capacity of the corporate sector, further increasing gap between deposit risk in the portfolio. 29 See Economic Survey 2016, Box 1.7. 49 « INDIA DEVELOPMENT UPDATE JUNE 2016 and lending rates, which may also Second, banks have been feeling the pressure to generate higher income to provision for an reflect limited increasing share of distressed assets, raising their incentives not to follow the recent reduction competition in the in the reference rates set by the RBI (Figure 75). As discussed below, in a perfectly competitive sector financial system, banks would find it difficult to avoid passing through lower rates to borrowers. However, frictions in the Indian banking system make this a plausible channel. As a result, it appears that pervasive levels of corporate sector distress currently in the economy may be one factor in weakening the monetary policy transmission process. The authorities have In early March, the Ministry of Finance reduced interest rates across various small savings taken successive schemes by 40-130 bps. The reduction in rates was larger for shorter-tenor instruments (~100- measures to address 130 bps). The objective was to close the gap with deposit rates and facilitate transmission. For the shortage of example, the gap between one-year post office time deposits (a small savings scheme) and the liquidity in the one-year bank deposit rate will narrow to 35 bps from 95 bps between April and March 2016. system In addition, in its most recent monetary policy meeting, RBI not only cut rates but also took a series of measures to enhance liquidity in the system. The RBI announced it was moving system liquidity from a structural deficit to neutral over a one-year period; narrowed the rate corridor; and reduced the minimum daily cash reserve ratio balance, which along with earlier reductions in the Statutory Liquidity Ratio (SLR) provide greater flexibility for banks to manage their funds. Addressing the challenges will require both short- and long-run interventions India’s financial Several aspects of India’s financial sector are performing well, especially when seen as a system sector appears to that is in the midst of a long-term structural transformation. Many apparent shortcomings are perform well on in fact transitional, and an underlying reform process is already underway in the direction of many dimensions making the sector more market-oriented. Moreover, there are few if any indications that the system is near crisis. Credit is stable in real terms, NPAs have been higher in the recent past without a banking crisis, the system is well-capitalized, RBI has demonstrated strong capacity to regulate the sector and ensure its soundness, and the fiscal position of the government – the ultimate backstop for the system – has strengthened. However, structural Addressing structural, long-term reforms as well as the immediate issue of rising stress in reforms and corporate and bank balance sheets are both imperative and urgent for three reasons. First, addressing the NPA action today is needed precisely to revert trends that, if left unchecked, could eventually lead to challenge remain a crisis; this is well recognized by the Indian authorities. Second, the financial sector will only urgent tasks be in a position to expand their balance sheets to support economic growth in a sustainable manner when their capital position has been restored; similarly, the demand of credit by the private sector will only pick up after corporate balance sheets are deleveraged. Third, enhancing the efficiency of India’s financial intermediaries will be critical for long-term productivity gains. The rest of this chapter is divided into two main parts. First, it places the analysis in the context of the ongoing structural transformation of the Indian financial sector and identifies areas where reforms could be accelerated. Second, it turns to the immediate problem of resolving stressed corporate and banking balance sheets, highlighting the need to address the underlying (real) causes of financial stress and governance and regulatory reforms that will provide the foundation for a stable and robust sector. INDIA DEVELOPMENT UPDATE JUNE 2016 » 50 ii. A transformation toward a market-oriented sector is underway and can be accelerated The Indian financial A number of current features of the financial sector reflect a long-term process of sector is in the midst transformation toward market-oriented institutions, instruments, and practices. First, the of a long-term system has been transitioning from one that was entirely state owned in the early 1990s toward reform process that one with greater participation of private banks and generally more competition. Second, is transforming it government mandates aimed at ensuring stable financing for key sectors of the economy (the into a modern, government itself, through the SLR requirement; and priority sectors, through priority sector market-oriented lending requirements) have been gradually relaxed. Third, non-banking financial institutions sector and capital markets are nascent, but expanding. Finally, remuneration of savings instruments, which have traditionally included gold and small savings certificates, increasingly reflect market rates, facilitating financial deepening, credit intermediation, and monetary policy transmission. Accelerating these transitions – for example by adopting more commercially oriented practices in PSBs – is likely to increase competition and efficiency. The participation of foreign and private banks has gradually increased, boosting competition Private banks have At the beginning of the 1990s, the banking sector was controlled by the public sector. Since been growing, but then, private and foreign banks have grown and now make up about 28 percent of banking significant public sector assets (Figure 76). These changes were the result of successive reforms aimed at sector holdings increasing efficiency, competition, and delivery of deeper penetration of financial services. remain a central Entry of private and foreign banks was facilitated, and between 1994 and 2000, seven new feature of the private banks entered the market. While the ownership of foreign banks is still limited to 74 financial sector percent of capital, since 2014 they have been allowed to incorporate domestically as wholly owned subsidiaries. With respect to PSBs, the government has floated 49 percent of the equity it held in many PSBs, but retained control of their executive boards and of the appointment of management officials. Figure 76. Private banks have increased their presence Figure 77. Top 10 banks control about 60% of the market Share of total credit, percent Percent 100 7.0 9.8 90 6.2 15.3 20.0 SBI Group 7.5 22.5 23.4 5% 80 8.4 4.9 8% Top 5 6.6 4.9 22% 70 Nationalized 60 Other PSB 50 14% 86.8 82.6 Top 3 Private 40 76.3 70.9 75.1 71.6 30 Other Private 20 22% 10 Foreign Banks 0 29% 1990 1995 2000 2005 2010 2015 PSBs Foreign Others (inc. private) Source: RBI and World Bank staff calculations Source: RBI and World Bank staff calculations Note: Shares based on gross advances of SCBs Despite the large The commercial banking sector currently comprises 214 scheduled commercial banks (SCBs), number of banks, which are banks with a full banking license, and 4 non-scheduled banks.30 The group of public assets are sector banks (PSBs) includes one large entity – the State Bank of India (SBI) – and its associates, concentrated in few which together control over 20 percent of banking system assets, and a number of small entities, institutions five of which are relatively large, accounting together for about a quarter of assets of the banking 30 As of February 16, 2016. 51 « INDIA DEVELOPMENT UPDATE JUNE 2016 system. Concentration is also high within the group of private and foreign banks, with three private banks accounting for two-thirds of total private bank assets and the top three foreign banks accounting for more than half of foreign bank assets. Figure 77 illustrates the structure of the sector using gross advances as a proxy for assets. The large market The large share of the market controlled by public banks, and the high asset concentration share of PSBs and within the other bank groups, have been identified as features that undermine competition in asset concentration the banking sector. Prior research on public sector banks (Viral et al., 2010) has shown that among other SCBs these institutions operate with implied or explicit sovereign guarantees, giving them a limit competition competitive edge over private counterparts. In India, the sector has historically displayed very low interest dispersion across banking entities (Figure 78 and Figure 79). This is suggestive of price dynamics being led by the large PSBs acting as “market leaders” and price setters, with the smaller competitors following suit. On the other hand, among banking groups, public banks acted counter-cyclically at the inception of the global financial crisis, maintaining sustained rates of credit growth. Figure 78. Interest rates show relatively low dispersion especially among the market-dominant PSBs Interest rates of scheduled commercial banks on term loans, average of maximum and minimum rates, percent 20 PSBs 20 Private Banks 20 Foreign Banks 18 18 18 16 16 16 14 14 14 12 12 12 10 10 10 8 8 8 6 6 6 Median Median Median 6-mo T-bills 4 4 4 Source: RBI and World Bank staff calculations Figure 79. A longer-term view of lending rates against the T-bill rate Interest rates, percent 14 12 10 8 6 4 2 t-bill Pub For Pvt 0 2008 2009 2010 2011 2012 2013 2014 Source: SECC and World Bank staff calculations INDIA DEVELOPMENT UPDATE JUNE 2016 » 52 Limited competition India has one of the highest interest spreads among a group of comparator countries,31 only may be among the followed by Russia, which had to raise interest rates on account of the financial turmoil in 2014 factors behind high (Figure 80). Partly reflecting such high rates, India’s affordability of financial services is relatively spreads between below that of its comparators. According to the Global Competitiveness Report for 2014–2015, lending and deposit India ranks 73rd in affordability of financial services, above only Mexico among its comparators rates in India and 83rd in availability of financial services behind all of its peers but following closely China, Mexico, and Russia (Figure 81). The relatively large gap between lending and deposit rates are suggestive of pricing power of banks, which as noted earlier may also be linked to the limited transmission of policy rate cuts to lending rates. Figure 80. Interest rate spreads in India are high Figure 81. Affordability and availability of financial compared to other emerging economies services lag peers Spread between lending and deposit rates, percentage points LHS: Rankings and RHS:index (0-7; 7=best) 8.0 CHN IND MEX RUS ZAF 7.0 0 7.0 Affordability Availability 6 19 7 21 37 35 6.0 6.0 33 43 42 41 46 36 50 63 5.0 5.0 73 73 83 69 71 50 106 4.0 100 4.0 3.0 150 2.0 3.0 1.0 2.0 200 0.0 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Source: WDI and World Bank staff calculations Source: Global Competitiveness Report Government mandates are gradually being phased out, which may help banks’ profitability The Statutory Liquidity Ratio has been progressively lowered A large share of SCBs are mandated to maintain a specified percentage of their total net demand and time bank assets was liabilities (DTLs) in the form of liquid assets, which include cash, gold, and approved securities absorbed by (including government securities). This Statutory Liquidity Ratio (SLR) is set by the Reserve government Bank of India, which has reduced it from 38.5 percent in 1980’s to 25 percent in 1990s; and securities, partly for more recently from 23 percent to 21.5 percent in 2015 (Figure 82). The SLR remains relatively regulatory reasons high as there is no comparable requirement in most countries beyond those for prudential reasons and which are generally much lower. Banks have been providing for government financing needs even beyond the SLR, absorbing additional liquid resources of the banking sector. About 20 percent of bank assets are absorbed by investments in government securities only, which is in addition to other cash and short-term assets they have. In total, about 25 percent of bank assets are absorbed by investments in government securities (Figure 83). This share of credit to government is high in comparison with peer countries, lower only compared to Brazil where interest rates on government securities are much higher (Figure 84). 31 According to loan and deposit data published by Finstats and the RBI. 53 « INDIA DEVELOPMENT UPDATE JUNE 2016 Figure 82. The SLR has been progressively reduced but Figure 83. About 25 percent of bank assets are absorbed remains relatively high by investments in government securities Statutory Liquidity Ratio, percent Cash and government securities, percent of total assets 40 FOREIGN BANKS NATIONALISED BANKS 38 45 PRIVATE SECTOR BANKS 36 40 STATE BANK OF INDIA & ITS ASSOCIATES ALL SCHEDULED COMMERCIAL BANKS 34 35 32 30 30 25 28 20 26 15 24 10 22 5 20 0 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2007 2008 2009 2010 2011 2012 2013 2014 2015 Source: RBI Source: RBI and World Bank staff calculations High concentration The relatively low interest rates in government securities would tend to limit profitability of the of assets in banking sector. Real interest rates on government securities have until recently been close to government zero (Figure 85). Low interest rates in turn are partly driven by the high statutory requirements securities hurt bank such as the SLR, as well as other institutional arrangements (requirements on insurance profitability companies and the employee provident fund, for example) that contribute toward de facto regulated interest rates on government borrowings. 32 Figure 84. Credit to government and SOEs is above most Figure 85. In 2012 and 2013, real interest rates on peers government securities had been negative Credit to Government and SOEs, percent of total credit Real interest rates in 6-month treasury bills, 3mma BRA CHN IND MEX 45 5 RUS TUR ZAF 40 4 35 3 30 2 25 1 20 0 15 -1 10 -2 5 -3 Jul-12 Jul-13 Jul-14 Jul-15 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16 Apr-12 Oct-12 Apr-13 Oct-13 Apr-14 Oct-14 Apr-15 Oct-15 0 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Source: WDI and World Bank staff calculations Source: WDI and World Bank staff calculations 32 The SLR also helps explain high interest spreads, as it keeps government borrowing costs lower but results in higher costs for the private sector. INDIA DEVELOPMENT UPDATE JUNE 2016 » 54 Priority Sector lending requirements can be relaxed as financial markets deepen to serve all markets The composition of In particular, there are specific credit targets for priority sectors, which are relatively high at 40 banks’ non- percent of the aggregate bank advances. The definition of priority sector has evolved over time government lending and currently includes small loans to small and marginal farmers for agriculture and allied portfolio is also activities, loans to micro and small enterprises, loans for small housing projects, education loans subject to and other small loans to low income households, all with the objective of enhancing financial mandatory inclusion in the country. Within these requirements, there are also sub-targets designed for requirements agriculture and other sectors, although the RBI has recently sought to simplify the structure. Subsidiaries of foreign banks and foreign banks with 20 or more branches in the country are subject to the same requirements while smaller foreign banks have the floors reduced. Regulatory requirements on minimum loan allocation to priority sectors have caused the loan book composition to change only slightly through the years with PSBs responding more to supporting developmental objectives and the infrastructure industries, while private banks have been concentrating more on loans for housing and consumption. The RBI has tried to However, asset allocation requirements on the composition of lending portfolios have a facilitate bank tendency to maintain high operating costs and reduce the risk-adjusted returns of loan lending to these portfolios, in particular because the origination process, monitoring, and risk-management of sectors by ensuring small and micro loans are particularly costly activities. Foreign banks on their part are expected adequate pricing of to bring in innovation and develop new models through their global experience to deliver credit credit to underserved sectors of the economy in a cost effective manner. Relaxing government mandates may be positive for profitability Profitability of the Despite the high interest spreads, the profitability of the banking sector, as measured by return Indian banking on assets (ROA) and return on equity (ROE), has remained below the average of its peers sector has been (Figure 86 and Figure 87). While inefficiencies and high structural costs have often been lower than its peers identified as causes of low profitability, the relatively high share of investments in government securities and high costs of priority sector lending also should be added as a source of low returns, along with, more recently, the need to increase provisions to cover NPAs. Accordingly, diversifying government funding sources and developing instruments and institutions to serve priority sectors can allow for a faster phasing out of government mandates, which may help bank profitability and efficiency. Figure 86. Government mandates one factor constraining Figure 87. …along with increasing NPAs in Public Sector profitability despite wide interest spreads… Banks Return on assets, percent Return on equity, percent BRA CHN IND MEX BRA CHN IND MEX RUS TUR ZAF RUS TUR ZAF 4.0 35 3.5 30 3.0 25 2.5 20 2.0 15 1.5 1.0 10 0.5 5 0.0 0 2005 2007 2009 2011 2013 2015 2005 2007 2009 2011 2013 2015 Source: Finstats Source: Finstats 55 « INDIA DEVELOPMENT UPDATE JUNE 2016 Non-banking institutions and capital markets have started to develop Non-banking intermediaries have expanded Non-bank financial Overall, financial sector assets have grown to account for about 140 percent of GDP, which is intermediaries have high in comparison with regional and income peers. Commercial banks have been playing the been expanding largest role, accounting for 94 percent of GDP or two-thirds of financial sector assets in India rapidly (Figure 88). In addition, other non-bank financial intermediaries such as insurance companies, pensions, mutual funds, cooperative banks, and other financial institutions now represent a sizeable share of the sector. The share of commercial banks in the financial sector is not out of line with other large emerging markets, but lags more advanced economies (Figure 89). Figure 88. Commercial banks dominate the financial Figure 89. The share of commercial banks is in line with sector peers but higher than in developed countries Share of commercial banks in total financial sector assets Percent of financial sector assets Cooperative Other 100 Banks 4% 88.3 89.5 90 79.5 6% Mutual Funds 80 69.7 67.0 6% 70 60 Pensions 43.9 5% 50 37.4 40 26.0 30 Insurance 20 companies Commercial 12% Banks 10 67% 0 Source: RBI Source: IMF FSAP documents for each country, except India (RBI) and Turkey (Central Bank of Turkey) Other financial Small banks with limited areas of operation, in particular credit cooperatives, were created to intermediaries have cater to the credit, processing and marketing needs of small and marginal farmers organized on been expanding and cooperative lines. Cooperatives expanded also in urban and semi-urban areas in the form of contributing to urban cooperative banks. Regional rural banks were created to bring together the positive credit growth and features of credit cooperatives and commercial banks, and specifically address credit needs of financial inclusion backward sections in rural areas. Further, there was an experiment of establishing local area banks, albeit on a smaller scale, to bridge the gap in credit availability and strengthen the institutional credit framework in rural and semi-urban areas. It will be important to insure that these non-bank intermediaries are also adequately regulated to avoid regulatory arbitrage with the commercial banking sector. Bond and equity markets are becoming increasingly important sources of financing Bond and equity Financing from capital markets has also grown. Stock market capitalization has reached over markets have grown, 57 percent of GDP (Figure 90). Debt markets reached about 33 percent of GDP in 2011, but but remain small government securities – issuances from India’s union and state governments – account for 65 percent of debt outstanding, while certificates of deposit issued by banks and commercial paper account for an additional 20 percent. The corporate bond market (INR 18.7 trillion outstanding as of September 2015) represents about 15 percent of debt outstanding or 5 percent of GDP, compared to 25 percent of GDP in China and nearly 90 percent in the United States (Figure 91). Bond markets have been successful especially for highly rated private placements. Public sector corporates represent 45 percent of total bonds issued, followed by banks (24 percent). INDIA DEVELOPMENT UPDATE JUNE 2016 » 56 Figure 90. India’s stock market capitalization is higher Figure 91. But bond markets are much smaller than in than China’s… China and advanced economies Stock market capitalization, percent of GDP Size of outstanding debt securities of financial and corporate issuers, percent of GDP BRA CHN IND MEX US China Germany India Italy Japan 105 RUS TUR ZAF 140 South Korea UK 95 120 85 100 75 80 65 55 60 45 40 35 20 25 0 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2005 2006 2007 2008 2009 2010 2011 Source: Finstats Source: Acharya et al. (2011) Developing the local bond market benefits corporates but also banks, allowing them to (i) better manage foreign currency risk and asset-liability mismatches, (ii) make loans at longer tenors; and (iii) face lower systemic risks as the availability of long-term local currency funding contributes to the overall stability of the financial sector. There are strict At present, qualified foreign portfolio investors (FPIs) can invest up to INR 2 trillion (about limits on foreign USD 30 billion) in government debt securities and USD 51 billion in bonds issued by Indian holdings of local companies. Other regulations preclude foreign investors from investing in corporate or bonds, limiting government security with residual maturity of less than three years. The government has been market size opening the market in recent years, allowing FPIs to invest in state development loans since October 2015, and providing a roadmap for higher limits on FPI participation in government debt markets. By March 2018 the quota for FPI investment in government securities is targeted to reach 5 percent of the total amount outstanding (up from 3.8 percent as of end 2015).33 Limits on foreign holdings, as well as the transaction costs of accessing the bond markets, constrain the share of foreign portfolio investment that may be intermediated through capital markets, thereby limiting the overall growth in the sector. Countries across the world, both advanced and emerging, seldom use quantitative limits, and foreign participation on local currency bond markets in many emerging countries has grown rapidly since the 2000s. Non- resident bond holdings in nine emerging markets have increased seven-fold between 2002 and 2006 (Peiris, 2010) and expanded even faster since then (Figure 92). While concerns about capital flow volatility are legitimate, further opening the market for FPIs increases demand for domestic assets, especially Government securities, and may help keep yields low even as domestic mandates on holding government securities are phased out. 33 Effective June 12, 2013, FIIs, QFIs, and long-term investors registered with the SEBI may invest in government securities up to USD 30 billion, after an earlier increase in January, 2013, to USD 25 billion from USD 20 billion earlier. The requirement that the government securities have a residual maturity of three years at the time of the purchase was also eliminated in January 2013. To simplify the investment limits for FIIs and long- term investors, the government debt limit merged the two sub-limits under government securities to (a) USD 10 billion for investment by FIIs in government securities, including treasury bills and (b) USD 15 billion for investment in dated government securities by FIIs and long-term investors. 57 « INDIA DEVELOPMENT UPDATE JUNE 2016 Figure 92. Foreign participation in Indian bond markets is small compared to most other economies Foreign holdings of local currency government debt securities, percent of total 45% China India Indonesia Brazil Mexico Russia 40% South Africa Turkey 35% 30% 25% 20% 15% 10% 5% 0% 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Source: Arslanalp and Tsuda (2014, 2015) Returns on savings instruments are increasingly market-determined Gold certificates have been introduced as an alternative to physical gold Financial As announced in the FY16 Union Budget, the government launched a Gold Monetization instruments Scheme (GMS) and Sovereign Gold Bonds (SGB) during the course of the year with the primary introduced by objectives of (i) integrating private savings in physical gold with financial markets; and (ii) authorities to reducing the reliance on gold imports. The RBI issues SGBs on behalf of government of India, channel gold in lieu of money and are linked to the international price of gold. The primary objective of demand into SGBs is to “monetize” the stock of privately owned physical gold, and convert the investment financial markets demand for physical gold into paper demand. According to the estimates of one investment bank, if the scheme is fully subscribed it will represent 27 percent of the 2014 investment demand.34 The government of India raised approximately INR 10 billon from gold bonds in FY16 and budgeted to raise another INR 100 billion during FY17. Simultaneously, the GMS allows private owners of gold to set up metal accounts with banks and earn interest (determined by banks) on their gold deposits. Banks can in turn use these deposits to lend to jewelers with a profit margin, qualify for the SLR/CRR mandates, and be exchanged for foreign currency or commodity exchange. The GMS, in addition to introducing gold holdings in circulation, will also help reduce import demand for gold by jewelers. Small savings certificate rates have been better aligned with market rates Interest rates on The central government introduced the national small savings scheme in 1999 with an objective small savings to provide social security and encourage household savings. Net small savings collections stood instruments have at INR 210 billion as at the end of November 2015, a 700 percent increase over the collections been deregulated, a year ago. Interest rates on the short-tenor instruments offered under this scheme were helping monetary deregulated in February by eliminating the previous 25 bps spread over government securities policy transmission of comparable maturities. In addition, the revision of interest rates on all small savings securities will be done more frequently on a quarterly basis, compared to annual earlier. While the spreads on longer-tenor securities were left unchanged (25–100 basis points above government securities of comparable maturities), more frequent revisions will allow closer alignment with market linked rates. The most recent revision took place in April, 2016, and 34 Nomura estimates. INDIA DEVELOPMENT UPDATE JUNE 2016 » 58 current interest on small savings securities ranges from 7.1–8.6 percent, down from 8.4–9.3 percent before the revision. These steps to integrate small savings instruments with market rates will allow better monetary policy transmission, as banks were reportedly unwilling to reduce deposit rates aggressively out of fear of having depositors channel savings to small savings schemes. iii. Corporate and bank balance sheets need to be repaired and strengthened Bank and corporate balance sheets are intertwined Corporate distress is A number of recent studies (RBI, 2015; Singh and Brar, 2016; Economic Survey 2016–17) point behind the observed to the crucial role of the deterioration in corporate balance sheets, especially in the infrastructure trends for bank sectors, in the trend of deteriorating asset quality in PSBs. For example, in a recent speech NPAs (April, 2016), Shri S. S. Mundra, Deputy Governor of RBI identifies among the four main causes of rising distress in PSBs “corporate imprudence” (over-leverage, and excessive growth and profit-seeking) and “corporate misdemeanors.” Corporate debt to equity ratios have been rising significantly in the last decade, as large infrastructure projects were increasingly funded through the banking sector, often as part of public-private partnerships (PPPs). Recently, debt-servicing capacity (interest coverage ratio) has been falling, particularly for large companies, as profitability has declined, a trend consistent with the observed increase in NPAs by those companies. Also consistent with the observed trends in NPA, corporate leverage has been higher for infrastructure-related industries, including construction, gas, water, electricity, and in the metal related-sectors (IMF, 2016). Financing of large Banks provided support to infrastructure development in various forms, both directly and by infrastructure investing in bonds issued by infrastructure companies. The government facilitated this support, projects relied including through regulatory forbearance to reduce the financing costs. As a result, total credit heavily on the to infrastructure has increased from 2 percent of total lending in 2006 to 14 percent in 2012, banking sector and has remained stable since (Figure 93). Within infrastructure, the power sector witnessed the fastest growth (Figure 94). Related to these developments, large corporates have been absorbing the largest share of credit, with their weight in total lending growing from 70 percent of all corporate non-food lending in 2007 to 82 percent in 2015. Figure 93. Infrastructure lending surged… Figure 94. …with large contributions from the power and telecom sectors Infrastructure lending as a percent of total lending Decomposition of year-on-year credit growth by sector, percentage points 16% 70 Power Roads 14% 14%14%14% 14% 13% 60 Other Infrastructure Infrastructure 12% 12% 50 10% 10% 13 9% 40 8% 8% 8% 7% 6% 30 10 6% 7 15 4 5% 5% 4 20 37 7 4% 6 5 4 3 2% 24 23 4 10 22 21 18 15 1 2% 12 14 12 13 10 8 0 0% -10 Source: RBI and World Bank staff calculations Source: RBI and World Bank staff calculations 59 « INDIA DEVELOPMENT UPDATE JUNE 2016 Infrastructure loans Within the industrial sector, mining, iron and steel, infrastructure and aviation, and textiles – especially in power together represented more than half of total stressed assets. The sector with the highest level and transport sectors of distress is aviation, but stressed assets in infrastructure have grown disproportionately. Loans – comprise a to the infrastructure sector account for 16 percent of total advances and 13 percent of NPAs, significant share of but 46 percent of restructured assets (as of end-December 2015; Figure 95). stressed assets… To some extent, distress in these sectors is linked to delays in the approval of projects, hurdles in land acquisition, and supply disruptions that justify the initial emphasis that was placed on loan restructuring attempts to better tune the project cash flows with the loan repayment schedules. However, in a number of cases, the project design has subsequently displayed significant flaws or, according to Shri S.S. Mundra of RBI, “(s)ome of the promoters have diverted borrowed funds for purposes other than for which the finance was availed.” … and are linked to a Related to growing NPAs in infrastructure sectors, credit growth slowed across most industrial slowdown in credit to sub-sectors since 2011, but particularly for infrastructure (Figure 96). The leveraged power those sectors sector was the largest contributor to the slowdown, especially in recent years, whereas the earlier slowdown was primarily driven by telecommunications and other infrastructure. Credit to the roads sector has also decelerated significantly, especially in 2015. Figure 95. Infrastructure loans are not a large share in Figure 96. As NPAs mounted, infrastructure was a key total advances, but a large share of those are distressed contributor to the credit growth decline Shares, percent Decomposition of industry credit growth by sub-sector, percentage points Construction & Infra (ex-power 50 Power Transport Telecom Other Power 27.5 Basic Metals 45 Textiles 40 22.5 Energy & Chemicals Food Processing 14.6 35 Others 17.5 30 25 12.5 20 15 7.5 29.3 10 3.2 3.8 2.5 5 9.2 5.0 0 -2.5 Sector's share in total Share of sector in Sector's share in advances Total NPAs - restructured standard aggregate level advances Source: RBI and World Bank staff calculations Source: RBI and World Bank staff calculations Headline figures for There are three primary reasons for this: (i) Banks are able to “evergreen” loans without limit NPAs do not tell the for a fixed time (2 years) after a project enters commercial operations because of regulatory full story of stress in forbearance that RBI has allowed until recently; this avoids classifying the loan as NPA. (ii) the portfolio of Banks have been able to shift some liabilities to Asset Reconstruction Companies (ARCs) that corporate debt have provided off-balance sheet repositories for poorly performing loans. (iii) Many projects especially of (particularly those let during the “boom years” have yet to reach completion when demand risk infrastructure will crystalize and repayment would begin. (iv) Finally, exposures to doubtful loans in different sectors infrastructure sectors are correlated, but so are asset values within these sectors (Figure 97). INDIA DEVELOPMENT UPDATE JUNE 2016 » 60 Figure 97. The exposure of SCBs to the power and roads sectors is large and correlated Y axis: Risk weighted exposure to roads (or roads and ports) as % of Basel III Tier 1 + Tier 2 capital; X axis: Risk weighted energy/power exposure as % of Basel III Tier 1 + Tier 2 capital 50% 45% 40% IDBI Bank Ltd.† United Bank of India† 35% Andhra Bank‡ 30% Indian Overseas Dena Bank Bank Oriental Bank of Canara Bank 25% Commerce† Punjab National Punjab and Sind Bank† Corporation Bank Bank 20% Bank of Mahar. Indian Bank† UCO Bank 15% Central Bank of India 10% Syndicate Bank SBI (all affiliates)† 5% Bank of Baroda 0% 0% 50% 100% 150% 200% 250% Sources: June 2015 Basel III filings for selected banks or analyst presentations where Basel III filings lack broken-out data Notes: All banks shown have more than 50 percent GOI shareholding Bubble sizes correspond to asset sizes. A sustainable Resolving the NPA issue will require therefore a two-pronged approach: on the one hand, solution to the NPA injecting additional capital will allow banks to expand lending while remaining within prudent problem requires capitalization ratios. On the other hand, without addressing the issue of corporate balance sheet additional capital stress the NPA problem is likely to recur. In addition, infrastructure companies will remain and fixing underly- unwilling or unable to borrow until legacy problems have been adequately resolved, dampening ing corporate stress investment growth. This section considers approaches for injecting capital. Governance reforms need to accompany fresh equity for banks PSBs will need fresh New capital will be required in PSBs because: (i) the income generated has not been sufficient capital injections to cover additional provisioning needs, and there has been both an erosion of capitalization levels; (ii) NPAs deplete banks’ liquid resources, and hence their lending capacity to support the activities of the corporate sector; (iii) banks will be forced to increase provisioning for restructured assets as regulatory forbearance is being phased out; and (iv) additional capitalization requirement due under the Basel III framework, which come fully into place by 2019, are being phased in (Figure 98). Recognizing the need to further boost capitalization levels in banks, the union government allocated INR 250 billion (USD 3.8 billion) to bank recapitalization in its FY17 budget as part of a USD 13.8 billion multiyear plan to recapitalize the sector. This amount may need to be supplemented, however. The ratings agency Moody’s has estimated that public sector banks need between USD 26 and USD 37 billion in new capital by 2019 to comply with RBI’s Basel III targets, assuming their assets remain sound and GDP growth is moderate. The recognition of Recently the RBI has taken significant steps to ensure that losses expected from distressed debt expected losses is are recognized on the balance sheets of banks participating in Corporate Debt Restructuring essential to gauge (CDR) and Joint Lenders Fora (JLFs) so that bank managers and shareholders will be able to the extent of fresh evaluate new capital needs. In particular, as of April 2015, the RBI has requested that distressed resources needed loans that have been restructured in the context of the CDR are no longer exempt from being classified as NPAs. They will have to be reported and provided for as pertaining to the category of loans to which they belonged when the restructuring proceedings initiated. RBI has given a compliance period for banks to make provisions for restructured loans, to ensure they will be setting aside sufficient resources to shoulder losses in those cases when the debt restructuring agreement has not succeeded in restoring financial health to the debtor. This initiative was particularly necessary because the provisions set aside by banks to reserve for potential losses 61 « INDIA DEVELOPMENT UPDATE JUNE 2016 from new NPAs and restructured debt have represented a particularly low share of the outstanding balances of distressed assets, especially for PSBs (Figure 99). As noted above, when losses materialized and provisioning has been insufficient, capital levels are quickly eroded. Figure 98. Some banks require additional capital to meet Figure 99. PSBs have made lower provisions compared to Basel III requirements private and foreign banks Percent Provisions, in percent of new NPAs and debt restructurings 20.0 60 Private banks Public Sector Private Sector Banks 18.0 50 Foreign Banks 16.0 PSBs 14.0 40 12.0 10.0 30 8.0 20 6.0 4.0 10 2.0 0.0 0 Tier 1 Tier 2 2008 2009 2010 2011 2012 2013 2014 2015 Source: RBI and World Bank staff calculations Source: RBI and World Bank staff calculations There is a need to RBI has been supporting the development of market practices to enable banks manage balance create tools for sheet stress. Considering the limits on the availability of recapitalization funds and the variety handling stress in of sectors, locations and specific situations of stressed assets, a large ‘arsenal’ of tools will be bank balance sheets, required to sustainably deal with the NPA problem.35 RBI has been supporting the development while guarding of market practices along two main types of tools for managing balance sheet stress: debt against moral restructuring frameworks and distressed asset sales. hazard. Strengthening governance and regulations Governance of PSBs When shareholders Using the opportunity of recapitalization to improve governance will ensure that the banks’ are asked to provide management teams and the organs of control are committed to restoring profitability and fresh resources to financial soundness through strong, commercially oriented business strategies. Before providing financial institutions the resources needed to support the continuation of the banks’ business models it is customary following losses for shareholders to question the role of the banks’ own management in the emergence of the generated by assets losses. There may be some extraordinary cases, in which the losses emerge because of gone bad, it is unforeseen circumstances outside management control. However, in most situations, the banks’ critical to review and own loan origination procedures, internal control practices, and risk management processes reform the banks’ contribute to the losses. Such procedures should be reviewed in the context of a general governance governance review of the banks. It is customary for shareholders to seek assurances that structures measures be taken regarding banks’ management and internal control models to ensure that new losses will not materialize as a result of old practices. The governance structure of banks is the way in which boards and senior management meet the obligation of accountability to their shareholders, which, in turn, affects how they set corporate objectives and operate the banks’ business on a day-to-day basis. The governance structure should be designed to ensure that banks autonomously implement business strategies 35 “You go to war with weapons, not bare hands: SBI’s Arundhati Bhattacharya”, Live Mint, June 13, 2016. INDIA DEVELOPMENT UPDATE JUNE 2016 » 62 that are directed at achieving high risk-adjusted returns while avoiding undue interference from non-commercially oriented objectives coming from the outside, as well as past mistakes. The Basel Principles for Corporate Governance provide the general framework to guide the design of the governance structures, and suggest how the management, control, audit, and risk management functions within the institution should interact to ensure banks will operate in a safe and sound manner and in compliance with applicable laws and regulations and thereby protect the interests of depositors. Recommendations In India, to guide potential governance reforms for banks receiving new capital, the RBI in of the Nayak report 2014 commissioned a review of “The Governance of Boards of Banks in India,” known as the commissioned by P. J. Nayak report after its chairman. The recommendations of the Nayak report reflect some RBI in 2014 to look best international practices on the structure of governance for banks. In particular, the report at banks’ board makes specific recommendations on possible changes to the structure of the government’s governance reflect holdings in public banks to enhance their autonomy in the pursuit of commercially oriented some of the best objectives, and therefore strengthen the protection of depositors and shareholders. Box 3 shows international some examples from international experience of state control in banks. practices Following up on the report, the GOI passed some important governance reforms, the so-called “Indradhanush” (rainbow) initiative in the context of providing for new capital resources for banks. To strengthen the governance of PSBs, “Indradhanush” introduces key changes in the process of appointment of managing director and non-executive chair, encourages the board to have increased autonomy in business decisions as well as strong and diverse composition with a mandate to improve governance, and introduces a framework of accountability for banks based on key performance indicators (KPI). In April 2016 the Government has set up a Bank Boards Bureau (BBB), as a precursor to a Bank Investment Company, to drive balance sheet improvement and consolidation in the state owned banking sector (see Box 4 for greater details). The BBB will evolve into an investment holding company for PSBs, separating ownership from management, helping address moral hazard in lending, bringing in professionalism in Boards and improving shareholder accountability by management. This will be critical to build institutional capacity in PSBs for risk management and prudent credit underwriting to contain the likelihood of future stresses. Box 3: State Control of Banks – Some International Examples 1. Government-as-Investor: (a). The Singapore Government controls DBS through Temasek. The board of DBS is fully empowered. For example, before the present CEO of DBS was appointed, a search committee was appointed and coordinated by the board of the bank. Neither the Singapore Government nor Temasek had a role in this. (b) The Malaysian Government controls CIMB and Maybank, the country’s largest banks, through Khazanah Nasional, an autonomous investment arm analogous to Temasek. (c). The UK Government controls RBS and Lloyds Bank through the UK Financial Investments Ltd (UKFI). UKFI was set up in the context of the government bail-out extended to these two troubled banks after the 2008 financial crisis, and the UK Treasury is its sole shareholder. The boards of both banks are subject to standard corporate law directors' duties, and fully empowered. UKFI acts as an informed shareholder. (d). The Government of Belgium controls Fortis and Dexia, two troubled banks, through SPFI-FPIM, a holding company. The latter pursues a proactive investment policy with a view to maximizing long-term financial returns. Here too, the boards of the two banks are fully empowered to govern and manage the banks. In each case, the role of Government-as-Investor is discharged through the construct of an intermediate investment company, seen as beneficial for both profitable and stressed banks . UKFI has signed a (revised) shareholder relationship agreement in 2010 with the UK Government containing three ‘overarching objectives’: maximizing sustainable value for the taxpayer, taking account of risk; maintaining financial stability; and promoting competition. UKFI’s mandate should therefore be seen as maximizing medium-term taxpayer returns adjusted for risk, subject to the maintenance of financial stability and competition. Its mandate also includes eventually disinvesting from the two banks. 63 « INDIA DEVELOPMENT UPDATE JUNE 2016 The investment or holding company for the ownership of government banks is widespread, including in China. Prominent exceptions, in the form of direct holdings by governments, are to be found in Germany, Indonesia, Brazil and India. 2. Government-as-Owner: (a). The Government of Brazil controls Banco do Brasil. It owns 59 per cent of the bank, held directly and without an intermediate holding company. The CEO is appointed by the Brazilian President, which diminishes the bank board's powers. Candidates applying to work in the bank need to pass a competitive examination similar to what is applicable to Government officers, further disempowering the bank's board in the manner in which it sets recruitment policy. (b). The Government of Brazil also controls Caixa, which the Government fully owns and in which it appoints all directors. Caixa is a special purpose bank serving as a tool for public investment and expansion of public access to financial services. As the developmental role is confined to wider objectives within financial services, the role is closer to Government-as-Investor rather than Government-as-Sovereign, though it also embodies some characteristics 3. Government-as-Sovereign: The Government of China controls China’s four largest banks, Industrial & Commercial Bank of China, Bank of China, China Construction Bank and Agricultural Bank of China, through an investment company, Central Huijin Investment Ltd. These banks represent the Government's instruments for furthering state policy. The Government appoints the bank Chairmen and Presidents, typically members of the Chinese Communist Party. Directed credit is very evident in these banks ’ lending, and periodic financial stress has necessitated recapitalization by the Government. Source: Excerpts Recommendations of the P J Nayak Committee Report to Review Governance of Bank Boards in India, May 2014 Box 4: Governance of Bank Boards in India: Bank Investment Company and Bank Board Bureau The May 2014 Nayak Committee Report proposes that the Government distance itself from several bank governance functions it presently discharges and that the Government transition towards establishing fully empowered boards in public sector banks, solely entrusted with the governance and oversight of the management of the banks. To give effect to this, the Committee recommends the setting up of a Bank Investment Company (BIC) incorporated under the Companies Act to hold equity stakes in banks that are presently held by the Government. Incorporating the BIC under the Companies Act would necessitate the repeal of statutes under which public sector banks are constituted and would require transfer of powers from the Government to BIC through a suitable shareholder agreement and relevant memorandum and articles of association. The Bank Investment Company (BIC) would be constituted as a core investment company under RBI registration and regulation, with BIC assured of its autonomy. It is strongly recommended that its CEO is a competitively selected financial sector professional. A three-phase transition process is recommended for transfer of the Government holding in banks to the Bank Investment Company (BIC) and the transitioning of powers to bank boards with the intent of fully empowering them, as follows: Phase 1 objectives: (a) Legislative amendments enacted to repeal the Acts through which public sector banks have been constituted as statutory bodies, the incorporation of these banks under the Companies Act, and the transfer of their ownership to BIC, with Government initially holding the entire equity in BIC. (b) A professional board constituted for BIC. (c) All existing ownership functions in relation to banks transferred from the Government to BIC. (d) All non-ownership functions, whether of a regulatory or development nature, transferred from the Government to RBI. (e) BIC commences the process of professionalizing and empowering bank boards. (f) Ownership functions taken over by BIC from the Government. In the context of the three-phase process proposed, the recommendation is to entrust the selection of the top management of public sector banks during Phase I to a newly constituted Bank Boards Bureau (BBB). It is desirable that BBB's recommendations be generally accepted by the Government and should therefore be mandated that BBB make a public disclosure of all cases of recommendations made which are rejected by the Government. Headed by the former Comptroller and Auditor General of India, the BBB comprises three ex-officio members (Secretary Financial Services; Secretary Public Enterprises; Deputy Governor RBI) and a mix of public and private sector bankers and independent financial sector professionals. Phase 2 objectives: (g) The reconstitution of bank boards coordinated by BIC. (h) Bank ownership functions continued to be executed by BIC. INDIA DEVELOPMENT UPDATE JUNE 2016 » 64 Phase 3 objectives: (i) All ownership functions transferred by BIC to the bank boards. The appointments of independent bank directors and whole-time directors (including the CEO) become the responsibility of bank boards. (j) BIC ensures that each bank splits the position of the bank’s Chairman into a non-executive Chairman (nominated by BIC) and a CEO (nominated by the board). (k) Strict compliance ensured with Clause 49 of SEBI’s Listing Guidelines, which stipulates a minimum number of independent directors. The Chairman, CEO, other whole time directors and BIC's nominee directors, would constitute the 'inside directors', those connected to the bank's principal shareholder (viz. the Government). All other board members would be 'outside directors', and therefore be characterized as independent. (l) A lead independent director would be nominated for each bank board by the set of independent directors. BIC would define the role of such directors. (m) BIC ceases to exercise ownership functions, and morphs instead into exercising investor functions. (n) Consequently, BIC is tasked with the responsibility of protecting the Government's financial investment in the banks, by raising the financial returns to the Government. Source: Excerpts Recommendations of the P J Nayak Committee Report to Review Governance of Bank Boards in India, May 2014 Financial Institution Restructuring, Resolution, and Exit When shareholders In cases in which shareholders do not consider renewing their engagement in financial sector do not renew their activities, it is important to have a strong legal framework that supports the restructuring of the engagement in activities and their passage to new shareholders or the orderly resolution of distressed financial financial sector institution.36 The current resolution framework in India is designed to protect the interest of activities, a strong depositors and financial stability at the same time. Yet the RBI has commissioned a specific legal framework working group to report on areas where the bank resolution framework may be brought more needs to be in place in line with the recent recommendations of the Financial Stability Board on the key attributes to support the of resolution regimes.37 restructuring Regarding PSBs, there may be specific cases where activities need to be scaled down to reduce the fiscal outlays associated with a recapitalization process. In this context, the share of the financial sector under direct government control may need to be further reduced, in line with the longer-term transformation highlighted in part ii of this chapter. In such event, there would need to be a strategic evaluation by the GOI of the best way the public sector can continue to support financial sector development in general while preserving developmental objectives and the need to ensure financial soundness in the public banks. Creditor rights and insolvency frameworks A well-developed The main instruments for the enforcement of creditor rights in India have been the Debt framework to protect Recovery Tribunals (DRTs) and the Securitization and Reconstruction Financial Assets and creditor rights Enforcement of Security Interests Act (SARFAESI), both of which require court-based ensures the process procedures for the enforcement of creditor rights. The DRTs were created to help financial to recover collateral institutions recover their dues in cases of unsecured debts, while the SARFAESI was designed is fast and effective, to support the collateral recovery efforts of secured bank lenders, but it is applicable only to and instills strong banks and housing finance companies registered with the National Housing Bank. credit discipline among borrowers The slow and complicated debt enforcement procedures in both cases have been delaying decisions. As a result, although amounts handled by both mechanisms have been increasing, 36 Following the passage of the Bankruptcy Code, the government announced plans in June 2016 to bring a new insolvency law for banks and financial institutions to allow for the quick winding up of stressed banks, NBFCs and microfinance institutions while safeguarding the interest of small savers. 37 These recommendations include creating a Financial Resolution Authority (FRA) with a legally backed mandate for resolution of all financial institutions and financial market infrastructures. The FRA would be independent from the regulator and the government, and would also provide deposit insurance, succeeding the Deposit Insurance and Credit Guarantee Corporation (DICGC) in this role. See RBI (2014), Chapter 4, “Framework for Resolution of Financial Institutions.” 65 « INDIA DEVELOPMENT UPDATE JUNE 2016 Table 12 shows that, the recovery rates remain low in the case of DRTs and have declined in the case of the SARFAESI. Indicators of debt recovery times tend to be relatively long compared with comparator countries (Figure 100). In this respect, the new Insolvency and Bankruptcy Code (IBC) promises to represent a significant step forward (see Box 5). Table 12: Banks have recovered about ¼ of claims through SARFAESI NPAs of SCBs recovered through various channels, INR billion Recovery Channel Year DRTs SARFAESI Act 2012–13 Amount involved 310 681 Amount recovered* 44 185 (in percent) 14.1 27.1 2013–14 Amount involved 553 953 Amount recovered* 53 253 (in percent) 9.5 26.6 2014–15 Amount involved 3789 4705 Amount recovered* 531 1152 (in percent) 14 24.5 Source: RBI Figure 100. The time to resolve insolvencies is long… Figure 101. …similar to the time to enforce a contract Time to resolve insolvencies (years) Time to enforce a contract (days) 1600 5 4.5 1400 4 1200 3.5 1000 3 800 2.5 2 600 1.5 400 1 200 0.5 0 0 Source WDI; Note: Line: South Asia average Source WDI; Note: Line: South Asia average Insolvency It is also important that insolvency frameworks recognize creditor’s priorities in payouts as the frameworks can extent to which the creditor priority is protected in the insolvency system will show in the increase certainty volatility of losses for lenders after enforcement proceedings. If the system delivers certainty through clearly that credit instruments with lower priority will be awarded higher losses in the event of default, establishing creditor and creditors with higher priority instruments will be able to recover higher percentages of their priorities loans, then lenders will be able to better price the costs of the different credit instruments. Overall, the lower uncertainty will tend to support lower interest rates across the system. Comprehensive and Collateral is a means of establishing priorities and creating certainty. It is thus important to have reliable credit registries for movable and immovable goods that can be used to secure credit, so that creditors registries that are able to search all such registries in a centralized manner to ensure nobody’s interest has been establish creditor inscribed before theirs. As in many countries, in India creditor priority has to be registered on priorities and timely specific credit registries. adjudication of registered collateral INDIA DEVELOPMENT UPDATE JUNE 2016 » 66 are cornerstones of There is a main central registry of transactions of securitization, reconstruction of financial effective insolvency assets, and creation of security interest over property called the “Central Registry of frameworks Securitisation Asset Reconstruction and Security Interest of India” (CERSAI) established by law. Additional registries (Registrar of Companies, Patents Registry, Trademarks Registry, and so forth) cover other types of transactions. However, there appear to be some shortfalls in the process of collateral adjudication, which are generating the very long periods necessary for collateral recovery reported in India and in contrast to those in comparator countries (Figure 101). In this respect, improved creditor access to a centralized credit registry database that allows for effective searches, so that creditors can ensure their correct priority, would likely help reduce collateral recovery times. Delayed and Delays and uncertainty about the level of losses banks are set to face upon a customer’s default protracted debt are likely to weigh heavily on the estimated cost of credit, which depends on both the probability resolution of default and loss given default, and therefore tend to maintain a high interest rate. Businesses procedures together lose quickly their franchise value, leading the creditor to have to break up and sell the assets with uncertainties separately. In addition, the complications of a long bureaucratic procedure may increase and delays in incentives for certain borrowers to act in bad faith by taking out loans they do not intend to collateral recovery repay. The more lenders trying to execute collateral and the more they are tied up in courts, the increase losses of greater the incentives to take out loans with fraudulent intentions, as expectations are that the lenders in cases of system will not be able to process large volumes of recoveries. insolvency Box 5: India’s New Insolvency and Bankruptcy Code India’s Insolvency and Bankruptcy Code (IBC) was passed by both houses of Parliament in May, 2016. The code creates a framework to oversee the bankruptcy process that: (1) creates a common process to govern all the existing stakeholders (distressed companies, their promoters, creditors, employees); (2) attempts to restore the balance of power – currently skewed toward borrowers – back toward lenders; and (3) attempts to hasten the speed of resolution, so as to best preserve the economic value of the underlying asset. Currently, the bankruptcy process is overseen by 11 different laws, with each of the different stakeholders typically governed by different legal processes. The IBC repeals some of these laws and amends others, such that there is a common process that governs all the stakeholders. Under the new law, for instance, bankruptcy can be triggered by any of the creditors, not just banks. Once a firm defaults, control would shift from the promoters to a committee of creditors, which will have 180 days to evaluate various proposals or else liquidate the assets. During this time, an interim “insolvency professional” will be appointed to take control of the debtor’s assets and company’s operations, collect the financial information of the debtor from information utilities, and constitute the creditors committee. The latter will take decisions with a 75% majority and appoint a permanent insolvency professional to conduct the resolution process. However, given the fundamental changes that the IBC proposes, successful implementation will depend on a host of new administrative structures that need to be created. Specifically, the IBC entails the creation of: (1) a new class of “insolve ncy professionals” (that will specialize in helping distressed companies; (2) “information utilities” that will collect information to prevent defaulters from misusing the system; and (3) a new regulator – Insolvency and Bankruptcy Board of India – to act as the regulator of these entities; and uses the existing infrastructure of National Company Law Tribunals (NCLT) and DRTs for adjudication, though the existing capacity of these tribunals will need to be significantly enhanced. Source: J.P. Morgan Regulation and Oversight The RBI has taken Continued strong financial sector oversight and regulation at the highest standards will many significant contribute to ensuring trust in the financial system, and that the new resources used for bank initiatives to recapitalization will be adequately used. Appropriate supervisory and regulatory architectures strengthen ensure a level playing field for banks irrelevant of their ownership, and reduce the extent of supervision and risky behavior brought about by moral hazard and structural informational asymmetries in the oversight to ensure credit markets. sound risk 67 « INDIA DEVELOPMENT UPDATE JUNE 2016 management in the Capital and liquidity regulation have recently been brought in line with the recommendations banking sector under Basel III, and a recent review of the capital framework (RCAP) has found India compliant with Basel III, which will be fully enforced in India by 2019, therefore strengthening the banking sector to support orderly credit growth. The structure of the Large companies have been the largest bank customers, absorbing more than 80 percent of Indian economy banks’ corporate portfolios, and feature among the largest categories of NPAs. This is to some requires paying extent to be expected because large exposures and potentially connected lending tend to particular attention generate higher losses to the banks in cases of insolvencies given the knock-on effect of one to the lending insolvency on the entire loan portfolio. Monitoring and controlling for large exposures and decisions regarding connected lending in banks’ loan portfolios is a particularly difficult task, and here India stands large companies out among comparable countries, as regulations pertaining to large exposures are not particularly restrictive (Figure 102). To address such concerns, and to improve the oversight and the regulatory environment of large exposures, the RBI has recently produced a consultation document to amend the regulations on large exposures limiting it to 25 percent of Tier 1 capital and bring it more in line with international comparators. Figure 102. India could consider reducing its single-borrower limit in line with peers Regulatory limit to lend to a single borrower or a group of inter-related borrowers, as percent of bank’s regulatory capital 45.0% 40.0% 40.0% 35.0% 30.0% 25.0% 25.0% 25.0% 25.0% 25.0% 25.0% 20.0% 15.0% 10.0% 5.0% 0.0% Brazil India Mexico Russia South Turkey Africa Source: 2013 Global Financial Development Report, Bank Regulation and Supervision Survey and RBI. Notes: India: The large exposure limit is 40% of capital funds in the case of a borrower group, which can be brought to 50 percent for infrastructure projects. The capital funds for the purpose comprises Tier 1 and Tier 2 capital as defined under capital adequacy standards. Russia: There is 800% (of the capital) limit on large exposures (in excess of 5% of the capital). South Africa: 20% – 25% for different groups. Developing tools for managing growing distressed assets while avoiding moral hazard Debt Restructuring Framework The Corporate Debt The CDR mechanism had been initially established in 2001, in the context of high corporate Restructuring distress levels. It was revamped in 2013 by the GOI to facilitate out-of-court agreements (CDR) was created between lenders and borrowers. The CDR mechanism is a voluntary non-statutory system to facilitate the based on Debtor-Creditor Agreements (DCAs) and Inter-Creditor Agreements, which restructuring of incorporate the principle that if the agreement is approved by creditors representing a distressed advances supermajority of 75 percent of total debt the agreement will be binding on the remaining 25 percent of creditors. The supermajority of 75 percent was recently brought down to 60 percent to favor the settling of agreements. The CDR mechanism covers only multiple banking accounts, syndication/consortium accounts, where all banks and institutions together have a large outstanding aggregate exposure. Under the CDR, Under the CDR, banks were also provided incentives to restructure assets and were allowed banks were provided until 2015 to classify restructured loans as performing and therefore release any provisions made INDIA DEVELOPMENT UPDATE JUNE 2016 » 68 incentives to on their account or avoid making provisions for them, as they had not been classified as NPAs. restructure assets This diverges to some extent from common loan classification practices for restructured loans, which typically require an observation period of good performance before allowing for the release of provisions. Until recently, banks By introducing a preferential treatment of restructured loans with the respect to the non- were able to performing loans, the CDR mechanism might have inadvertently supported banks in “evergreen” bad “evergreening” their portfolio to avoid confronting losses from bad debt while instead allowing loans through them to delay recognition by negotiating restructured deals, even when they still proved to be multiple untenable by the borrower. Evergreening is a process whereby banks execute multiple restructurings restructurings of a non-performing loan to avoid classifying it as a non-performing asset. Each restructuring enables the bank to classify the loan as healthy until repayments fall behind again and the loan must be re-restructured. RBI reports that PSBs restructured USD37.6 billion in loans in 2012–2013. This figure represented 94 percent of all loan restructurings in the Indian banking sector during that year. RBI intends to withdraw the regulatory forbearance for restructured loans so that all restructured loans are classified as NPAs attracting a provisioning of 15 percent of the loan outstanding, from the earlier requirement of 5 percent. RBI set up a new Against a background of increasing borrower distress and growing volumes of NPAs on banks’ framework for balance sheets, in 2014 the RBI set up a “Framework for Revitalizing Distressed Assets in the resolving distressed Economy.” The framework incorporates a number of elements to incentivize the early assets in 2014 detection of problematic accounts, the timely restructuring of distressed but solvent borrowers, and the steps for recovery or sale of unviable accounts. As a first step it This has become a very important way of collecting, storing, and disseminating loan required the classification information on large borrowings to all lenders. This enables early detection of establishment of the potentials signs of stress in one borrower, even if he or she is current in all his or her borrowing Central Repository from a particular credit institution. The initiative was further strengthened in April 2015 by of Information of requiring banks to identify so-called special mention assets. These are assets that are not yet Large Credits classified as nonperforming, implying they are less than 90 days overdue, but which, (CRILC) nevertheless, are exhibiting early signs of distress either because they are more than 30 days overdue or because the borrower is known to be in distress. In the second step, Lenders of accounts reported to CRILC as overdue over 90 days or of ‘special mention assets’ lenders of overdue are asked to form a so-called Joint Lenders Forum (JLF), a consortium of creditors that need accounts in CRILC to participate in the debt restructuring negotiations. In this way, re-negotiation discussions begin must form a creditor speedily with the borrower. If a debt restructuring agreement cannot be reached within the group timelines set out in the framework, the JLF will initiate procedures for recovery. Finally, new To facilitate restructuring discussions, the framework contains new guidelines on the structuring guidelines on of project loans so that they should include a periodic refinancing option, to allow the project structuring project loans to be restructured so that the repayment schedule would match the cash flows generated loans are to be by all the different phases of the project. followed RBI also provided a To further support debtors and lenders in the debt restructuring process, the RBI designed the scheme to allow for “Strategic Debt Restructuring” (SDR) scheme in 2014. The SDR scheme allows the JLF to debt-to-equity swaps become the majority shareholder of the distressed borrower corporation, effectively working as a scheme for the conversion of debt into equity. To facilitate the participation of banks in the SDR scheme, the RBI exempted shares acquired in the context of the scheme from being 69 « INDIA DEVELOPMENT UPDATE JUNE 2016 subject to the limits stated under the Bank Regulation Act of 1949.38 However, to preserve the soundness of the banking sector and to prevent banks from holding excessive equity shares in corporations, the RBI also asked that when banks possess equity participations in excess of statutory amounts they should also have a plan for their disposal. In effect, it has been the experience in past cases of corporate distress at an international level that large equity participation by banks has also generated distress, as returns and liquidity of such assets may be limited in times when corporate sector earnings have not recovered. 39 Distressed Asset Sales Asset Another important instrument that helps financial institutions deal with distressed assets was Reconstruction initially set up in 2003 by establishing ARCs, which are entities created to take on non- Companies (ARCs) performing assets and optimize their value. Their role is to acquire the distressed assets from acquire the the banks and proceed to collection, through insolvency procedures and resolution of the distressed assets acquired assets. As volumes of NPAs have increased, so have the assets under management of from the banks and the ARCs since 2013. The ARCs are securitization companies (or “reconstruction companies”) proceed to collection registered under the Securitization and Reconstruction Financial Assets and Enforcement of Security Interests (SARFAESI) Act, which governs the recovery of secured debts (see below). The act provides the legal framework for the activities of the ARCs and places them under the supervision of the RBI. In a recent initiative, the RBI has simplified the process to offload assets to the ARCs, including through setting floors on transaction prices (to prevent collusion with promoters in buy back deals), and allowing for greater foreign participation in ARCs’ capital of up to 74 percent.40 ARCs have been However, several individual or consortiums of banks have set up ARCs directly as subsidiary used to take NPAs corporations using a “capital light” model. Under the “capital light” approach an ARC buys off banks’ balance stressed assets from its parents but pays for these assets primarily using Security Receipts (i.e. sheets without rights to future cash flows) rather than cash. In effect, the ARC buys the parents’ non- transferring the performing loans with a promise to pay using the future cash flow that the non-performing associated risk, but loans themselves might generate. The net effect of the capital light model to ARCs is simply to RBI has taken transfer the non-performing asset off the bank’s balance sheet without transferring the action associated risk of the asset itself, and the valuation of the Security Receipts can have a major bearing on the extent of loss recognition in the banks’ balance sheet. RBI has expressed concern with this model and issued circulars in 2013 and 2014 requiring ARC’s to hold at least 15 percent of the debt liabilities (security receipts) they issue to finance the purchase of the distressed assets. By improving incentives, it is expected they will acquire better assets. However, it still seems to be the case that more than 80 percent of the liability issuance of the ARCs are held by financial institutions, 41 therefore the benefits to banks of selling distressed assets to the ARCs are diminished as their liquidity position is not improved as a result of the sale, nor is the average quality of their assets. 38 The Act limits the extent of participation of banks in equity investments to 30 percent of their total equity or 30 percent of a single company, whichever is less. 39 In June, 2016 RBI announced another restructuring window allowing banks to undertake deep restructuring even with existing promoters. The key proposal of the scheme is to segregate loans into sustainable (part A) and unsustainable parts (part B), with Part B less than half of total loans. The unsustainable debt can then be converted into a cumulative preference share. Thus, banks can effectively take up to 50 percent effective haircut in loans by conversion to equity or preference share. (Source: CreditSuisse). 40 See Keynote address by Mr. Gandhi, Deputy Governor of the Reserve Bank of India, at the “The Economic Times Re Model in India –Asset Reconstruction and NPA Management Summit” Mumbai, 15 September 2015. 41 Keynote address by Mr. Gandhi “Asset Reconstruction and NPA management in India .” INDIA DEVELOPMENT UPDATE JUNE 2016 » 70 Sale of distressed In recent international experience of banking distress, where special purpose vehicles have been assets causes the set up to acquire distressed assets from banks, it has been the case that government-sponsored lender to recognize programs have provided necessary capitalization and liquidity resources to accelerate the potential losses process of dealing with bad debt.42 It is important to note that the sale of distressed assets causes stemming from the lender to have to recognize immediately potential losses stemming from NPAs. This will NPAs that were not occur because likely buyers will not acquire assets for prices above their reasonable recovery adequately value inclusive of collection costs. Thus, in the event the banks have not fully provisioned for provisioned for – all the losses given the default of the borrower (for example because they availed themselves of and may increase the exemptions to do so for debt restructured under the SDR scheme), their capital levels will the demand for be significantly affected by such sales. This will be the case unless there are explicit provisions capital resources for how to recognize such losses. In this respect, increased asset sales by public banks may well be also increasing the demand for further capital resources. Addressing underlying corporate stress is important to closing the pipeline of future NPAs To ensure the As the earlier discussion highlighted, the NPA problem is concentrated in loans by PSBs to pipeline of NPAs is corporates in the infrastructure segment. Sectoral credit stress tests in the RBI’s financial cut off, issues in the stability report confirm that shocks to the infrastructure sector would significantly affect the power and roads system with the most important effect of the single sector shock being on the power and sectors will need to transport sectors (Table 13). Therefore, a sustainable resolution to the NPA issue that ensures be addressed. India’s financial sector can be a reliable pillar of growth will necessitate addressing the underlying problems that led to the accumulation of NPAs in these sectors. Table 13: The power and transport sectors pose significant risks Sectoral credit risk: power and transport; ratios as of September 2015 Financial Power Transport System Sector's share in total advances 9.2 3.2 100.0 Sector's share in system restructured standard advances 29.3 14.6 100.0 Sector's share in system NPAs 5.0 3.8 100.0 Sectoral Restructured Standard Advances Ratio 20.7 29.5 6.5 Shock: Increase in NPAs ratio of the sector 1/ 6.5 9.4 1.0 Shock: Losses as share of system-wide profit 1/ 33.3 16.9 .. Source: RBI (2015) Note: 1/ under most extreme shock Ensuring financial sustainability in the power sector The most financially The power sector accounts for 9.2 percent of total advances and 5 percent of NPAs but 29 stressed sector percent of restructured assets as of September 2015 (Table 13). Power sector loans outweigh appears to be power loans in the transport sector by approximately 3.4 times.43 Within the power sector, the weakest – especially link in the electricity value chain is distribution. Accumulated losses and outstanding liabilities distribution of distribution companies (DISCOMs) amounted to INR 3,800 billion (USD 58 billion) and INR 4,300 billion (USD 66 billion or 0.4 percent of GDP), respectively, as of March 2015. Given that much of the debt had been restructured, DISCOMs had been paying interest rates up to 14–15 percent. The RBI’s Financial Stability Report 2014 (p. 24) noted that “Considering the inadequate fiscal space, it is quite likely that the state governments might not be in a position to repay the overdue principal/installments [of restructured power distribution company loans] in time and banks may be forced to continue classifying these loans as [overdue between 61-90 42 IFC has invested about 3 billion in distressed assets since 2008, and its experience in this area has shown the following to be essential for a well-functioning distressed asset market: (i) servicing capacity outside the banking sector - specialized by asset class, and (ii) the ability to correctly price portfolios of assets. 43 RBI data on industry-wide deployment of bank credit as on July 24, 2015 71 « INDIA DEVELOPMENT UPDATE JUNE 2016 days] (…). Probability of slippage of this exposure into NPAs is very high considering the implementation of new regulatory norms on restructuring of loans and advances…” The Central In this context, the government announced a package of financial assistance to the power Government sector, known as the Ujwal DISCOM Assurance Yojna (UDAY), in November 2015. The announced a package includes debt relief but also aims to achieve long-term financial sustainability by financial bailout addressing inefficiencies in DISCOM operations. The scheme was designed to enable the scheme to address DISCOMs to break even in 2 to 3 years through a combination of financial support and power the stresses faced by sector reforms aimed at improving operational efficiency, reducing the cost of procuring power, state power lowering interest expenditures and ensuring financial discipline by making the state take over distribution the losses and outstanding debts on their books. States will take over 75 percent of DISCOMs’ companies outstanding debt over two years – 50 percent in FY16 and 25 percent in FY17. 44 Under the scheme, the debt taken over will be excluded from the calculation of fiscal deficit (as required under the FRBM) in FY16 and FY17 – allowing the states to maintain current levels of capital and social expenditures. As of April 2016, 15 states have joined the scheme. UDAY provides The UDAY scheme is the third financial restructuring package for DISCOMs in the past 15 better incentives and years (Figure 103). Given the concern of limiting moral hazard, it differs from previous ensures better packages in that it puts greater responsibility on the state governments and provides stronger monitoring than incentives for better monitoring and supervision of DISCOM operations. Specifically, UDAY previous bailouts requires that from FY18, state budgets will have to absorb DISCOMs’ losses without further accommodation by the central government on deficit constraints under the FRBM. UDAY is better designed to encourage participation by states: (i) it involves an incentive scheme for states and DISCOMs to improve DISCOM administration;45 and (ii) leverages the GOI’s centrally owned electricity companies and fuel resources to lower the cost of bulk electricity supply. Commitment to a number of other micro-level reforms aimed at aligning incentives (greater use of technology, HR incentives, and tariff petition commitments) are also part of the package. Figure 103. Buoyant credit growth to the power sector Figure 104. …largely because of persistently high declined as the sector was under stress… technical losses Change from the previous year, percent AT&C losses (latest of FY13 and FY14), percent 90 50 46 80 45 42 39 70 40 34 35 32 33 60 30 27 28 50 25 25 22 22 23 40 20 18 19 16 16 14 15 15 15 30 15 20 10 7 10 5 0 0 MH MP MZ PJ TN AP KL KA BH HP UK CG UP RJ Norm (min) WB GJ HR JH OD Norm (max) Bailout Power Sector Credit Growth Source: RBI, World Bank staff calculations Source: Ministry of Power 44 According to the UDAY scheme, states will issue non-SLR bonds (SDL) with maturity period of 10–15 years with a moratorium on principal up to 5 years. Ten-year Bond Pricing: 7.92% (as per last RBI auction of state SDLs) + 0.25% spread for non-SLR status on semi-annual compounding basis (market driven, subject to cap of 10 yr. G-Sec + 50 + 25 bps). 45 States accepting UDAY and performing as per operational milestones will be given additional/priority funding through Deendayal Upadhyaya Gram Jyoti Yojana (DDUGJY), Integrated Power Development Scheme (IPDS), Power Sector Development Fund (PSDF), or other such schemes of Ministry of Power and Ministry of New and Renewable Energy. INDIA DEVELOPMENT UPDATE JUNE 2016 » 72 UDAY’s success will The financial restructuring actions are estimated to reduce costs for the distribution companies depend on reducing by about 10 percent during the first year and an additional 5 percent the second year. More AT&C losses of importantly, reaching financial break-even status within three to four years with relatively DISCOMs, and risks moderate tariff increases will require that the operational targets for loss reduction to also be are significant met. Therefore, the success of UDAY will be measured largely by the ability of DISCOMs to reduce their Average Technical & Commercial (AT&C) losses, which currently stand well above international norms (Figure 104). Depending on the particular performance of state utilities, a turnaround of distribution companies might take longer than the current ambitious timetable. Public-Private Partnerships in the highway sector The challenge More than half of all The World Bank conducted research into India’s PPP programs from 2012 through 2014 and highway PPP estimates that more than half of all highway PPP projects are under stress. One aspect of this projects are work entailed examining stress in the portfolio of tendered contracts. The working definition estimated to be of stress included in this research was as follows: (i) operational projects with cash flows that under stress cannot consistently fund debt or premium obligations; (ii) projects under construction that are facing significant delays or cost escalations; and (iii) awarded projects unable to achieve financial closure. It is estimated that more than half of all projects are under stress, with half of stressed projects currently under construction (Figure 105). Meanwhile roughly two-thirds of all projects under development at the time of this research were showing significant signs of stress. Figure 105. Breakdown of stress in the portfolio of national highways contracts (2014) Portfolio under stress by km (includes completed and under development); Status of stressed contracts (by km) Under construction Stable 49% Completed 45% 11% Under stress 55% Construction not started Terminated 30% 10% Source: 2014 World Bank research Key causes of stress The cause of stress in the portfolio of contracts varies by project. Some of the more typical are delays, revenue causes include: (i) delays owing to approvals, land acquisition, and other authority shortfalls, and responsibilities; and (ii) revenue shortfalls combined with large debt service requirements where speculative factors developers have overestimated traffic volumes. There have also been instances where policy in the bidding model decisions in tolling or popular discontent have affected project revenues. Singh and Brar (2016) point to the inherent role of speculative factors in the bidding model for contracts and related potential for fraud. They argue that “some Concessionaires increase the total project cost (TPC) by two or three times the original TPC prepared by the procuring authority at the time of bidding to obtain higher grant and debt from PSBs or the procuring authority itself. This allows the concessionaire to practically fund the entire construction cost with public funds, while itself contributing zero or negative equity.” Meanwhile, in the case of 73 « INDIA DEVELOPMENT UPDATE JUNE 2016 defaults, banks have a difficult time enforcing their security interest on the holding company of the concessionaire, which may have been able to take out funds from the concessionaire, for example by hiring related companies to provide services. Impact on banks Stress in the The stress in the portfolio of highways PPPs presents a challenge for the government because portfolio of more than 80 percent of debt financing to highways PPPs has come from PSBs. Figure 106 highways PPP summarizes the exposure of individual PSBs that hold the majority of highways PPP-related projects is flowing debt based on Basel III regulatory filings. It is important to note that the figures presented offer through to the banks an imperfect view. Public sector banks are currently using different reporting formats for their that have financed Basel III regulatory filings and some co-mingle portfolios such as roads, rail, and ports whereas highways PPPs others break out sub-sectors in more detail. Nevertheless, the figures below suggest that the State Bank of India group potentially holds the largest share of highways related debt with IDBI Bank second. Figure 106. SBI concentrates a large share of roads and transport assets Funded exposure to roads and other transport as noted, INR million 450,000 SBI holds about US$ 6.3 billion in roads and ports debt.≈ ¼ of credit 400,000 deployed to the sector nationally. Note the large year on year change between 2014 and 2015 when SBI shifted exposure off balance 350,000 sheet 300,000 250,000 2014 exposures 200,000 2015 exposures 150,000 100,000 † 50,000 † † - Source: Basel III filings and analyst presentations (where Basel III data is incomplete) from June 30, 2015 Notes: All banks shown have more than 50 percent GOI shareholding. Mingled roads and ports figures marked†, mingled roads and rail marked‡ for individual years (above bar) or both years (next to bank name). Some of the banks Some of the banks with lower total exposure to highways debt are actually more vulnerable to with lower total changes in the value of those assets because of smaller capital bases (e.g. Andhra Bank, United exposure to Bank of India, IDBI, Corporation Bank). In contrast, SBI (which appears to hold more highways debt may highways debt than others) has proportionately more capital to cushion potential losses (See in fact be more Box 6). SBI’s size also offers potential advantages in the ability to influence sponsors and public vulnerable policy. There is a risk that smaller banks may actually be most vulnerable to stress in the portfolio of highways debt but least capable of influencing sponsors policymakers. Box 6: Impact of infrastructure loans on SBI The State Bank of India (SBI) offers an important look into the status of highways-related credit owing to both the size of its portfolio and the level of detail offered in its regulatory filings. Specifically, the Basel III reports from SBI include a breakdown of non- performing assets (NPAs) by sector and subsector. This provides a time-series view of how stress is developing in the portfolio of highways related debt. Figure 107 summarizes SBI’s rates of non-performing assets across its overall infrastructure portfolio. Consistent increases in the rate of gross NPAs for infrastructure and roads in particular throughout 2014 suggest a deteriorating trend in the quality of highways PPP related debt. The apparent decrease in March 2015 corresponds to a shifting of NPAs off balance sheet to an ARC, as described earlier. Had this shift not taken place, the trend line may have been opposite for this quarter. INDIA DEVELOPMENT UPDATE JUNE 2016 » 74 An analysis of SBI’s lending to infrastructure highlights that the market for financing highways has effectively stalled while other sub- sectors have continued to source new financing commitments (Figure 108). The steady decline in SBI’s funded portfolio of infrastructure loans following the selloff of assets in 2014 also suggests that projects may not be drawing down on their unfunded financing commitments. This could suggest that projects under development that already have financing commitments are stagnating and failing to make physical progress. Figure 107. Roads and ports are the most stressed part of Figure 108. SBI funded exposure to highways and other SBI’s infrastructure portfolio infrastructure sectors Trends in SBI’s NPAs by infra. Sector; NPAs as % of sector portfolio Indexed total exposure (funded + unfunded Dec. 20013 =1) 12% 1.90 Note: SBI made a large sale of NPAs to an asset All infrastrcuture reconstruction company(ies) in this quarter. 1.80 Power 10% Telecoms 1.70 Roads and ports 8% 1.60 1.50 6% 1.40 4% 1.30 2% 1.20 1.10 0% Dec-13 Mar-14 Jun-14 Sep-14 Dec-14 Mar-15 Jun-15 1.00 0.90 All infrastrcuture Power Telecoms Roads and ports Source: SBI Basel III filings for dates noted Source: CEIC, World Bank analysis of SBI group Basel III filings for dates noted Resolving the stress India’s challenge of Many countries have had to realign their highways PPP programs after encountering financial addressing stress in challenges. However, macroeconomic shocks or currency devaluations have usually precipitated the portfolio of restructuring efforts (e.g. France in the 1980s, Mexico in 1994, Malaysia in 1997, Brazil in 1999, highways PPP and Argentina in 2001). The stress in India’s PPP portfolio can potentially remain latent for contracts is not longer as the PPP program is almost entirely financed in domestic currency and less susceptible unique to an external catalyst that would create an immediate need for restructuring. It is also important to note that India’s highways PPP program is much larger than Latin American, European, or Asian examples, which means the costs of waiting are also likely to be higher (Figure 109). Figure 109. Sizing up programs under stress Program size in USD million, adjusted to 2014 prices 48,000 India (2014) 38,000 NH only 28,000 18,000 Mexico (1994) Chile (1998) 8,000 France (1982) Brazil (1998) Colombia (1998) (2,000) 0 5,000 10,000 15,000 20,000 25,000 30,000 Km Concessioned Source: (i) Mexico program size: PPI database; (ii) Brazil program size from: de Castro; (iii) French program size from: Fayard Notes: (i) size of the French program estimated as book value of nationalized company’s debt and equity; (ii) program sizes adjusted to 2014 prices using IMF inflation data from the WEO; (iii) India figure only includes national highways program. 75 « INDIA DEVELOPMENT UPDATE JUNE 2016 International Two mechanisms are summarized in Table 14. Option A follows a model of voluntary workouts experience provides and renegotiations that has typically been used on smaller programs where changing factors insights regarding such as toll rates, contract periods, and/or interest on project debt provides enough relief to two specific make concession companies solvent and liquid again. Option B entails a more drastic mechanisms that intervention by the government to take possession of contracts explicitly and recapitalize them India may consider (usually under a temporary holding company). Once stable, a government may then re-divest for relieving stress of individual concessions or bundles of concessions (e.g. clustered by geographical area). This approach has been used for larger programs where insolvency has been more acute. Table 14: Options for relieving stress in the portfolio of active contracts Option B: Take over, clean up, Option A: Voluntary workouts & renegotiations get out  Government seeks to renegotiate contracts on  Government leads an effort to assume What would it voluntary basis. This may include developing a control of concessions, restructure them to entail? special financing window for replacing short term optimize their value and drive further debt with government provided long term debt to investment. GOI could then re-divest to improve project cash flows private investors at an opportune time  Malaysia, Who has tried  France  Chile something similar?  Mexico  Colombia Both approaches It is important to note that the options above are not mutually exclusive. Specifically, Option carry advantages A (workouts and renegotiations) could precede Option B (temporary takeover) if additional and risks stress relief became necessary. One caveat however is that the optics of being seen to fail at implementing workouts and renegotiations may compromise other efforts. Enhanced reporting of fiscal risks and shifting of financing sources complement sector-specific approaches Improved public Improved monitoring and disclosure of the build-up of contingent liabilities of the public sector accounting of PPPs embedded in the PSBs’ funding of public infrastructure would be welcome by inter-alia making will also help avoid provisions for such expenses in annual government budgets and in a better sharing of risks the resurgence of under the PPPs and SOEs (such as DISCOMs). At present, such contingent liabilities are NPAs implicit in the balance sheet and the income statements of the PSBs, with the losses eventually recognized in the form of write-downs of bank or SOE balance sheets. Moreover, improved transparency would provide a stronger basis to judge the performance of PSBs. Diversifying A key challenge for banks to act as primary financiers of infrastructure is the fact that core bank financing sources of liabilities – namely deposits – are by nature short-term, while project loans are by nature long- infrastructure would maturity assets. The RBI has put in place a comprehensive framework for asset and liability also help management as part of its norms to ensure the soundness of the banking system, and developed enabling regulations for raising long-term bonds by banks for financing project loans. However, in the recent Financial Stability Report (2015) the RBI cautioned that “(…) banks’ processes and business models may not yet be adequately prepared to make, monitor and manage long- term project loans. Therefore, entities with long term investible resources such as pension funds and insurance companies need to be encouraged in this space” INDIA DEVELOPMENT UPDATE JUNE 2016 » 76 iv. Summary of policy options to ensure the financial sector remains a pillar of growth Structural transformation of the financial Sector 1. Increase India can continue to pursue an organic growth in the participation of private and foreign banks commercial relative to PSBs, while increasing the commercial orientation and practices of public sector orientation of banks by implementing the recommendations of the Nayak report with respect to governance the banking and separation of ownership and management of PSBs. sector 2. Relax mandates The government may consider continuing to reduce the SLR and the share of banks’ portfolio on banks mandated to serve priority sectors, and consider a roadmap for phasing out those mandates altogether, subject to meeting certain benchmarks. For example, lowering of the SLR may be tied to gradually raising limits in foreign participation in domestic currency government bonds and a stabilization of the government’s fiscal deficit (and funding needs). 3. Accelerate Limits could be raised not only on the participation of foreign investors in local currency development of government bond markets but also in corporate bond markets. Naturally, this needs to be capital markets pursued with due consideration for the potential impact of financial flow volatility on the Indian economy, and could also be part of a long-term roadmap. In addition to reviewing the quantitative limits in place, the government may also review the process by which foreign institutional investors access the Indian markets. Dealing with rising NPAs 1. Ensure success To ensure that the power sector no longer provides a pipeline of NPAs to the banking sector of the UDAY it will be important to ensure the success of the UDAY scheme, especially in reducing the scheme AT&C losses of DISCOMs. Preparing and publishing financial results of DISCOMs in a timely manner will be helpful in increasing transparency and facilitating accountability of DISCOM management to the achievement of the targets agreed as part of the scheme. 2. Consider deeper In the case of infrastructure, a gradual approach has been taken to resolve PPP projects under reforms in road stress in the roads sector. The government may consider a more proactive action along the lines PPPs of Options A and B in Table 14. In addition, it will be important that India moves toward a new PPP model for the roads sector that will prevent renewed accumulation of NPAs in the future, while ensuring the effective delivery of much-needed infrastructure. 3. Pursue The need for capital infusion to be performance-linked and accompanied by substantial substantive governance reforms cannot be overemphasized. Bank recapitalization provides an important governance opportunity to pursue deeper governance reforms and strengthen institutional capacity in reforms in critical functions including risk management and credit origination to meet the objective of exchange for having a more commercially oriented public banking sector. Ensuring high standards for the fresh capital selection of top executives and the board, as well as aligning staff and management incentives toward profit generation, are important areas for governance reforms. 4. Enhance The IBC will be an important step in improving India’s insolvency framework. Ensuring that insolvency and the required administrative structures are in place in a timely manner, and that the law is contract implemented effectively in the courts, will be equally important to ensure that its objectives are enforcement met. Moreover, providing banks with adequate tools to manage balance sheet stress, while frameworks and avoiding moral hazard, will be necessary to reduce the amount of new capital that banks will tools require to resume orderly expansion of their balance sheets and of credit to the economy. 77 « INDIA DEVELOPMENT UPDATE JUNE 2016 References Acharya, Viral V.; Agarwal, Anukaran; Kulkarni, Nirupama “State Ownership and Systemic Risk: Evidence from the Indian Financial Sector during 2007-2009” August 23, 2010. Ang, James B. “Financial liberalization, financial sector development and growth: Evidence from Malaysia” Journal of Development Economics 2007. Arslanalp, Serken and Takahiro Tsuda “Tracking Global Demand for Emerging Market Sovereign Debt”, IMF 2014, Working Paper 14/39 Bahal, et al (2015), “Crowding-out or crowding-in? Public and Private Investment in India”, IMF Working Paper (WP/15/264) Calderon, Cesar and Lin Liu “The direction of causality between financial development and economic growth” Journal of Development Economics 72 (2003) 321-334 Cecchetti, Stephen G and Erisse Kharroubi “Reassesing the impact of finance on growth” BIS Working Papers (2012) No. 381 Chakraborty, Pinaki. “Restructuring of Central Grants”, EPW, Feb 6, 2016. Chinoy, Sajjid Z “India: Landmark bankruptcy law gets a step closer to implementation” J.P. Morgan Asia Pacific Emerging Markets Research, May 2016 Cole, Harold L., Jeremy Greenwood, and Juan M. Sanchez. “Why Doesn't Technology Flow from Rich to Poor Countries?” No. w20856. National Bureau of Economic Research, 2015. Economic Survey 2015-16, Chapter 9 “Reforming the Fertilizer Sector”, p.131, 2016 Fayard, Alain “Les Autoroutes et Leur Financement.” la Documentation française, 1980 (Notes et études documentaires, n°4597-4598, 10 décembre 1980). Financial Sector Assessment Program, IMF country report no. 14/98, 2014 Financial Stability Report, Reserve Bank of India, December 2015 Global competitiveness report 2014-15, World Economic Forum page 44 Global Economic Prospects: Spillovers amid weak growth” World Bank January 2016 Global Financial Development Report, 2013, World Bank Hasan, Iftekhar, Roman Horvath, and Jan Mares. "What type of finance matters for growth? Bayesian model averaging evidence." World Bank Policy Research Working Paper 7645 (April, 2016). India Development Update: Fiscal Policy for Equitable Growth, World Bank. October 2015. India, Selected Issues, IMF country report no. 16/76, February 2016. Keynote address “Asset reconstruction and NPA management in India” by Mr. R Gandhi, Deputy Governor of the Reserve Bank of India at the The Economic Times ReMode in India –Asset Reconstruction and NPA Management Summit, Mumbai, 15 September 2015. King, Robert G. and Ross Levine “Finance and Growth: Schumpeter Might be Right” Quarterly Journal of Economics, Vol 108, No. 3 (August 1993), pp 717-737 Levine, Ross; Norman Loayza and Thorsten Beck “Financial intermediation and growth: Causality and causes” Journal of Monetary Economics 46 (2000) 31-77 Levine, Ross “Finance and growth: Theory and Evidence” Published in the Handbook of Economic Growth, edited by Aghion and Durlauf, 2005 INDIA DEVELOPMENT UPDATE JUNE 2016 » 78 Moody’s investor service, “Indian Banks Could Need US$26-37 billion in External Capital for Basel III Compliance; Rated public-sector banks currently meet minimum capital requirements, but barely.” September 19, 2014 Mundra, S.S. “Asset Quality Challenmges in India: diagnosis and prognosis”, Deputy Governor of RBI at the Edelweiss Credit Conclave, Mumbai, April 2016 Newton de Castro, “Privatization of the transport sector in Brazil,” 2002, A privatização no Brasil: o caso dos serviços de utilidade pública. OCDE/BNDES Peiris, SJ “Foreign participation in emerging markets local currency bonds”, IMF 2010 – WP/10/88 Report of Working Group on Resolution Regime for Financial Institutions, Chapter 4, “Framework for Resolution of Financial Institutions.”, RBI (2014), Report of the PJ Nayak Committee to Review Governance of Boards of Banks in India, Reserve Bank of India, May 2014 Schumpeter, J.A., [1911] (2008), “The Theory of Economic Development: An Inquiry into Profits, Capital, Credit, Interest and the Business Cycle” Transaction Publishers Singh, Charan and JS Brar “Stressed Assets and Banking in India”, IIM Bangalore Working Paper No. 507 , 2016 Stiglitz, J. E. 2000. “Capital Market Liberalization, Economic Growth, and Instability.” World Development 28, 1075– 86. “You go to war with weapons, not bare hands: SBI’s Arundhati Bhattacharya”, Live Mint, June 13, 2016. 79 « INDIA DEVELOPMENT UPDATE JUNE 2016