## _wp1468 ## Source details **Canonical URL:** [_wp1468](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2014/_wp1468.pdf) ## Other formats - [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2014/_wp1468.pdf.md) - [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2014/_wp1468.pdf.json) --- ### I. Introduction - India recorded robust annual growth of 9 percent plus during 2004-08, accompanied by consolidation of key macroeconomic indicators. - North Atlantic financial crisis (NAFC) in 2008 caused a setback: growth rebounded after large monetary and fiscal stimuli but slowed significantly subsequently. - Post-2008 the economy experienced a substantial widening of current account and fiscal deficits and elevated inflation. - Declines in domestic saving and investment rates raise concerns that India’s potential growth rate has fallen significantly (IMF, 2013; Mishra, 2013). - Alternative view: the 2004-09 high growth may have been a debt-led cyclical boom supported by unprecedented capital inflows coinciding with exceptional world growth (Nagaraj, 2013). ### II. Recent macroeconomic trends: India in a global perspective - Growth and macro outcomes: - Real GDP Growth (factor cost) and selected indicators (Table 1, selected years and values): - 2003-04: Real GDP Growth (factor cost) 8.0; Real GDP Growth (market prices) 7.9; GFD/GDP (Centre) 4.3; CAB/GDP 2.3; Non-oil CAB/GDP 5.0; WPI Inflation 5.5; CPI Inflation 3.9; REER Index 96.8; Real Policy Rate -0.4 - 2004-05: Real GDP Growth (factor cost) 7.1; Real GDP Growth (market prices) 7.9; GFD/GDP (Centre) 3.9; CAB/GDP -0.3; Non-oil CAB/GDP 2.8; WPI Inflation 6.5; CPI Inflation 3.8; REER Index 99.9; Real Policy Rate -0.6 - 2005-06: Real GDP Growth (factor cost) 9.5; Real GDP Growth (market prices) 9.3; GFD/GDP (Centre) 4.0; CAB/GDP -1.2; Non-oil CAB/GDP 2.7; WPI Inflation 4.4; CPI Inflation 4.4; REER Index 102.7; Real Policy Rate 1.7 - 2006-07: Real GDP Growth (factor cost) 9.6; Real GDP Growth (market prices) 9.3; GFD/GDP (Centre) 3.3; CAB/GDP -1.0; Non-oil CAB/GDP 3.0; WPI Inflation 6.6; CPI Inflation 6.7; REER Index 101.0; Real Policy Rate 3.1 - 2007-08: Real GDP Growth (factor cost) 9.3; Real GDP Growth (market prices) 9.8; GFD/GDP (Centre) 2.5; CAB/GDP -1.3; Non-oil CAB/GDP 2.9; WPI Inflation 4.7; CPI Inflation 6.2; REER Index 108.6; Real Policy Rate 2.2 - 2008-09: Real GDP Growth (factor cost) 6.7; Real GDP Growth (market prices) 3.9; GFD/GDP (Centre) 6.0; CAB/GDP -2.3; Non-oil CAB/GDP 3.1; WPI Inflation 8.1; CPI Inflation 9.1; REER Index 97.8; Real Policy Rate 0.9 - 2009-10: Real GDP Growth (factor cost) 8.6; Real GDP Growth (market prices) 8.5; GFD/GDP (Centre) 6.5; CAB/GDP -2.8; Non-oil CAB/GDP 1.5; WPI Inflation 3.8; CPI Inflation 12.2; REER Index 95.3; Real Policy Rate 1.5 - 2010-11: Real GDP Growth (factor cost) 8.9; Real GDP Growth (market prices) 10.3; GFD/GDP (Centre) 4.8; CAB/GDP -2.7; Non-oil CAB/GDP 1.1; WPI Inflation 9.6; CPI Inflation 10.5; REER Index 103.5; Real Policy Rate 2.0 - 2011-12: Real GDP Growth (factor cost) 6.7; Real GDP Growth (market prices) 6.6; GFD/GDP (Centre) 5.7; CAB/GDP -4.2; Non-oil CAB/GDP 1.1; WPI Inflation 8.9; CPI Inflation 8.4; REER Index 100.7; Real Policy Rate 1.0 - 2012-13: Real GDP Growth (factor cost) 4.5; Real GDP Growth (market prices) 4.7; GFD/GDP (Centre) 4.9; CAB/GDP -4.7; Non-oil CAB/GDP 1.1; WPI Inflation 7.4; CPI Inflation 10.4; REER Index 96.3; Real Policy Rate 1.9 - 2013-14: Real GDP Growth (factor cost) 4.9; Real GDP Growth (market prices) 4.6; GFD/GDP (Centre) 4.6; CAB/GDP -2.3 (April-December) - Global context: - Global growth fell from an annual average of 4.8 percent during 2003-07 to an average of 2.9 percent during 2008-12. - In 2012, among G-20 economies India had the third largest CAD after Turkey and South Africa; India’s fiscal deficit was the second largest after Japan. ### III. Domestic macroeconomic policies and growth slowdown - Stimulus and withdrawal: - Monetary and liquidity stimulus between September 2008 and April 2009: policy rate cut from 9.00 percent to 3.25 percent; cash reserve ratio cut from 9.0 percent to 5.0 percent. - Other measures collectively had potential to release liquidity of more than 10 percent of GDP (Mohan, 2011c). - Fiscal measures: CENVAT rate cut from 14 percent to 8 percent between December 2008 and February 2009; increased plan expenditure; Pay Commission implementation; agriculture debt waiver. - Centre’s headline gross fiscal deficit increased from 2.5 percent of GDP in 2007-08 to 6.0 percent in 2008-09; including bonds issued in lieu of cash subsidies, GFD/GDP rose from 3.1 percent to 8.2 percent. - Net effect: both monetary and fiscal policy provided strong support—deemed excessive with hindsight—to domestic demand in 2008-09. - Stimulus withdrawal was phased and gradual during 2010-11 and 2011-12; initial monetary accommodation and delayed withdrawal contributed to high inflation and later slowdown. ### IV. Monetary policy, inflation, and real interest rates - Inflation dynamics: - Stronger-than-expected growth in 2009-10 and 2010-11 translated into high inflation, initially food inflation by end-2009 and underlying inflation by April 2010. - Non-food non-fuel CPI inflation (rural and urban combined based on the new CPI series) has been around 8 percent since June 2012 (it was around 10 percent during January-May 2012). - Real policy and lending rates (selected numbers preserved): - Real policy rate fell from an average of 2.2 percent in 2007-08 to 0.9 percent in 2008-09. - Real policy rate edged up to 2.0 per cent in 2010-11. - Real policy rate fell back to an average of 1.0 percent in 2011-12. - Real policy rate edged higher to 1.9 percent in 2012-13. - The real policy rate in terms of CPI inflation has been generally negative since mid-2008 (RBI, 2014). - Real deposit rates (depositors): - Real deposit rates fell from an average of (+) 1.5 percent during 2003-08 to (-) 1.9 percent during 2008-13. - Using households’ inflation expectations, real deposit rate averaged (-) 2.1 percent using households’ “current” inflation expectations and (-) 3.3 percent using households’ “one-year ahead” inflation expectations during 2008-13. - Monetary policy assessment: - While nominal policy rate increased gradually during 2010 and 2011, the real policy rate was highly negative (w.r.t. headline WPI) and marginally negative (w.r.t. NFMP), implying monetary policy was still accommodative over this phase. ### V. External sector, CAD, and exchange rates - CAD dynamics and drivers: - CAD averaged less than 1 percent of GDP during 1992-2008; widened to just under 5 percent in 2012-13; improved to 2.3 percent in the first three quarters of 2013-14 following policy actions. - Net oil imports rose from 4.1 percent of GDP in 2007-08 to 5.9 percent by 2012-13. - Gold imports increased from around 1.5 percent of GDP during 2004-08 to 3.0 percent in 2011-12 and 2012-13. - Balance of payments (selected entries, percent of GDP): - Merchandise exports: 2003-04 10.7; 2007-08 13.4; 2011-12 16.5; 2012-13 16.5 - Merchandise imports: 2003-04 12.9; 2007-08 20.8; 2011-12 26.6; 2012-13 27.0 - Trade balance: 2003-04 -2.2; 2007-08 -7.4; 2011-12 -10.1; 2012-13 -10.5 - Current account balance: 2003-04 2.3; 2007-08 -1.3; 2011-12 -4.2; 2012-13 -4.7 - Capital flows, net: 2003-04 2.7; 2007-08 8.6; 2011-12 3.6; 2012-13 4.8 - Foreign exchange reserves (US $ billion): 2003-04 113; 2007-08 305; 2012-13 294 - Memo US$ billion: Current account balance 2003-04 14; 2007-08 -16; 2011-12 -78; 2012-13 -88 - Financing risks and composition: - Inward FDI flows fell from 3.4 percent of GDP in 2008-09 to 1.5 percent in 2012-13. - Portfolio flows and debt flows financed almost two-thirds of the CAD in 2012-13. - Stock of external commercial borrowings rose from around US 30 billion during 2000-04 to US $115 billion in 2007-08 and to US $225 billion in 2012-13. - REER measurement divergence (March 2008 to March 2013): - RBI index: real depreciation of 12 percent. - BIS index: real depreciation of 4 percent. - OECD and IMF indices: real appreciation of 12-14 percent. - Measurement differences driven by choice of price deflators (RBI uses WPI for India; OECD/IMF use CPI). ### VI. Interpretation of the slowdown and quantitative estimates - Two broad interpretations: - Return-to-potential: pre-NAFC period had global growth above potential; post-NAFC slowdown is a return to underlying potential. - Domestic factors: extent of slowdown, elevated inflation, and domestic imbalances suggest domestic factors added to global headwinds. - Quantifying impacts (Annex 2 empirical results): - An increase of 100 bps in the real interest rate leads to a reduction of 26-33 bps (around 30 bps) in GDP growth with a lag of two quarters (Annex Table 2 columns). - A one percentage point reduction in global exports reduces domestic growth by almost 13 bps. - A one percentage point reduction in global GDP growth reduces domestic growth by almost 40-56 bps. - Fiscal stimulus and consolidation: - India’s structural primary deficit increased by 4.0 percentage points of GDP in 2008-09. - Withdrawal of the stimulus was 1.4 percentage points during 2010-13 (annual average 0.35 percentage point of GDP). - Assuming fiscal multiplier ~0.5, the 4 percent of GDP stimulus added almost 2 percentage points to growth in 2009-10; annual fiscal consolidation implied less than 20 bps per annum adverse impact on growth. ### VII. Real sector: investment, savings, and corporate performance - Corporate sector: - Corporate savings fell from 9.4 percent of GDP in 2007-08 to 7.1 percent in 2012-13. - Corporate investment fell from 17.3 percent of GDP in 2007-08 to 9.2 percent in 2012-13. - Outward FDI by Indian corporates: 0.3 percent of GDP in 2003-04; 1.5 percent in 2007-08; 0.4 percent in 2012-13. - Gross capital formation and saving-investment balance (selected values): - Gross capital formation (percent of GDP): 2003-04 26.1; 2004-05 32.5; 2005-06 34.4; 2006-07 35.9; 2007-08 38.0; 2008-09 35.5; 2009-10 36.3; 2010-11 36.5; 2011-12 35.5; 2012-13 34.7 - Saving-Investment Balance, net (4-9): 2003-04 2.2; 2004-05 -0.4; 2005-06 -1.2; 2006-07 -1.1; 2007-08 -1.3; 2008-09 -2.3; 2009-10 -2.8; 2010-11 -2.8; 2011-12 -4.2; 2012-13 -4.7 - Households: - Household total savings (percent of GDP): 2003-04 23.1; 2004-05 23.6; 2005-06 23.5; 2006-07 23.2; 2007-08 22.4; 2008-09 23.6; 2009-10 25.2; 2010-11 23.1; 2011-12 22.8; 2012-13 21.9 - Household financial saving (gross) (percent of GDP): 2003-04 11.0; 2004-05 10.1; 2005-06 11.9; 2006-07 11.3; 2007-08 11.6; 2008-09 10.1; 2009-10 12.0; 2010-11 9.9; 2011-12 7.0; 2012-13 7.1 - Household financial savings (net) available for private corporate sector fell from 15.5 percent of GDP in 2007-08 to 10.8 percent in 2012-13 (narrative figure). - Households’ physical investment in gold increased from an average of 1.1 percent of GDP during 2003-08 to 2.6 percent by 2012-13. - Crowding out and channels: - Public savings deteriorated from 5 percent of GDP in 2007-08 to 1.2 percent in 2012-13. - Net financial savings available to private corporate sector fell from average 6.3 percent of GDP during 2003-08 to 1.9 percent during 2008-13. - Reasons for crowding out include higher government borrowing needs (near-trebling), monetization via OMOs (28 percent of overall borrowing requirement on average), and capital flows. ### VIII. Food, MSP, and supply constraints - Minimum Support Prices (MSP) and WPI (selected series, 2003=100): - Rice WPI: Mar-03 100.0; Mar-04 97.7; Mar-05 101.1; Mar-06 104.0; Mar-07 113.4; Mar-08 129.7; Mar-09 149.7; Mar-10 161.7; Mar-11 165.4; Mar-12 173.7; Mar-13 204.2 - Rice MSP: Mar-03 100.0; Mar-04 103.8; Mar-05 105.7; Mar-06 107.5; Mar-07 109.4; Mar-08 121.7; Mar-09 169.8; Mar-10 188.7; Mar-11 188.7; Mar-12 203.8; Mar-13 235.8 - MSP gaps and analogous series for Wheat, Gram, Arhar, Moong, Urad preserved in source tables. - Drivers of pulse price increases: - Strong domestic growth from 2003-04; plateauing domestic pulse production; shift in dietary patterns toward protein-rich items; support from MGNREGA. - Policy implication: - Productivity gains in food production are suggested as a durable, non-inflationary solution. - Monetary policy can keep overall inflation within its comfort zone in the face of persistently high food inflation only at large output cost. ### IX. Exports, imports, REER and empirical evidence (Annex 3 summary) - Export demand estimates (annual 1980-81 to 2007-08): - An increase of 1 percent in world real exports is associated with an increase of 1.1-1.4 percent in India’s exports in the long-run. - Real exchange rate effect: 10 percent real appreciation leads to a reduction of almost 2.1-6.5 percent in the volume of exports in the long-run (Annex Table 3). - Quarterly post-reforms estimates (1996-97 to 2007-08): - External demand elasticity: 1.6-1.9 (world exports) and 2.6-3.6 (world GDP). - REER elasticity: absolute coefficient on REER is substantially higher in quarterly estimates; coefficient sensitive to choice of REER and inclusion/exclusion of oil exports. - Imports: - Income elasticity of imports with respect to domestic industrial GDP around 1.55 in the long-run (Annex Table 6). - REER coefficient ranges from 0.19 to 0.42; therefore 10 percent real appreciation increases imports by almost 2-4 percent across specifications. - Value-added in exports (OECD-WTO): - India’s domestic value-added ratio averaged 82.5 percent during 1995-2009. - India’s ratio declined by 12.3 percentage points between 1995 and 2009. - Lower imported intermediate content in India’s exports suggests exchange rate channel may be effective. ### X. Determinants of gold imports (Annex 4 summary and findings) - Stylized facts: - Gold imports jumped from monthly average 62 tonnes during 2004-08 to 82 tonnes during 2009-13. - US$ terms: from US $1.2 billion per month to US $4.1 billion per month. - Empirical findings (Annex Table 7, selected coefficients and interpretation): - Higher gold prices have a significant dampening impact in the first few months and a lagged positive impact in fourth/fifth month; cumulative impact negative. - One percent increase in CPI inflation leads to an increase of 9 percent in gold imports in the long-run (model interpretation). - Exchange rate depreciation reduces gold demand; effect generally statistically significant. - Higher domestic nominal deposit rates unexpectedly associated with more gold demand (real deposit rates generally negative in sample period). - Policy implication: - Success with inflation management and appropriate real returns on bank deposits should help reduce gold imports. ### XI. Fiscal composition, subsidies, and policy multipliers - Subsidies and fiscal impact: - Subsidies (cash outgo) increased from 1.6 percent of GDP in 2003-04 to 2.3 percent in 2012-13. - Oil subsidies increased from 0.1 percent of GDP during 2003-08 to 1.0 percent in 2012-13. - Bonds issued in lieu of subsidies increased from 0.5 percent of GDP in 2005-06 to 1.7 percent in 2008-09. - Cash subsidies increased by one percentage point of GDP in 2008-09; increase almost 2 percentage points when bonds included. - Expenditure composition and multipliers: - Revenue expenditure of central government increased from 11.9 percent of GDP in 2007-08 to 14.1 percent in 2008-09. - Capital outlays of the centre broadly stagnant at around 1.5 percent of GDP. - Empirical multiplier evidence (Jain and Kumar, 2013; Tapsoba, 2013): fiscal multipliers for government capital outlays exceed government consumption expenditure in the long-run; first-year multiplier for government consumption can be unity then turn negative; government investment has first-year and long-run multipliers more than unity. - Policy implication: higher capital outlays would have a more durable growth impact than revenue subsidies/tax cuts. ### XII. Capital account liberalization, debt flows, and macroprudential concerns - Liberalization steps and risks: - Increase in interest rate ceilings on foreign currency non-resident deposits; removal of ceiling on non-resident external rupee deposits; increase in local-currency investment limits by non-residents. - Opening government securities markets to non-resident investors poses volatility risks; example June-September 2013: foreign sell-off in debt securities around US $10 billion vs. equity US $2 billion. - Debt flows and CAD financed by portfolio and debt flows increase vulnerability. - Institutional and macroprudential framework: - Debate on central bank role and macroprudential mandates; IMF models summarized: - Model 1: Macroprudential mandate assigned to central bank Board. - Model 2: Macroprudential mandate to dedicated committee within central bank. - Model 3: Macroprudential mandate to committee outside central bank with central bank participation. - Recommendation: careful design to preserve synergy between monetary policy and financial stability; caution against narrowing central bank mandate solely to price stability. ### XIII. Fiscal consolidation, quality, and growth outlook - Need and composition of consolidation: - Fiscal consolidation necessary for sustained growth and stability; focus on shifting expenditure from revenue (subsidies) to capital outlays. - Estimates: fiscal multiplier for non-defense capital outlays 2.1 vs. 0.2 for revenue expenditure for the Central government (Jain and Kumar, 2013). - Suggested path: aim for FRBM target 3 percent GFD/GDP by 2017-18 and consider more ambitious medium-term consolidation; aim for balanced budgets by end of decade considered judicious. - Expected macro effects: - Fiscal consolidation and recovery in public savings expected to raise gross domestic savings rate by around 2-3 percent of GDP. - Durable reduction in inflation and low positive real interest rates would moderate gold imports and CAD. - Growth prospects: - Conservative sustained gross domestic saving rate of about 35 percent would facilitate growth of 8.5 percent, given moderate incremental capital output ratios, contingent on removal of impediments to domestic investment and recovery in global economy. - Projected potential growth: 8.5 percent, contingent on removal of investment impediments and global recovery. - Key policy priorities to restore sustained growth in excess of 8 percent: - Restore macroeconomic stability by reducing medium term inflation to pre-NAFC decade levels. - Coordinate supply-side and demand management: fiscal consolidation, appropriate monetary policy, active measures to loosen supply constraints. - Market reforms: labor and land reforms to increase flexibility. - Manufacturing revival: realistic exchange rate policy combined with labor flexibility and skill development to achieve 10 per cent plus manufacturing growth. - Infrastructure: enhance public investment; reinforce economic user charges and curtail non-merit subsidies. - Public-sector delivery and second-generation reforms: focus on agricultural development, urban development, human resource development, and management of public services. ### XIV. Key concise findings - Growth: - India’s growth rebounded after NAFC but decelerated sharply after 2011; high growth phase 2004-08 (9 percent plus) followed by slowing and concerns about lower potential growth. - Fiscal and external: - Centre GFD rose from 2.5 percent of GDP in 2007-08 to 6.0 percent in 2008-09; broader measure including in-kind bonds rose to 8.2 percent. - CAD widened to just under 5 percent in 2012-13 and corrected to 2.3 percent in first three quarters of 2013-14. - Inflation and monetary stance: - Inflation remained persistently high post-2008; monetary policy moved from rapid easing in late-2008/early-2009 to phased withdrawal during 2010-11 and 2011-12. - Financial sector: - Indian banking and financial markets showed resilience during the NAFC with limited direct exposure to toxic assets. - Policy emphasis: - Quality fiscal consolidation (shift to capital outlays), inflation control, repair of public savings, and removal of supply-side bottlenecks are priorities to restore investment and achieve sustained growth. *Source: _wp1468 - References (excerpted content supplied).* ### References ............................................................................................................. ### _wp1468 - References ............................................................................................................. ### I. Introduction - India recorded robust annual growth of 9 percent plus during 2004-08, accompanied by consolidation of key macroeconomic indicators. - The North Atlantic financial crisis (NAFC) in 2008 caused a setback: growth rebounded after large monetary and fiscal stimuli but slowed significantly subsequently. - Post-2008 the economy experienced a substantial widening of current account and fiscal deficits and elevated inflation. - Declines in domestic saving and investment rates raise concerns that India’s potential growth rate has fallen significantly (IMF, 2013; Mishra, 2013). - Concerns about a possible balance of payments crisis intensified during June-August 2013 after the US Federal Reserve’s mention of tapering; these concerns receded somewhat after a significant correction in the current account deficit since Q2 2013-14. - Alternative views: the 2004-09 high growth may have been a debt-led cyclical boom supported by unprecedented capital inflows coinciding with exceptional world growth (Nagaraj, 2013). ### II. Recent macroeconomic trends: India in a global perspective - After NAFC, India’s real GDP growth rebounded during 2009-11 but decelerated significantly in the following three years. - Deceleration accompanied by: - reversal of fiscal consolidation (fiscal deficit in 2013-14 remained well above pre-crisis levels), - CAD widening from less than 1 percent of GDP during 1992-2008 to just under 5 percent in 2012-13 (later halved to 2.3 percent in the first three quarters of 2013-14), - persistently high headline and consumer inflation post-crisis, - significant decline in private corporate investment. - Table 1 (Key Macroeconomic Indicators: 2003-14) highlights selected exact annual values: - 2003-04: Real GDP Growth (factor cost) 8.0; Real GDP Growth (market prices) 7.9; GFD/GDP (Centre) 4.3; CAB/GDP 2.3; Non-oil CAB/GDP 5.0; WPI Inflation 5.5; CPI Inflation 3.9; REER Index 96.8; Real Policy Rate -0.4 - 2004-05: Real GDP Growth (factor cost) 7.1; Real GDP Growth (market prices) 7.9; GFD/GDP (Centre) 3.9; CAB/GDP -0.3; Non-oil CAB/GDP 2.8; WPI Inflation 6.5; CPI Inflation 3.8; REER Index 99.9; Real Policy Rate -0.6 - 2005-06: Real GDP Growth (factor cost) 9.5; Real GDP Growth (market prices) 9.3; GFD/GDP (Centre) 4.0; CAB/GDP -1.2; Non-oil CAB/GDP 2.7; WPI Inflation 4.4; CPI Inflation 4.4; REER Index 102.7; Real Policy Rate 1.7 - 2006-07: Real GDP Growth (factor cost) 9.6; Real GDP Growth (market prices) 9.3; GFD/GDP (Centre) 3.3; CAB/GDP -1.0; Non-oil CAB/GDP 3.0; WPI Inflation 6.6; CPI Inflation 6.7; REER Index 101.0; Real Policy Rate 3.1 - 2007-08: Real GDP Growth (factor cost) 9.3; Real GDP Growth (market prices) 9.8; GFD/GDP (Centre) 2.5; CAB/GDP -1.3; Non-oil CAB/GDP 2.9; WPI Inflation 4.7; CPI Inflation 6.2; REER Index 108.6; Real Policy Rate 2.2 - 2008-09: Real GDP Growth (factor cost) 6.7; Real GDP Growth (market prices) 3.9; GFD/GDP (Centre) 6.0; CAB/GDP -2.3; Non-oil CAB/GDP 3.1; WPI Inflation 8.1; CPI Inflation 9.1; REER Index 97.8; Real Policy Rate 0.9 - 2009-10: Real GDP Growth (factor cost) 8.6; Real GDP Growth (market prices) 8.5; GFD/GDP (Centre) 6.5; CAB/GDP -2.8; Non-oil CAB/GDP 1.5; WPI Inflation 3.8; CPI Inflation 12.2; REER Index 95.3; Real Policy Rate 1.5 - 2010-11: Real GDP Growth (factor cost) 8.9; Real GDP Growth (market prices) 10.3; GFD/GDP (Centre) 4.8; CAB/GDP -2.7; Non-oil CAB/GDP 1.1; WPI Inflation 9.6; CPI Inflation 10.5; REER Index 103.5; Real Policy Rate 2.0 - 2011-12: Real GDP Growth (factor cost) 6.7; Real GDP Growth (market prices) 6.6; GFD/GDP (Centre) 5.7; CAB/GDP -4.2; Non-oil CAB/GDP 1.1; WPI Inflation 8.9; CPI Inflation 8.4; REER Index 100.7; Real Policy Rate 1.0 - 2012-13: Real GDP Growth (factor cost) 4.5; Real GDP Growth (market prices) 4.7; GFD/GDP (Centre) 4.9; CAB/GDP -4.7; Non-oil CAB/GDP 1.1; WPI Inflation 7.4; CPI Inflation 10.4; REER Index 96.3; Real Policy Rate 1.9 - 2013-14: Real GDP Growth (factor cost) 4.9; Real GDP Growth (market prices) 4.6; GFD/GDP (Centre) 4.6; CAB/GDP -2.3 (April-December) - Global context: - Global growth fell from an annual average of 4.8 percent during 2003-07 to an average of 2.9 percent during 2008-12. - Many advanced economies recorded real currency depreciations while EME currencies appreciated post-NAFC; many EMEs exhibited high CADs. - In 2012, among G-20 economies India had the third largest CAD after Turkey and South Africa; India’s fiscal deficit was the second largest after Japan. - World Bank estimates cited: potential growth for developing countries will be 5.5 percent for 2012-2015, compared with potential growth of 6.3 percent during 2005-07 and actual growth averaging 8.3 percent during that period. ### III. Domestic macroeconomic policies and growth slowdown - Part of the 2011-14 slowdown reflects withdrawal of large monetary and fiscal stimulus administered immediately after the crisis. - Following Lehman’s collapse in October 2008: - Large capital outflows from India occurred due to foreign institutional investors selling in domestic stock market. - No direct impact on Indian banking system due to limited exposure to toxic assets, reflecting a prudent regulatory framework. - Indian financial markets continued to function, though with elevated volatility. - Stimulus measures after the 2008 downturn: - Monetary and liquidity stimulus between September 2008 and April 2009 was large; effective policy rate cut from 9.00 percent in September 2008 to 3.25 percent by April 2009. - Cash reserve ratio reduced from 9.0 percent to 5.0 percent over the same period. - Other measures collectively had potential to release liquidity of more than 10 percent of GDP (Mohan, 2011c). - Fiscal measures included cut in CENVAT rate from 14 percent to 8 percent between December 2008 and February 2009 and increased plan expenditure, in addition to Pay Commission implementation and agriculture debt waiver. - Central government headline gross fiscal deficit (GFD) increased from 2.5 percent of GDP in 2007-08 to 6.0 percent in 2008-09; including bonds issued in lieu of cash subsidies, GFD/GDP rose from 3.1 percent to 8.2 percent. - Net effect: both monetary and fiscal policy provided strong support—deemed excessive with hindsight—to domestic demand in 2008-09. ### Monetary policy (post-stimulus dynamics) - Stronger-than-expected growth in 2009-10 and 2010-11 translated into high inflation, initially food inflation by end-2009 and underlying inflation by April 2010. - Elevated international commodity prices and domestic structural imbalances in availability of select food items (pulses and other protein items) added to inflationary pressures. - Monetary accommodation continued until early 2010; subsequent withdrawal was phased and gradual during 2010-11 and 2011-12 due to: - high uncertainty about global and domestic outlook, - perception that initial high inflation was food-driven, - real-time data under-estimating domestic demand strength (Subbarao, 2011). - The gradual withdrawal contrasted with the rapid stimulus phase from September 2008 to April 2009. ### IV. External sector and vulnerabilities - CAD dynamics: - CAD averaged less than 1 percent of GDP during 1992-2008. - Widened to just under 5 percent in 2012-13. - Improved to 2.3 percent in the first three quarters of 2013-14 following policy actions. - Cross-country vulnerabilities: - Large twin deficits (fiscal and current account) pose sources of vulnerability; literature shows countries with high debt/GDP ratios and persistent CADs are vulnerable to rapid fiscal deterioration (Greenlaw et al., 2013). - Debt thresholds are typically lower for emerging economies (Reinhart and Rogoff, 2009). - External vulnerabilities and domestic credit booms have been central to crises in EMEs (Lane and Milesi-Ferretti, 2010). - The June-August 2013 episode evidenced vulnerabilities associated with large CADs in EMEs. ### V. Interpretation and factors shaping the slowdown - Two broad interpretations of the post-crisis slowdown: - Return-to-potential view: pre-NAFC period had global growth above potential; post-NAFC slowdown is a return to underlying potential growth, with potential itself lower than pre-crisis estimates. - Example: potential growth of developing countries was 6.3 percent during 2005-07; actual growth in that period averaged 8.3 percent; output gap reached 3.5 percent in 2007 (World Bank, 2013). World Bank projects potential growth for developing countries at 5.5 percent for 2012-2015. - Domestic factors view: the extent of India’s slowdown, elevated inflation, and magnitude of domestic imbalances suggest domestic factors have added to global headwinds. - Noted resilience: the domestic financial sector exhibited striking resilience to NAFC, reflecting a prudent approach to domestic and external financial liberalization (Mohan 2011). ### VI. Key findings (concise) - Growth: - India’s growth rebounded after NAFC but decelerated sharply after 2011. - High growth phase 2004-08 (9 percent plus) was followed by slowing and concerns about lower potential growth. - Fiscal and external: - Gross fiscal deficit (Centre) rose from 2.5 percent of GDP in 2007-08 to 6.0 percent in 2008-09; broader measure including in-kind bonds rose to 8.2 percent. - CAD widened to just under 5 percent in 2012-13 and corrected to 2.3 percent in first three quarters of 2013-14. - Inflation and monetary stance: - Inflation remained persistently high post-2008, unlike many economies where inflation moderated. - Monetary policy moved from rapid easing in late-2008/early-2009 to a phased withdrawal during 2010-11 and 2011-12. - External context: - Global growth slowdown and currency movements post-NAFC contributed to India’s slowdown and CAD widening, but domestic policy choices and structural issues were important additional factors. - Financial sector: - Indian banking and financial markets showed resilience during the NAFC, with limited direct exposure to toxic assets. *Source: _wp1468 - References (excerpted content supplied).* ### 2009. The quantum and the pace of the monetary stimulus in that period was greater than in ### _wp1468 - 2009. The quantum and the pace of the monetary stimulus in that period was greater than in ### Monetary stimulus, capital flows, and macro outcomes - In the face of large and increasing capital flows – from 2.7 percent of GDP in 2003-04 to 8.6 percent in 2007-08 – the Reserve Bank deployed a range of instruments to manage these capital flows, including sterilized interventions. - Growth in broad money averaged more than 21 percent per annum during 2005-08. - Growth in non-food credit averaged 28 percent during 2005-08. - Real GDP growth averaged 9.5 percent during 2005-08. - Actual growth in monetary and credit aggregates was above the Reserve Bank’s indicative projections at the beginning of each financial year. - These indicators suggest signs of overheating in the pre-NAFC period; inflation indicators began increasing in 2007-08 and were compounded by increasing oil prices. - Stimulus measures adopted after the NAFC added to incipient inflationary pressures. ### Real GDP forecasts and actuals (selected indicators) - Table 3 (excerpted): various institutions’ forecasts and latest estimates for Real GDP Growth across fiscal years 2008-09 through 2012-13 (Professional Forecasters, PMEAC, RBI). (Table structure preserved in source.) ### Food inflation and Minimum Support Prices (MSP) - Large increases in MSP for various crops occurred in recent years, especially since 2008-09, but causation is arguable. - For pulses: actual prices (WPI) rose substantially during 2005-06 and 2006-07 while MSP rose only moderately; MSP was increased in 2007-08 and substantially in 2008-09. - Cumulative variation in MSP between March 2005 and March 2010 trailed cumulative variation in actual prices for arhar, moong, urad, and wheat (Table 5). - Plausible drivers of the pulse price increases beginning 2005-06: - Strong domestic growth from 2003-04 onwards amid near plateauing of domestic production of pulses. - Shift in dietary patterns toward protein-rich items with higher incomes. - Support from the MGNREGA scheme. - Increases in MSP can be viewed as government attempts to incentivize higher domestic production of pulses to meet rising demand. - Higher-order initial increases in MSP may have been necessitated by higher food demand on the back of high growth and rising incomes. - Non-food non-fuel CPI inflation (rural and urban combined based on the new CPI series) has been around 8 percent since June 2012 (it was around 10 percent during January-May 2012), indicative of generalized pressures. - Monetary policy can keep overall inflation within its comfort zone in the face of persistently high food inflation, but doing so would involve excessive tightening and large output costs for other sectors. - MGNREGA may have contributed to rural bargaining power but accounts for only a small fraction of rural wage increase and the effect is waning. - Productivity gains in food production are suggested as a more durable, non-inflationary solution to increase food production. - Higher outlays on MGNREGA and higher food subsidy bills are ultimately reflected in revenue deficits, which add to domestic demand. ### Minimum Support Prices and WPI (selected series, 2003=100, Table 5) - Rice WPI: Mar-03 100.0; Mar-04 97.7; Mar-05 101.1; Mar-06 104.0; Mar-07 113.4; Mar-08 129.7; Mar-09 149.7; Mar-10 161.7; Mar-11 165.4; Mar-12 173.7; Mar-13 204.2 - Rice MSP: Mar-03 100.0; Mar-04 103.8; Mar-05 105.7; Mar-06 107.5; Mar-07 109.4; Mar-08 121.7; Mar-09 169.8; Mar-10 188.7; Mar-11 188.7; Mar-12 203.8; Mar-13 235.8 - The row 'Gap' gives difference (percent) of MSP over WPI (Table 5 entries preserved). - (Table 5 includes analogous WPI, MSP, and Gap series for Wheat, Gram, Arhar, Moong, Urad; values preserved in source.) ### Real interest rates — borrowers - Real lending rates assessed using headline WPI inflation and non-food manufactured products (NFMP) WPI inflation; a 12-month moving average of y-o-y inflation is used as an indicator of inflation expectations in parts of the analysis. - Using 12-month moving average and NFMP inflation: - Real policy rate fell from an average of 2.2 percent in 2007-08 to 0.9 percent in 2008-09. - Real policy rate edged up to 2.0 per cent in 2010-11. - Real policy rate fell back to an average of 1.0 percent in 2011-12 (reflecting higher-than-expected increase in NFMP inflation). - Real policy rate edged higher to 1.9 percent in 2012-13 (on the back of higher policy rate and some moderation in NFMP inflation). - The real policy rate in terms of CPI inflation has been generally negative since mid-2008 (RBI, 2014). - While nominal policy rate increased gradually during 2010 and 2011, the real policy rate was highly negative (w.r.t. headline WPI) and marginally negative (w.r.t. NFMP), implying monetary policy was still accommodative over this phase despite commentary that it was too tight. - Real commercial paper rates increased during 2012 and were higher than in the pre-crisis period, especially in terms of core inflation. - Commercial bank lending rates movements are broadly similar to commercial paper trends; assessment complicated by shift from benchmark prime lending rate to base rate system in July 2010. - A 100 bps increase in the policy interest rate is associated, on average, with a growth slowdown of 25-50 bps (econometric evidence cited). The actual recent slowdown was much greater than suggested by these estimates. ### Corporate sector performance - Corporate savings fell from 9.4 percent of GDP in 2007-08 to 7.1 percent in 2012-13. - Corporate investment fell from 17.3 percent of GDP in 2007-08 to 9.2 percent in 2012-13 (Table 6). - Possible explanations for larger decline in corporate investment vis-à-vis savings since 2007-08: - Policy bottlenecks (environmental permissions, fuel linkages, land acquisition) led to stalling of large projects and discouraged new investment. - Large increase in fiscal deficit and near-trebling of government borrowing requirements appears to have led to some crowding out of the private sector. - Outward FDI by Indian corporates increased from 0.3 percent of GDP in 2003-04 to 1.5 percent in 2007-08, but since then outward FDI fell to 0.4 percent of GDP in 2012-13, aligned with declining domestic investment; decline in domestic investment since 2007-08 cannot be attributed to greater investment abroad. - Charted indicators (from source): Gross Profit to Sales, Profit After Tax to Sales, Interest to Sales, Interest to Gross Profits (CEIC data referenced). ### Savings, investment, and household behavior (Table 6 highlights) - Household sector total savings (percent of GDP): 2003-04 23.1; 2004-05 23.6; 2005-06 23.5; 2006-07 23.2; 2007-08 22.4; 2008-09 23.6; 2009-10 25.2; 2010-11 23.1; 2011-12 22.8; 2012-13 21.9 - Household financial saving (gross) (percent of GDP): 2003-04 11.0; 2004-05 10.1; 2005-06 11.9; 2006-07 11.3; 2007-08 11.6; 2008-09 10.1; 2009-10 12.0; 2010-11 9.9; 2011-12 7.0; 2012-13 7.1 - Financial assets and liabilities, and composition (bank deposits, etc.) preserved in Table 6. - Gross capital formation (percent of GDP): 2003-04 26.1; 2004-05 32.5; 2005-06 34.4; 2006-07 35.9; 2007-08 38.0; 2008-09 35.5; 2009-10 36.3; 2010-11 36.5; 2011-12 35.5; 2012-13 34.7 - Saving-Investment Balance, net (4-9): 2003-04 2.2; 2004-05 -0.4; 2005-06 -1.2; 2006-07 -1.1; 2007-08 -1.3; 2008-09 -2.3; 2009-10 -2.8; 2010-11 -2.8; 2011-12 -4.2; 2012-13 -4.7 - Household financial savings (net) available for private corporate sector (10a + 10c): 2003-04 5.7; 2004-05 5.0; 2005-06 6.3; 2006-07 6.5; 2007-08 7.8; 2008-09 1.7; 2009-10 3.0; 2010-11 4.1; 2011-12 0.5; 2012-13 0.2 - Decline in household financial savings between 2007-08 and 2012-13: from 15.5 percent of GDP in 2007-08 to 10.8 percent in 2012-13 (Table 6 narrative). - Households’ physical investment in gold increased from an average of 1.1 percent of GDP during 2003-08 to 2.6 percent by 2012-13. - Overall household savings remained around 21.9 percent of GDP in 2012-13, almost the same as in 2007-08, despite the economic slowdown. ### Real interest rates — depositors - Real deposit rates (nominal deposit rate less y-o-y CPI (Industrial Workers) inflation) have been broadly negative since 2008-09 due to persistently elevated consumer inflation driven by high food inflation. - Real deposit rates fell from an average of (+) 1.5 percent during 2003-08 to (-) 1.9 percent during 2008-13. - Using households’ inflation expectations (available from 2006 onwards), real deposit rate averaged (-) 2.1 percent using households’ “current” inflation expectations and (-) 3.3 percent using households’ “one-year ahead” inflation expectations during 2008-13. - Administered interest rates on small savings have also been negative in real terms in recent years and growth in small savings has been low or negative in this period. - Negative real deposit rates, along with the growth slowdown, appear to have contributed to the decline in household financial savings and a switch toward physical assets (gold and property). - Historical context: household savings increased between 1997-98 (18 percent of GDP) and 2003-04 (23 percent of GDP) even as nominal deposit rates fell (from ~11 percent to ~5-6 percent), but that period saw a significant fall in inflation and real deposit interest rates were positive. ### Industrial data divergence and implications for policy - Divergence between Index of Industrial Production (IIP) and Annual Survey of Industries (ASI): IIP growth rates were significantly lower than ASI growth rates in most years (PMEAC, 2013). - During 2003-12, IIP growth averaged almost 4 percentage points lower than real growth of gross value added from ASI data, with pronounced difference in 2011-12. - Because IIP is available at monthly frequency and informs short-term policy, substantial revisions or measurement problems in IIP can lead to incorrect policy inferences and actions. - Priority recommended: understand and reconcile differences between IIP and ASI, and correct problems in collecting IIP data. ### Fiscal policy, stimulus, and borrowing costs - Centre’s headline fiscal deficit/GDP ratio more than doubled to 6.0 percent in 2008-09 (Table 8). - Deficit including bonds issued in lieu of cash subsidies rose from 3.1 percent of GDP in 2007-08 to 8.2 percent in 2008-09, moderating to 6.6 percent in 2009-10. - Centre’s fiscal deficit in nominal terms: Rs.1,269 billion in 2007-08; Rs.3,370 billion in 2008-09 (budgeted amount was Rs.1,333 billion); Rs.4,185 billion in 2009-10 — an increase of 230 percent in 2 years (2007-08 to 2009-10). - Despite the increase in borrowing requirements, weighted average yield on Central government’s dated securities fell from 8.12 percent in 2007-08 to 7.23 percent in 2009-10, helped by accommodative monetary policy and large open market operations by the Reserve Bank. - Fiscal stimulus began to be withdrawn in 2011-12 and 2012-13. - Quality of fiscal stimulus in aftermath of the NAFC appears to have exacerbated the slowdown in 2011-13: revenue expenditure of central government increased from 11.9 percent of GDP in 2007-08 to 14.1 percent in 2008-09 (and maintained at this level in 2009-10); part of this increase was on account of subsidies. *Italicized source attribution: Content drawn from the supplied IMF PDF chapter/section.* ### 2.3 percent in 2008-09 and remained around this level till 2012-13. The increase in subsidies ### _wp1468 - 2.3 percent in 2008-09 and remained around this level till 2012-13. The increase in subsidies ### Subsidies and fiscal impact - Subsidies (cash outgo) increased from 1.6 percent of GDP in 2003-04 to 2.3 percent in 2012-13 (Table 8). - Oil subsidies increased from 0.1 percent of GDP during 2003-08 to 1.0 percent in 2012-13. - Bonds issued in lieu of subsidies to oil companies, fertilizer companies and Food Corporation of India increased from 0.5 percent of GDP in 2005-06 to 1.7 percent in 2008-09. - Cash subsidies increased by one percentage point of GDP in 2008-09; the increase was almost 2 percentage points once bonds issued in lieu of cash subsidies are included. - Note: Data on Central government finances for 2013-14 pertain to revised estimates. Data on combined government finances for 2012-13 and 2013-14 pertain to revised and budget estimates, respectively. Figures in parentheses are GFD including off-budget liabilities. ### Oil demand elasticities, under-recoveries, and impact on CAD - Estimated price elasticities of demand: - Petrol: (-) 0.66 - Diesel: (-) 0.36 - Kerosene oil: (-) 0.54 - Diesel accounts for almost 45 percent of domestic petroleum consumption. - Under-recoveries for diesel in 2011-12 and 2012-13 are estimated at around Rs.11 per liter (around 25 percent of the actual prevailing prices). - Total domestic consumption of diesel: 65 million tonnes (MT) in 2011-12 and 69 MT in 2012-13; corresponding under-recoveries were Rs.812 billion and Rs.921 billion. Thus, the under-recovery was around Rs.13 per kg or Rs. 11 per liter. - If diesel prices had been raised to eliminate under-recoveries, the estimated price elasticity of 0.36 implies diesel consumption would have been around 9 percent lower. - Ceteris paribus, this reduction in diesel consumption would have lowered overall imports and the current account deficit by around 0.5 percent of GDP each in 2011-12 and 2012-13. ### Fiscal composition, expenditure quality, and multipliers - Revenue expenditure trended upward; capital outlays of the centre were broadly stagnant at around 1.5 percent of GDP over this period (Table 8). - Empirical multiplier evidence: - Fiscal multipliers for government capital outlays exceed government consumption expenditure in the long-run in India (Jain and Kumar, 2013). - Tapsoba (2013): fiscal multiplier for government consumption is unity in the first year, then turns negative; long-run impact negative. First-year and long-run multipliers for government investment are more than unity. - Policy implication: Higher revenue expenditures provided only short-lived boost to activity; higher capital outlays would have had a more durable impact. - Stimulus measures focused on revenue expenditure/tax cuts with stagnant capital outlays, imparting volatility to the growth path. ### Tax revenue and indirect taxes - Revenue receipts: 9.3, 9.4, 9.4, 10.1, 10.9, 9.6, 8.8, 10.1, 8.3, 8.7, 9.1 (percent to GDP for 2003-04 through 2013-14 as in Table 8). - Gross tax (percent to GDP): 9.0, 9.4, 9.9, 11.0, 11.9, 10.8, 9.6, 10.2, 9.9, 10.2, 10.2. - Direct taxes (income and corporation) ratio to GDP broadly unchanged from 2007-08; decline in tax/GDP ratio driven by indirect taxes, especially excise collections. - CENVAT rate movements: increased to 10 per cent in the Union Budget 2010–11 (February 2011) and to 12 percent in the Union Budget 2011-12 (March 2012), still below pre-NAFC level of 14 percent. ### Saving-Investment balance and crowding out - Public savings deteriorated from 5 percent of GDP in 2007-08 to 1.2 percent in 2012-13. - Overall savings rate fell from 36.8 percent to 30.1 percent, with large chunk of decline occurring in 2011-13. - Public and private corporate investment fell by almost 9 percentage points between 2007-08 and 2012-13. - Household investment increased (reflecting physical savings in gold and property), cushioning overall investment rate decline from 38.1 percent to 34.8 percent. - Net financial savings: - Household sector net financial savings fell from 11.6 percent of GDP in 2007-08 to 7.1 percent in 2012-13. - Public sector net S-I deficit increased from (-) 3.9 percent to (-) 6.9 percent over this period. - Net financial savings available to private corporate sector fell from an average of 6.3 percent of GDP during 2003-08 to 1.9 percent during 2008-13, indicating significant crowding out of the private sector in the post-NAFC period. - The increase in the public S-I gap is attributable to the decline in public savings and a large part owes to the increase in subsidies, especially oil; higher government subsidies were not a response to lower corporate investment. ### Channels and counterarguments on crowding out - Counter-argument: higher public S-I gap reflected government responding to collapse in corporate sector investment. Text argues this is not supported because public investment rate declined from its 2007-08 level. - Reasons crowding out could occur without higher domestic interest rates: - Monetary policy was accommodative initially; de facto monetization of fiscal deficit to the extent of 28 percent of overall borrowing requirement on average via injections of primary liquidity through OMOs (RBI, 2014, para III.53). - Economy open to capital flows: net capital flows averaged 4.2 percent of GDP during 2011-13 vs. 4.6 percent during 2003-08, even as domestic growth slowed from 8.7 percent to 5.6 percent. - Unconventional monetary policies in advanced economies led to record low global interest rates, constraining domestic interest rate increases. ### Quantifying the growth slowdown - Annex 2 empirical results: - An increase of 100 bps in the real interest rate leads to a reduction of around 30 bps in GDP growth (non-agricultural non-community services GDP) with a lag of two quarters (Annex Table 2). - A one percentage point reduction in global exports reduces domestic growth by almost 13 bps. - A one percentage point reduction in global GDP growth reduces domestic growth by almost 50 bps. - The estimated equations track actual growth relatively well for the 2003-08 period, but deviations are observed since then; the estimated equation under-predicts the actual outturn during 2009-10 by around (text truncated). *Source: Reserve Bank of India; Union Budget documents; and, Economic Advisory Council to the Prime Minister.* ### 3.5 percentage points and over-predicts by around 3 percentage points on average during 2011- ### _wp1468 - 3.5 percentage points and over-predicts by around 3 percentage points on average during 2011- ### Growth deviations and fiscal policy - The model includes monetary policy and global demand impacts; a potential reason for the deviation since 2009 is the role of fiscal policy, which has not been included in the explanatory variables. - India’s structural primary deficit increased by 4.0 percentage points of GDP in 2008-09. - The withdrawal of the stimulus was 1.4 percentage points during the 3-year period 2010-13, or an annual average of 0.35 percentage point of GDP. - Assuming a fiscal multiplier for India similar to other EMEs (around 0.5), the stimulus of 4 percent of GDP added almost 2 percentage points to growth in 2009-10. - The annual fiscal consolidation of 0.35 percent of GDP during 2010-11 to 2012-13, combined with a multiplier of 0.5, would suggest an adverse impact of less than 20 bps per annum on growth. - Even if the fiscal multiplier is assumed at unity, the annual growth impact would be around 35 bps, abstracting from lags. - Conclusion: fiscal stimulus can largely explain high growth in 2009-10; subsequent consolidation explains only a very modest part of the slowdown. Policy bottlenecks (environmental permissions, fuel linkages, land acquisition) likely contributed to stalling large projects and discouraged new investment. ### The external sector: current account (overview) - Large current account deficits in the 1980s averaged around 2 percent of GDP. - India’s current account deficit averaged 0.6 percent of GDP during 1991-2 to 2007-08. - Widening of the current account deficit to 4.7 percent of GDP in 2012-13. - Net oil imports rose from 4.1 percent of GDP in 2007-08 to 5.9 percent by 2012-13. - Imports of coal increased from around 0.5 percent of GDP during 2004-08 to 0.9 percent in 2011-12. - Gold imports increased from around 1.5 percent of GDP during 2004-08 to 3.0 percent in 2011-12 and 2012-13. - The non-oil non-gold current account balance remained around a surplus of around 4 percent between 2007-08 and 2012-13. - The non-oil non-gold trade balance remained broadly around a deficit of around 2 percent between 2007-08 and 2012-13. - The deterioration in overall trade balance and non-oil non-gold trade balance began well before the NAFC. ### Balance of Payments: key figures (selected entries from Table 9; percent of GDP unless noted) - Merchandise exports: 2003-04 10.7; 2007-08 13.4; 2011-12 16.5; 2012-13 16.5 - Oil component of exports: 2003-04 0.6; 2012-13 3.2 - Merchandise imports: 2003-04 12.9; 2007-08 20.8; 2011-12 26.6; 2012-13 27.0 - Oil component of imports: 2003-04 3.3; 2007-08 6.4; 2012-13 9.1 - Gold component of imports: 2003-04 1.1; 2011-12 3.0; 2012-13 2.9 - Non-oil non-gold imports: 2003-04 8.6; 2007-08 15.8; 2012-13 15.0 - Net oil imports: 2003-04 2.7; 2007-08 4.1; 2011-12 5.3; 2012-13 5.9 - Trade balance: 2003-04 -2.2; 2007-08 -7.4; 2011-12 -10.1; 2012-13 -10.5 - Non-oil balance: 2003-04 0.5; 2007-08 -3.2; 2012-13 -4.6 - Non-oil non-gold balance: 2003-04 1.6; 2007-08 -1.9; 2012-13 -1.8 - Invisibles, net: 2003-04 4.5; 2007-08 6.1; 2012-13 5.8 - Services, net: 2003-04 1.6; 2012-13 3.5 - Private transfers, net: 2003-04 3.5; 2012-13 3.5 - Investment income, net: 2003-04 -0.6; 2012-13 -1.2 - Current account balance: 2003-04 2.3; 2007-08 -1.3; 2011-12 -4.2; 2012-13 -4.7 - Non-oil current account balance: 2003-04 5.0; 2012-13 1.1 - Non-oil non-gold current account balance: 2003-04 6.1; 2012-13 4.0 - Capital flows, net: 2003-04 2.7; 2007-08 8.6; 2011-12 3.6; 2012-13 4.8 - Capital inflows: 2003-04 12.3; 2012-13 25.4 - Capital outflows: 2003-04 9.6; 2012-13 20.6 - Foreign investment, net: 2003-04 2.2; 2008-09 0.7; 2012-13 2.5 - FDI inward: 2003-04 0.7; 2008-09 3.4; 2012-13 1.5 - Portfolio: 2003-04 1.8; 2008-09 -1.1; 2012-13 1.5 - Debt flows, net: 2003-04 -0.1; 2007-08 3.3; 2012-13 1.7 - External assistance: 2003-04 -0.4; 2012-13 0.1 - External commercial borrowings: 2003-04 -0.5; 2007-08 1.8; 2012-13 0.5 - Short-term trade credits: 2003-04 0.2; 2012-13 1.2 - Non-resident deposits: 2003-04 0.6; 2012-13 0.8 - Overall balance: 2003-04 5.1; 2007-08 7.4; 2011-12 -0.7; 2012-13 0.2 - Memo (US $ billion): - Current account balance: 2003-04 14; 2007-08 -16; 2011-12 -78; 2012-13 -88 - Capital flows, net: 2003-04 17; 2007-08 77; 2011-12 62; 2012-13 89 - End-March external debt (US $ billion): 2003-04 17.2; 2007-08 18.0; 2012-13 21.0 - Short-term debt (residual maturity): 2003-04 1.6; 2012-13 6.6 - Import cover of forex reserves - goods and services (months): 2003-04 14.0; 2012-13 6.0 - Forex reserves/Short-term debt by residual maturity (Percent): 2003-04 107; 2012-13 37 - Foreign exchange reserves (US $ billion): 2003-04 113; 2007-08 305; 2012-13 294 ### Drivers of the widened CAD and financing risks - Factors contributing to widening CAD: - Sluggish global growth since 2009 reducing India’s export markets. - Elevated international commodity prices, particularly affecting a net oil importer. - Domestic supply and policy constraints increasing imports (e.g., coal) and reducing exports (e.g., iron ore mining restrictions). - Large increases in gold imports driven by high domestic inflation, negative real deposit rates, and international gold price gains. - Real exchange rate movements (see next subsection). - Financing composition and risks: - Inward FDI flows fell from 3.4 percent of GDP in 2008-09 to 1.5 percent in 2012-13. - Portfolio flows and debt flows financed almost two-thirds of the CAD in 2012-13. - Stock of external commercial borrowings rose from around US 30 billion (5-7 percent of GDP) during 2000-04 to US $115 billion (9.2 percent of GDP) in 2007-08 and to US $225 billion (12.2 percent of GDP) in 2012-13. - Drivers of increased external borrowings include unconventional monetary policies in advanced economies, phased liberalization of ceilings on external borrowings and non-resident investments, and possible crowding out of private sector by high government borrowings since the NAFC. ### Real exchange rate (REER) and trade competitiveness - Divergences across REER measures between March 2008 and March 2013: - RBI index: real depreciation of 12 percent. - BIS index: real depreciation of 4 percent. - OECD and IMF indices: real appreciation of 12-14 percent. - Measurement differences arise because: - RBI index uses WPI inflation for India and CPI inflation for partner countries. - OECD/IMF measures use CPI inflation for all countries. - Indian CPI inflation has been higher than WPI inflation in recent years, causing OECD/IMF indices to show higher real appreciation. - Literature note: WPI- and ULC-based REERs may be better indicators of price competitiveness than CPI-based measures (Bayoumi, Harmsen and Turunen, 2011; Chinn, 2006). - Debate on exchange rate policy and CAD: - PMEAC (2013) notes that allowing the currency to adjust downward may not produce expected import/export responses because: - Large gold and oil imports are relatively insensitive to exchange rate movements. - Exports integrated in supply chains may see depreciation neutralized by price re-negotiations. - Example cited: a substantial depreciation of 20 percent in the external value of the rupee did not boost exports, while China registered strong dollar export growth in late 2012 and early 2013. *Source: Reserve Bank of India; Ministry of Finance.* ### 2013. The Chinese yuan appreciated against the dollar over the past year by about 3 per cent” ### _wp1468 - 2013. The Chinese yuan appreciated against the dollar over the past year by about 3 per cent” ### Exchange rate movements and export performance - The rupee depreciated by more than 20 percent in nominal terms between mid-2011 and end-2012; a large part of this offset higher inflation differentials. - Based on US dollar terms data, India’s exports performed poorly vis-à-vis China in 2012, but a medium-term perspective is appropriate given lags with which exchange rate movements impact trade volumes. - India’s average export growth since the NAFC is comparable to that of China, whereas it was much lower in the pre-NAFC period (Table 10). - REER movements (BIS data basis): - Indian REER recorded an average annual depreciation of 1.0 percent during 2008-12. - Chinese currency recorded an annual average appreciation of 4.1 percent during 2008-12. - Conclusion: Given common global factors, the REER movements favored India in the post-NAFC period and this appears reflected in relative export performance. - Caveats and complementary factors: - Weak domestic supply response due to domestic rigidities and adverse investment climate could have offset expected impact of depreciation. - Single-equation export demand estimates implicitly assume elastic export supplies; if supplies are inelastic, depreciation effects are dampened. - The appropriate counterfactual to judge efficacy of exchange rate is the path of exports had there been no real depreciation in 2012. - Other policies—taxation changes, government efforts towards product- and market-diversification—also play important roles. ### Value added in exports (OECD-WTO data insights) - The imported intermediate content of exports can dampen competitive effects of depreciations; the OECD-WTO data provide domestic value-added embodied in exports as per cent of exports. - Interpretation: Lower domestic value-added ratio → higher foreign content → higher importance of imports to exports. - Aggregate findings for India: - The value added ratio for India averaged 82.5 percent during 1995-2009 (averaged over the five data points: 1995, 2000, 2005, 2008, and 2009). - India’s ratio was above that of European countries and emerging Asian countries such as China, Korea, Malaysia, Philippines, and Thailand (Table 11). - Between 1995 and 2009, the value added ratio declined for India by 12.3 percentage points; declines for China and Korea were 20.8 and 16.9 percentage points, respectively. - Sectoral patterns: - For most manufacturing industries (food products; ‘textiles, leather and footwear’; chemicals; metals; machinery and equipment; electrical equipment; transport equipment), domestic value-added ratio is higher for India vis-à-vis other major economies. - Exception: residual manufacturing group (‘manufacturing nec; recycling’) where India lags; ratio declined from 85.7 percent in 1995 to 50.9 percent in 2008 and 2009—possibly reflecting higher exports of oil products and increased crude oil imports due to domestic refining capacity. - Services exports: India’s value-added ratios are relatively lower than other economies (though ratios for services are higher than for manufacturing activities). - Implication: Given relatively lower imported content of India’s export inputs, the hypothesis that exchange rate depreciation is ineffective because of high imported content is unlikely to bind in cross-country perspective. If exchange rate channel is effective elsewhere, data suggest potential efficacy in India as well. ### Price and income elasticities: cross-country evidence - Cross-country analysis finds significant impacts of prices and demand conditions on exports and imports. - Representative long-run elasticities (Senhadji, 1998; Senhadji and Montenegro, 1999) for a sample of developing and industrial countries: - Price elasticity: (-) 1.0 (mean) - Income elasticity: 1.5 (mean) - Reported minima and maxima vary; studies summarized in Table 12 show heterogeneity. - Firm-level evidence highlights heterogeneity: - Berman, Martin and Mayer (2012) using French firm-level data find average export elasticity of (-) 0.4 with respect to the exchange rate; high-performance firms increase markup and raise export volumes less in response to depreciation. - Tang and Zhang (2012) estimate an exchange rate elasticity of (-) 0.4 for exports using Chinese firm-level data. - Broader findings: - Currency undervaluation stimulates economic growth, particularly for developing countries (Rodrik, 2008). - Export surges in developing countries tend to be preceded by large real depreciation associated with resource reallocation into exports (Freund and Pierola, 2012). - Real appreciation leads to deterioration in current account balance, savings and exports; impact is more pronounced in developing countries (Kappler et al, 2013). - Hysteresis: appreciation can cause persistent loss of export markets even if exchange rate later returns to neutral (Baldwin and Krugman, 1989). - Summary view (Auboin and Ruta, 2013): currency undervaluation generally has positive short-term impact on exports; persistence of effects appears limited to developing countries. Price elasticities are statistically significant but magnitudes differ and are sensitive to variable choice. Exchange rate is one among many factors affecting export performance; sector-specific policies and industrial policy are often critical (China 2000-2007 noted as example). ### Recent Indian empirical evidence and interpretation - Recent studies show mixed evidence on efficacy of exchange rate for Indian exports (Rangarajan and Mishra, 2013); studies in 1990s generally found higher elasticities. - Selected study results: - Aziz and Chenoy (2012), quarterly 1996-2008: external demand (real GDP growth in partner countries) coefficient = 4.6 (positive, statistically significant); REER coefficient = (-) 0.6 (negative sign, statistically insignificant). No manufacturing sub-sector (including textiles, leather, gems and jewellery) showed statistically significant sensitivity to exchange rate. - IMF (2012): long-run REER elasticity on exports = (-) 0.1 for 1982-2011; (-) 0.2 for post-1990s period. Corresponding long-run elasticities on external demand = 2.9 and 2.2 for respective periods. - Reasons why studies may find weak export–exchange rate elasticities: - Difficulty estimating elasticities with macro data due to endogeneity between exports and exchange rates. - Macro equations do not allow elasticity to vary with position of aggregate supply curve. - Policy recommendation distilled from literature: ensure the rupee does not appreciate in real terms and further worsen the trade balance, while factoring potential impact of depreciation on capital flows and balance sheets of corporate and financial institutions. ### Re-estimation of export and import demand (study’s empirical results) - The authors revisit issue and estimate export and import demand equations (Annex 3). - Annual estimates (1980-81 to 2007-08): - An increase of 1 percent in world real exports is associated with an increase of 1.1-1.4 percent in India’s exports in the long-run. - Real exchange rate effect: 10 percent real appreciation leads to a reduction of almost 2.1-6.5 percent in the volume of exports in the long-run (Annex Table 3). - Post-reforms quarterly estimates (1996-97 to 2007-08): - Elasticity with respect to external demand: - 1.6-1.9 when world exports used as indicator of external demand. - 2.6-3.6 when world GDP used as indicator of external demand. - REER elasticity: - Absolute coefficient on REER is substantially higher than annual estimates. - Coefficient is more than unity for RBI’s REER and even higher for REERs of OECD and IMF. - Results are sensitive to inclusion of oil exports (Annex 3). *Source: _wp1468 - 2013. The Chinese yuan appreciated against the dollar over the past year by about 3 per cent” (PDF chapter/section content provided).* ### Annex Table 4). Third, the results for overall exports broadly hold for exports of the major ### _wp1468 - Annex Table 4). Third, the results for overall exports broadly hold for exports of the major ### Exports and imports: empirical findings - Exports: - Estimates show a stronger impact of global conditions – both demand and price factors – on India’s exports. - The phased opening of the Indian economy and domestic deregulation/liberalization result in a greater role for market forces. - Quarterly results are subject to a shorter sample period and associated estimation uncertainties. - Imports: - Long-run coefficients are on expected lines: real appreciation and stronger domestic activity both contribute to higher imports. - Income elasticity of imports with respect to domestic industrial GDP is around 1.55 in the long-run in all cases. - Implication: the volume of imports changes more than proportionally in relation to variations in domestic demand. - The coefficient on the REER ranges from 0.19 to 0.42 and is higher for the RBI-REER. - Therefore, 10 percent real appreciation increases imports by almost 2-4 percent across the various specifications (Annex Table 6). - Overall: - Both economic activity (home and abroad) and real exchange rate movements have the expected impact on exports and imports. - Real appreciation reduces exports and increases imports. - REER appreciation can lead to a widening of the trade and current account balance, and vice versa for real depreciation. ### Determinants of gold imports (empirical results and interpretation) - Stylized facts: - Gold imports jumped from a monthly average of 62 tonnes during 2004-08 to 82 tonnes during 2009-13. - Increase in US dollar terms: from US $ 1.2 billion per month to US $ 4.1 billion per month over the same period. - Potential contributors examined: high domestic inflation, low domestic deposit rates, availability of bank loans for gold, substantial increase in international gold prices, movements in the Indian rupee, pent-up demand, stock prices, and real estate (real estate excluded due to lack of time-series). - Econometric results summary (Annex Table 7): - Higher gold prices (nominal and real) have a significant dampening impact on gold imports in the first few months, but a lagged positive impact on demand in the fourth/fifth month. - The cumulative impact of gold price increases on gold imports is negative and statistically significant. - Channels identified: - Price increase reduces immediate quantity demanded (normal good effect). - Recent price increases may increase investment interest if interpreted as signaling further increases (Starr and Tran channel); both channels present but first predominates. - Higher domestic CPI inflation has a significant impact on gold imports; no impact found for WPI inflation. - Estimates show that one percent increase in CPI inflation leads to an increase of 9 percent in gold imports in the long-run. - Higher domestic deposit rates, unexpectedly, lead to more gold demand (possible explanation: real deposit rates were generally negative over the last four years of the sample; nominal deposit rate rises did not produce expected dampening because real return remained negative). - Stock market gains lead to higher gold demand with a lag (possibly wealth diversification motive). - Exchange rate depreciation reduces gold demand in all specifications and the effect is generally statistically significant; the model also includes gold prices in rupee terms and exchange rate depreciation has an additional direct negative impact. - Model performance and limits: - Model captures various turning points well but has difficulties explaining the increase in imports in 2010-11 (Chart 8); possible role of volatility and uncertain environment post-NAFC. - Policy implications from gold-import findings: - Higher domestic CPI inflation and continued negative real deposit rates could have contributed to higher gold imports, adding to external vulnerability and reducing domestic financial savings. - Success with inflation management and appropriate real returns on bank deposits should provide a durable solution to the surge in gold imports. - Tarapore (2013b) noted curbing gold imports requires very attractive instruments better than the return on gold; example given: a 3 per cent real rate plus consumer price inflation of, say, 9 per cent would yield a nominal return of 12 per cent plus inflation adjustment for the capital and such an instrument would knock down demand for gold. As the CAD and inflation come down, the cost of such an instrument would also come down. ### External sector and current account deficit (CAD) - Sustainable CAD: - Level of CAD during 2012-13 and early 2013-14 was well-above estimates of a sustainable level, which is in the range of 2.3-2.5 percent of GDP (Rangarajan and Mishra, 2013; RBI, 2012). - Sustainable CAD estimates are an upper limit and rely on stabilization of external liabilities/GDP ratio; financing volatility makes even sustainable levels challenging. - Even a 2.3 percent CAD requires net annual capital inflows of US $ 50-70 billion at present GDP levels (Rangarajan and Mishra). - Oil and gold price volatility: - Large oil and gold imports and fluctuations in international prices mean sustainable deficit needs reassessment; plan for lower deficit to contain vulnerability from price shocks. - Proposal of Dr. Y.V. Reddy to aim for an average current account balance of zero merits further consideration. - Role of foreign savings: - Given large domestic savings, foreign savings play only a marginal role in investment and growth. - A zero average or low current account deficit is not an argument against current account openness; openness to gross imports/exports and FDI matters for competition, productivity, investment and growth. - Empirical drivers of external balance: - Exports and imports respond significantly to domestic and foreign income variables and to real exchange rate dynamics. - Income elasticities are greater than price elasticities; growth dynamics at home and abroad are key driving forces. - Sustained higher domestic inflation over the past 4-5 years had a negative impact on external competitiveness through a higher REER. - Low and stable inflation aids maintenance of external competitiveness and reduces pressures for nominal depreciation. ### Capital account liberalization, external vulnerability, and debt flows - Recent liberalization steps (to meet financing needs) included: - Increase in interest rate ceilings on foreign currency non-resident deposits. - Removal of interest rate ceiling on non-resident external rupee deposits and capping them at the same level as domestic rupee deposits. - Increase in local-currency investment limits by non-residents (Table 13). - Consequences and risks: - External-sector sustainability indicators (import cover and short-term debt to reserves) recorded some deterioration (RBI, 2013b) though they remained at comfortable levels, adequacy questioned during June-August 2013. - Opening capital account to meet persistently high CAD potentially hinders adjustment: more foreign capital may lead to real appreciation, higher CAD, and enhanced external vulnerability (Panagariya, 2013). - Empirical evidence from the euro area supports this channel: abundant capital flows, high credit growth and real appreciation largely determined current account balances in run up to the 2008 crisis (Atoyan, Manning and Rahman, 2013). - Offering more incentives to foreign capital in current global uncertainty and unprecedented external imbalance risks exposure to short-term volatile capital (Nagaraj, 2013; Subramanian, 2013). - Debt versus equity flows: - Debt capital flows increase vulnerability to future crises; EDE regions with large debt flows and large CADs faced crises after the NAFC. - One factor reducing India’s external vulnerability is that public debt is largely internally held. - Caution on opening government securities market: - Further opening up of government securities market to non-resident investment needs careful watching. - Debt investments by non-residents in domestic securities are more volatile than equity and can add to foreign exchange market pressures; react to monetary policy developments in advanced economies. - Example: during June-September 2013, foreign institutional investors sold both equities and debt securities, with the sell-off in debt securities (around US $ 10 billion) five-times that of equity investments (US $ 2 billion). - Open debt markets can lead to large destabilizing capital inflows and outflows, complicating domestic macroeconomic management. - On local-currency non-resident investment risk: - View that foreign-currency borrowing fears do not apply to non-residents’ investments in local-currency bonds was disproved during June-August 2013 turmoil. - Sudden large sell-off by non-residents in local-currency bonds puts sharp downward pressure on the currency with adverse implications for corporates, the fisc, and inflation. - Large borrowings from foreign sources, whether foreign-currency or local-currency denominated, add to domestic vulnerability. ### Price and financial stability: institutional issues and macroprudential arrangements - Inflation context: - Inflation in India had averaged 5-6 percent per annum between mid-1990s and late 2000s, down from 8-9 percent in previous three decades. - Post-NAFC, inflation increased again despite low global inflation. CPI inflation remains well above comfort levels. - There is a gap between headline WPI and core WPI, and between headline CPI and headline WPI, posing challenges for monetary policy formulation and communication. - Weight of food and fuel in CPI: - In the new CPI, food items have a weight of 59 percent (rural), 37 percent (urban) and 50 percent (all-India). - Fuel group weight: 8-10 percent. - Combined weight of food and fuel: 46-70 percent across the three indices. - Given these large weights and high volatility in food prices, an inflation targeting framework focused on a single indicator is conceptually and practically challenging in India (Mohan, 2011). - Monetary policy framework: - Despite challenges with conventional inflation targeting, price stability (low and stable inflation) should remain the key objective of monetary policy. - The Expert Committee to Revise and Strengthen the Monetary Policy Framework recommended headline CPI inflation as the nominal anchor and predominant objective for monetary policy in India (RBI, 2014). - Patel Committee recommendations aim to avoid turning the RBI into agents focused solely on reducing inflation at the expense of financial stability (Rajan, 2014). - Financial stability and central bank mandate: - NAFC showed price stability alone does not guarantee financial stability; central banks mandated only with price stability/inflation targeting may endanger financial stability (Eichengreen et al., 2011). - Global trend post-crisis: many countries are integrating prudential functions into central banks or revisiting regulatory architecture. - Cross-country survey on supervisory arrangements (Barth, Caprio and Levine, 2013): - Central bank is the only supervisory authority in 89 countries. - Central bank is not a supervisory authority in 38 countries. - In 9 countries, central bank is among multiple supervisors (United States is one). - Broader financial system supervision: 101 countries have multiple authorities; 25 countries have a single authority covering the entire financial sector. - Macroprudential policymaking models (IMF, 2013b): - Model 1: Macroprudential mandate assigned to central bank, with decisions by its Board (e.g., Malaysia, Czech Republic). - Model 2: Macroprudential mandate assigned to a dedicated committee within central bank (e.g., U.K.). - Model 3: Macroprudential mandate assigned to a committee outside the central bank, with central bank participation (e.g., Australia, France, U.S.); mitigation of Ministry of Finance dominance via central bank chair, strong voice, or veto in some countries. - Institutional debate in India: - Financial Sector Legislative Reforms Commission (FSLRC) proposals to restrict Reserve Bank’s role mainly to monetary policy and price stability need serious debate (Tarapore, 2013a). - FSLRC recommendation to vest executive responsibility for safeguarding systemic risk with the FSDC Board runs counter to post-crisis trend where collegial bodies are for coordination and recommendation; extending executive authority to FSDC Board could compromise synergy between monetary policy and financial stability. - International experience cautions against narrowing central bank mandates to price stability alone; central bank should retain multiple objectives and a broad toolkit (Reinhart and Rogoff). ### Fiscal policy: consolidation, composition and buffers - Need for fiscal consolidation: - Fiscal consolidation is necessary for sustained growth in an environment of macroeconomic and financial stability. - High fiscal deficits and debt limit policy flexibility and contributed to worsening economic conditions during the NAFC. - Fuel subsidies in India are badly targeted; richest ten percent of households receive seven times more in benefits than the poorest ten percent (Anand et al, 2013); scope exists for better targeting while protecting lower-income households. - Quality of consolidation: - Fiscal consolidation should focus on quality: shifting expenditure from revenue (subsidies) to capital outlays. - Estimates by Jain and Kumar (2013): fiscal multiplier for non-defense capital outlays is 2.1 vs. 0.2 for revenue expenditure for the Central government. Directing expenditures toward capital outlays could have a less negative, or even positive, impact on growth. - Such consolidation would provide monetary policy greater maneuverability and could be short-term growth positive. - Policy buffers and targets: - Policymakers should maintain significant policy buffers to use in crises; India’s buffers were limited at onset of the NAFC. - A more ambitious medium-term fiscal consolidation plan beyond 2017-18 is critical. - Earlier FRBM plan targeted a 3 percent GFD/GDP ratio for the central government; revised path also targets 3 percent by 2017-18. - It would be judicious to aim for balanced budget targets by the end of the decade. - Expected effects: - Stronger fiscal consolidation, with attention to quality, would increase public sector savings, lower domestic interest rates, improve domestic savings, and allow recovery in private sector investment. ### Conclusion and growth outlook - Background: - Current growth slowdown followed almost a decade of consistent high growth and sharp recovery from 2008-09. - High growth during pre-NAFC (especially 2003-08) was underpinned by fiscal correction, increasing public savings, low inflation, anchored expectations, low nominal and real interest rates, and higher corporate profitability and investment. - Market-determined exchange rate system, cautious capital account management, sterilized interventions, and progressive deregulation/liberalization aided export growth and moderate CAD. - Sources of slowdown: - Overshooting of post-NAFC stimulus: very high growth averaging around 9 percent during 2009-10 and 2010-11 sowed seeds for inflation and CAD pressures. - Quality of fiscal stimulus (focus on tax cuts and increased revenue expenditure/subsidies with stagnant capital outlays) added to demand pressures and high inflation. - Withdrawal of stimulus was gradual and incomplete (especially fiscal measures); delayed and incomplete pass-through of international oil prices to domestic prices added to fiscal pressures and impeded expenditure adjustment. - Large monetary stimulus facilitated financing of near-trebling of government borrowing needs at lower yields; delayed fiscal withdrawal led to crowding out of private sector and contributed to massive decline in private corporate investment. - High inflation and negative real deposit rates caused switch from financial savings to gold, raising gold imports and CAD. - Domestic policy bottlenecks led to more coal imports and lower iron ore exports. - CAD evolution: - Combination of domestic and global factors pushed CAD to 4.7 percent by 2012-13. - Rapid policy response after mid-2013 market turbulence reduced CAD to 2.3 percent of GDP for 2013-14 (April-December). - Policy recommendations and outlook: - Appropriate policies on domestic oil prices are needed to contain fiscal subsidies and oil imports; recent government steps are welcome and should be persevered with. - Restoration of tax/GDP ratios, implementation of goods and services tax, introduction of direct tax code, and measures to contain subsidies should help contain fiscal deficit in next couple of years. - Fiscal correction and recovery in public sector savings are expected to raise gross domestic savings rate by around 2-3 percent of GDP. - Fiscal consolidation would allow durable reduction in inflation and yield low and positive real interest rates for depositors and borrowers, moderating gold imports and CAD. - The slowdown has a large cyclical component reflecting domestic and global factors; structural drivers of growth—favorable demographics and high savings and investment rates—remain broadly intact. - Based on these assumptions, a conservative estimate would result in a sustained gross domestic saving rate of about 35 percent, which should facilitate growth of 8- (text truncated). *Source: _wp1468 - Annex Table 4). Third, the results for overall exports broadly hold for exports of the major (IMF PDF content provided).* ### 8.5 percent, given moderate incremental capital output ratios. This would be contingent on the ### _wp1468 - 8.5 percent, given moderate incremental capital output ratios. This would be contingent on the ### Growth prospects and contingencies - Projected potential growth: 8.5 percent, given moderate incremental capital output ratios. - Condition: contingent on the removal of the recent impediments to domestic investment activity and recovery in the global economy. - Interest rate outlook: interest rates in India can be expected to remain above those in advanced economies; prudent approach needed with regard to opening up of debt flows to foreign investors. ### Key policy priorities to restore sustained growth in excess of 8 percent - Restore macroeconomic stability by reducing medium term inflation to levels achieved in the decade prior to the North Atlantic financial crisis (NAFC). - Coordinate supply-side and demand management measures: - Fiscal consolidation. - Appropriate monetary policy. - Active measures to loosen supply constraints. - Factor market reforms: - Labor and land reforms to increase flexibility in the face of burgeoning demand from growing incomes. - Manufacturing revival: - Realistic exchange rate policy combined with policies promoting labor flexibility and skill development to promote manufacturing growth. - Objective: achieve 10 per cent plus growth in the manufacturing sector; without this, sustained GDP growth of 8 per cent plus is not feasible. - Revive the “animal spirits” in Indian industry inspired by the 1991 reforms in a focused manner. ### Infrastructure, fiscal policy, and revenue mobilization - Infrastructure investment critical to loosening supply-side constraints and promoting manufacturing. - Need to enhance public investment in infrastructure on a sustained basis despite increasing private investment. - Two prerequisites for sustainable public infrastructure investment: - Reinforce the culture of economic user charges to make infrastructure investment remunerative. - Curtail expenditures on non-merit subsidies and redirect savings toward infrastructure investments as incomes rise. - Fiscal consolidation critique: - Recent trend overly focused on reduction in capital expenditure rather than revenue expenditure such as subsidies; this needs to be reversed. - Tax/GDP and revenue receipts: - Indian tax/GDP ratio has been relatively stagnant for a decade despite high GDP growth and rising incomes. - Revenue receipts/GDP ratio of the Central government is now below the levels prevailing in the late 1980s. - Public investment in physical and social infrastructure will be difficult without revenue enhancement consistent with income growth. ### Public-sector delivery and second-generation reforms - Ongoing need for reforms in agriculture, physical and social infrastructure, including urban infrastructure. - Concern that private sector–led acceleration may have reached limits and constrained by lack of public investment. - Underlying constraint: inadequate delivery of public services in both quality and quantity. - Second-generation reforms should focus on empowering the public sector (broadly defined across local, state, and national levels) to deliver public goods and services. - Four priority public-sector-delivered areas (often via public–private partnerships): - Agricultural development. - Urban development. - Human resource development. - Management of public services. - Common problem: lack of competence in public systems governing these areas; governance issues call for significant improvement in public service delivery. ### Annex 1 — Oil consumption and prices: model and data - Empirical specification estimated (April 2001 to March 2013): - Log(Ct) = a1 + a2.log(Yt) + a3.log(Pt/WPIt) + et - C = consumption of specific petroleum item or total petroleum products. - Y = real GDP. - P = price of the specific petroleum item (indices in WPI basket). - WPI = overall price level in the country. - Methodology: - Variables non-stationary; cointegration framework using autoregressive distributed lag (ARDL) approach to cointegration. - Robustness tested via Johansen-Juselius approach. - Key period: April 2001 to March 2013. ### Annex 1 — Elasticities and empirical findings - Income elasticities: - Petrol: around 1.2. - Diesel: almost unity. - LPG: below unity (indicative of supply constraints). - Kerosene: negative (role diminishing); share of kerosene in total petroleum products’ consumption halved from around 10 percent to 5 percent over the study period (2002-12). - Overall petroleum products: 0.5. - Price elasticities (statistically significant): - Petrol: -0.66. - Kerosene: -0.54. - Diesel: -0.36. - Interpretation: an increase of 10 percent in domestic prices is associated with a reduction of 3.6-6.6 percent for these three products. - LPG and overall petroleum consumption price elasticities: negative but not significant. - Implication: insignificant price elasticity of total petroleum consumption despite significant elasticities for major components suggests merits of a disaggregated approach. ### Annex Table 1 — Selected estimated coefficients and statistics (as reported) - LGDPRSA coefficients (income elasticities) by product: - Diesel (LHSDA): 0.96 (t-statistic: 37.37) - Petrol (LPETROLSA): 1.18 (t-statistic: 19.33) - Kerosene (LSKOSA): -0.49 (t-statistic: 9.18) - LPG (LLPGSA): 0.66 (t-statistic: 8.61) - POL total (LPOLTOTSA): 0.50 (t-statistic: 32.70) - Price elasticities (LP_...): - Diesel (LP_HSDSA): -0.36 (t-statistic: 5.62) - Petrol (LP_PETROLSA): -0.66 (t-statistic: 3.70) - Kerosene (LP_SKOSA): -0.54 (t-statistic: 4.78) - LPG (LP_LPGSA): -0.35 (t-statistic: 1.20) - Mineral oils (LMINOILSA): -0.06 (t-statistic: 1.48) - R-bar^2 (levels regressions): - Diesel: 0.996 - Petrol: 0.997 - Kerosene: 0.940 - LPG: 0.986 - POL total: 0.984 - Short-run ECM(-1) coefficients: - Diesel: -0.42 (t-statistic: 6.10) - Petrol: -0.18 (t-statistic: 3.23) - Kerosene: -0.38 (t-statistic: 3.73) - LPG: -0.29 (t-statistic: 2.69) - Note: Estimates based on ARDL methodology; quarterly data for 2001:2 to 2013:1; figures in parentheses are t-statistics. ### Annex 2 — Growth slowdown: framework and findings - Context: accommodative monetary and fiscal policies promoted growth during 2009-11; phased reversal contributed to slowdown in 2011-12 and 2012-13. Anemic global growth also a factor. - Modeling approach: - IS-curve framework following Kapur and Behera (2012). - Dependent variable: GDP excluding agriculture and community, social and personal services. - Explanatory variables: real domestic interest rates, external demand, real exchange rate, real bank credit. - Real interest rate defined as effective nominal policy rate less 4-quarter average of non-food manufactured products inflation. - External demand proxied by real world exports or global GDP. - Real exchange rate: RBI’s 36-currency export-weighted real effective exchange rate index. - Equation estimated for 1996-97 to 2007-08 (quarterly) and used to forecast out-of-sample. - Variable treatment: all variables except real interest rate are in growth terms (quarter-on-quarter, seasonally adjusted) and stationary. ### Annex 2 — Quantitative impacts on domestic growth - Real interest rate: - An increase of 100 bps in the real interest rate leads to a reduction of 26-33 bps in GDP growth with a lag of two quarters (Annex Table 2, columns 2 to 5). - External demand: - One percentage point reduction in global exports reduces domestic growth by almost 13 bps. - One percentage point reduction in global GDP growth reduces domestic growth by almost 40-56 bps. - Note: coefficient on world exports lower than on world GDP owing to different average growth magnitudes (annual growth in real world exports averaged 7.7 percent vs. 3.8 percent in global GDP over sample 1997:3-2008:1). - Real exchange rate: - Appreciation reduces domestic output; depreciation boosts output. - Regression diagnostics reported as satisfactory. ### Annex Table 2 — Reported coefficients and diagnostics (selected) - Constant terms (columns 1–4): 6.38, 6.45, 5.62, 5.15 (t-statistics: (5.91), (6.10), (4.78), (4.10)) - RIRQ(-2) coefficients (real interest rate lagged 2): - Column 1: -0.32 (t-statistic: 3.26) - Column 2: -0.33 (t-statistic: 3.23) - Column 3: -0.26 (t-statistic: 2.76) - Column 4: -0.27 (t-statistic: 2.82) - WEXPRGQ (world exports growth) coefficients (columns shown): 0.13 (t-statistics: (2.40), (2.43)) - GDPWORLDQ (world GDP growth) coefficients: 0.40 and 0.56 (t-statistics: (1.99), (2.42)) - REERXQ (real exchange rate q-o-q): -0.09 (t-statistic: (1.74)); REERXQ(-3): -0.08 (t-statistic: (1.95)) - NFCRQ(-2) (real non-food credit growth lagged 2): 0.14 (t-statistics: (3.21), (3.15), (3.26), (3.10) across columns) - R-bar^2: 0.30, 0.32, 0.29, 0.31 - Diagnostic tests (selected): - DW: 1.97, 1.76, 2.15, 2.12 - White test p-values: 0.43, 0.62, 0.69, 0.90 - JB test p-values: 0.91, 0.89, 0.91, 0.69 - LB-Q test p-values: 0.34, 0.20, 0.47, 0.33 - Dependent variable: GDPRGQ = growth (q-o-q) in real GDP excluding agriculture and community services. - Sample period: 1996:2–2008:1. *Source: Authors' estimates, as presented in the source document.* ### Annex 3: Determinants of Exports and Imports ### Annex 3: Determinants of Exports and Imports ### Methodology - Export demand specification: depends on external demand and real exchange rate. - Import demand specification: depends on domestic activity and real exchange rate. - Estimation framework: cointegration and vector error correction mechanism (VECM) using Johansen-Juselius methodology. - Estimated short-run VECM equations (first-differences, logs): - ∆LEXPQt = a1 + a2.ECM(t-1) + ∑a3i.∆LEXPQ(t-i) + ∑a4i.∆LREER(t-i) + a5.∆LWEXPR(t-i) + ut - ∆LIMPQt = b1 + b2.ECM(t-1) + ∑b3i.∆LIMPQ(t-i) + ∑b4i.∆LREER(t-i) + b5.∆LY(t-i) + vt - ECM interpretation: ECM terms capture deviations from long-run equilibrium: - Exports ECM: (LEXPQ – c1 – c2.LREER – c3.LWEXPR) - Imports ECM: (LIMPQ – d1 – d2.LREER – d3.LY) - Expected sign: coefficients on ECM terms are expected to be negative (speed of adjustment). - Weak exogeneity findings: world real exports treated as weakly exogenous in exports VAR; domestic activity found weakly exogenous in imports VAR. ### Data and Sample Periods - Baseline annual sample: 1980-81 to 2007-08. - Robustness/extended annual sample: up to 2011-12. - Quarterly post-reforms sample: 1996-97 to 2007-08 (quarterly trade volumes; DGCI&S quantum index for quarter-end months). - Rationale for sample choices: - 1980-81 start coincides with early 1980s reform phase. - Focus on pre-NAFC period due to NAFC and global disruptions. - Post-1990s reforms increased market role; price signal responsiveness may have risen. - Annual data may under-estimate price elasticities due to intra-year REER volatility; hence quarterly robustness. - REER measures used: REER-RBI (36-currency export weighted), REER-OECD, REER-IMF. - Oil exports treatment: augmented specifications include real oil exports (nominal oil exports divided by international crude oil prices) and dummies for outliers. ### Exports — Annual Estimates (1980-81 to 2007-08) - Cointegration tests: - Null hypothesis of at least one cointegrating vector cannot be rejected for both basic and augmented specifications for REER-RBI, and for the augmented specification for REER-OECD and REER-IMF. - Long-run elasticities: - World real exports: an increase of 1 percent in world real exports is associated with an increase of 1.1-1.4 percent in India’s exports in the long-run. - REER long-run coefficient: ranges from 0.21 to 0.65 across alternative specifications; coefficients lower for OECD and IMF REER measures. - Short-run dynamics: - ECM coefficients for exports and REER equations are correctly signed and significant. - Deviations from long-run equilibrium are adjusted through adjustments in both exports and the REER. - Short-run coefficients indicate significant contemporaneous impact of external demand on exports, although partly offset in the next year. - Regression diagnostics for the short-run model are satisfactory. - Extension to 2011-12: - Results broadly carry through when sample extended to 2011-12 and are more supportive. - Selected coefficient table excerpts (Annex Table 3): - L(EXPQ): 1.00 (presented repeatedly) - L(REER): 0.37, 0.65, 0.27, 0.21 (across columns) - L(WEXPR): -1.33, -1.10, -1.35, -1.42 - L(OILR): 0.06, 0.04, 0.02 - Constants shown: 3.93, 1.02, 4.50, 5.26 - Short-run ECM( -1) examples: -1.02, 0.37, -0.45, 0.69, -0.60, 0.21, -0.47, 0.09 - DLWEXPR contemporaneous coefficients: 1.08, -0.14, 0.76, -0.44, 0.40, 0.37, 0.38, 0.01 - DLOILR contemporaneous coefficients: 0.03, 0.00, 0.02, 0.00, 0.02, 0.00 - Diagnostic p-values (examples): - PV (r=0): 0.02, 0.04, 0.05, 0.05 - PV (r=1): 0.21, 0.89, 0.47, 0.41 ### Exports — Quarterly Estimates (1996:2 to 2008:1) - Elasticities with world exports (WEXPR) (post-reforms quarterly): - Elasticity range with respect to world exports: 1.6-1.9 across alternative estimates (higher than annual estimates). - Elasticity with respect to world GDP: 2.6-3.6. - REER sensitivity: - Coefficient on REER sensitive to choice of REER and inclusion of oil exports. - With REER-RBI: - (Absolute) coefficient is more than unity and significant when oil exports are included. - With REER-OECD and REER-IMF: - (Absolute) coefficient is more than 1.6 and statistically significant only when oil exports are excluded. - Major manufactured product categories: - Income elasticities (with respect to world exports): range from 1.5 to 2.8 for chemicals, manufactures, and machinery. - Income elasticities (with respect to world GDP): range from 2.9 to 5.1 for the same categories. - Manufactured goods classified chiefly by material subgroup: - Elasticity of exports with respect to REER: 1.6-3.8 across REER indicators. - Machinery and transport equipment: - Price elasticities: 2.4-3.3 for OECD and IMF indicators; estimate wrongly signed for RBI REER measure. - Chemicals and related products subgroup: - Exchange rate elasticity: 1.7-2.9 for OECD-IMF indicators; not significant (but correctly signed) for RBI REER. - Short-run dynamics: - ECM( -1) examples: -0.47, -0.63, -0.71, -0.64, -0.61, -0.65, -0.81, -0.90, -0.79, -0.40, -0.75, -0.54, -0.70, -0.58, -0.75 (all significant with t-statistics reported). - Diagnostics and fit (selected): - R2 range examples: 0.33, 0.33, 0.39, 0.51, 0.36, 0.38, 0.40, 0.45, 0.42, 0.44, 0.44, 0.31, 0.40, 0.35, 0.45 - PV (r=0) examples: 0.09, 0.06, 0.01, 0.02, 0.01, 0.00, 0.00, 0.00, 0.000, 0.03, 0.01, 0.02, 0.03, 0.01, 0.01 ### Robustness: Quarterly DOLS Estimates (Annex Table 5) - DOLS used as robustness check given small sample size. - DOLS specifications: one lead and lag and two leads and lags; rolling samples starting from indicated quarter (effective sample begins quarter ended December 1996 given lags). - DOLS results: - Estimates using REER-RBI, REER-OECD, and REER-IMF presented for sample starting quarters from Dec-96 through Mar-01. - Examples of reported coefficients and t-statistics (selected cells, REER coefficients shown): - REER-RBI: Dec-96 0.22 (t 0.16), Mar-97 -0.31 (t -0.23), Jun-97 -0.81 (t -0.62), Sep-97 -1.56 (t -1.12), Dec-97 -1.99 (t -1.31) - REER-OECD: Dec-96 -2.70 (t -2.92), Mar-97 -2.36 (t -2.35), Jun-97 -2.18 (t -2.01) - REER-IMF: Dec-96 -3.03 (t -3.69), Mar-97 -2.74 (t -3.27), Jun-97 -2.60 (t -2.99) - DOLS estimates with one lead and lag are broadly in line with Johansen-Juselius methodology results. - Diagnostic p-values for autocorrelation (LB-Q(2), LB-Q(4)) reported in table for each rolling sample and specification. ### Imports - Domestic activity indicator: - Industrial GDP found to be a better indicator than overall GDP, reflecting import composition tilt towards industrial raw materials. - Cointegration: - Null hypothesis of one cointegrating vector cannot be rejected. - Long-run coefficients: - Income elasticity of imports with respect to domestic industrial GDP: around 1.55 in all cases (Annex Table 6). - REER coefficient: ranges from 0.19 to 0.42; higher for RBI-REER. - Short-run dynamics: - ECM terms correctly signed. - Both imports and REER adjust to deviations from equilibrium. *Source: Authors' estimates in Annex 3: Determinants of Exports and Imports.* ### Annex Table 6: Determinants of Imports (Annual Estimates) ### Annex Table 6: Determinants of Imports (Annual Estimates) ### Long-run cointegrating estimates (results reported for REER-OECD) - L(IMPQ): 1.00 1.00 1.00 1.00 1.00 - L(REER): - 0.39 -0.42 -0.32 - 0.19 -0.21 - (3.70) ( 3.74) ( 5.85) (4.90) ( 5.17) - L(INDR): - 1.57 -1.57 -1.57 - 1.55 -1.55 - (32.33) ( 32.04) ( 45.04) ( 59.84) ( 60.60) - CONSTANT: 15.58 15.70 15.45 14.67 14.63 - (15.12) ( 14.86) ( 23.26) ( 29.34) ( 29.68) ### Short-run ECM model (selected coefficients and t-statistics) - ECM(- 1): - 0.38 0.48 -0.33 0.48 -0.30 0.48 - 0.93 0.54 -0.74 0.60 - (1.80) ( 4.56) ( 1.77) ( 4.87) ( 1.67) (5.65) (2.93) (2.39) ( 2.53) (2.77) - DLIMPQ(- 1): 0.17 -0.37 0.21 -0.37 0.04 - 0.43 0.39 - 0.12 0.34 - 0.16 - (0.68) ( 3.08) ( 0.94) ( 3.17) ( 0.21) (4.39) (1.58) (0.71) ( 1.48) (0.91) - DLREER(-1): 0.09 -0.17 0.17 -0.17 0.35 0.16 0.47 - 0.03 0.50 - 0.06 - (0.32) ( 1.16) ( 0.63) ( 1.16) ( 1.15) (1.08) (1.56) (0.13) ( 1.75) (0.27) - DLIMPQ(- 2): 0.30 -0.01 0.21 -0.04 0.38 0.23 0.27 0.17 - (1.42) ( 0.09) ( 1.10) ( 0.40) (1.99) (1.66) ( 1.47) (1.26) - DLREER(-2): 0.32 -0.45 0.22 -0.48 0.53 - 0.08 0.36 - 0.13 - (1.10) ( 3.06) ( 0.83) ( 3.35) (1.84) (0.37) ( 1.30) (0.65) - DLINDR: 2.32 0.23 1.98 0.21 2.18 0.16 2.44 - 0.37 2.10 - 0.36 - (4.76) ( 0.92) ( 4.32) ( 0.87) ( 4.81) (0.74) (5.46) (1.14) ( 4.92) (1.14) - DLINDR(- 1): - 2.27 0.92 -1.84 1.03 -0.87 0.45 - 2.17 - 0.01 -1.70 0.12 - (3.17) ( 2.55) ( 2.74) ( 2.89) ( 1.98) (2.19) (3.40) (0.01) ( 2.71) (0.25) - DLINDR(- 2): 0.28 -0.38 0.32 -0.32 0.01 0.11 0.05 0.23 - (0.48) ( 1.30) ( 0.59) ( 1.10) (0.02) (0.28) ( 0.09) (0.59) - dum84: -0.17 -0.22 0.02 -0.16 - 0.03 - ( 2.48) ( 2.87) (0.41) ( 2.40) (0.63) ### Model diagnostics (reported across specifications) - Serial Correlation @: 0.21 0.06 0.32 0.14 0.04 - ARCH @: 0.15 0.78 0.23 0.89 0.59 0.16 0.43 0.78 0.37 0.61 - Normality @: 0.41 0.50 0.20 0.46 0.59 0.06 0.17 0.92 0.11 0.86 - R2: 0.48 0.64 0.57 0.67 0.52 0.77 0.56 0.58 0.61 0.61 - VAR lags: 3 3 2 3 3 - PV (r=0): 0.01 0.02 0.01 0.03 0.03 - PV (r=1): 0.48 0.81 0.39 0.57 0.78 ### Notes on Annex Table 6 - Sample period for the estimation is 1980- 81 to 2007- 08. - Variables defined: - IMPQ = quantum index of India's imports; - INDR = India's real industrial GDP; - REER = real effective exchange rates compiled by RBI (36-currenncy trade weighted), OECD and IMF, respectively. - dum84 is dummy for 1984-85. - Rows PV (r=0) and PV (r=1) give p-values (Bartlett-corrected) for the null of no and one cointegrating vector, respectively. - @: p-values for the null hypotheses of no serial correlation, no conditional heteroskedasticity and normality of residuals. - Figures in parentheses are t-statistics. - Source: Authors' estimates. ### Annex 4 — Determinants of Gold Imports (overview of empirical approach) - Monthly data used: April 2004 to December 2012 (choice driven by availability of monthly value of gold imports). - Two baseline specifications estimated: 1. Nominal specification with domestic price variables: - MGt = a + trend + b. ∆LPGt + c. ∆LPt + d.RNt + e. ∆LBSEt + f. ∆Et + ut 2. Real specification: - MGt = a1 + trend + b1. ∆LPGRt + d1.RRt + e1. ∆LBSERt + f1. ∆ERt + ut - Definitions and transformations: - ∆L = first difference of log terms (variables enter in first differences as they turn out to be stationary). - MG = volume of monthly gold imports. - PG = local currency gold price. - P = domestic prices (measured by WPI or CPI). - RN = nominal deposit interest rate. - BSE = Bombay Stock Exchange (BSE Sensitive) index. - E = exchange rate (Rupees per US dollar). - PGR = local currency gold price in real terms. - RR = real deposit interest rate. - BSER = real BSE index. - ER = real exchange rate. - Real variables obtained by deflating nominal variables with either WPI or CPI. - Real interest rate computed as nominal deposit rate less year-on-year WPI (or CPI) inflation. - Monthly dummies included to capture seasonality; trend included as proxy for higher incomes. - Augmented Dickey-Fuller (ADF) tests: - Null of unit root cannot be rejected for level series (other than monthly imports and WPI), but can be rejected at the 5 percent level for their first-differences. - Estimation details: - Four specifications estimated (Annex Table 7): columns 2 and 3 estimate equation (1) for CPI and WPI, respectively; columns 4 and 5 estimate equation (2) for real variables deflated by CPI and WPI, respectively. - Equations estimated with six lags of each variable and insignificant variables excluded using the STWISE command in WinRATS. - General result statement: - All specifications have relatively good explanatory power and regression diagnostics are satisfactory. ### Annex Table 7: Determinants of Gold Demand (selected coefficients and diagnostics) - Dependent Variable: Monthly Gold Imports (LMG). Sample period: April 2004 to December 2012. Regressions include monthly dummies. - Reported coefficient blocks (columns labeled 1 2 3 4 5 in source): - Constant: 1.77 1.73 1.79 1.76 - (5.14) (4.98) (4.77) (4.83) - TREND: 0.00 0.00 0.00 0.00 - (3.83) (4.25) (3.01) (3.47) - LMG(-1): 0.25 0.24 0.34 0.31 - (3.86) (3.67) (5.18) (4.54) - DLPG: - 6.24 - 6.43 - (8.39) (8.62) - DLPG(-3): - 1.05 - 0.99 - (1.55) (1.45) - DLPG(-4): 1.23 0.91 - (1.76) (1.33) - DLPG(-5): 1.56 1.61 - (2.19) (2.24) - DLPG_R: -6.39 -6.77 - (8.42) ( 8.99) - DLPG_R(-3): -1.33 - (1.86) - DLPG_R(-5): 1.82 1.45 - (2.40) ( 1.95) - DLCPI(-1): 6.84 - (1.83) - DDRATE(-1): 0.26 0.20 - (1.79) (1.38) - DDRATE(-2): 0.43 0.46 - (3.13) (3.27) - DDRATE_R(-2): 0.08 - (2.39) - DLBSES(-4): 0.58 0.77 - (1.38) (1.86) - DLBSES_R(-4): 0.74 - (1.70) - DLEXCH: - 2.58 - 1.92 - (1.81) (1.38) - DLEXCH_R(-1): -1.94 - (1.39) - DLEXCH_R(-5): -3.21 -3.48 - (2.12) ( 2.49) - DUM2006M6: - 1.63 - 1.65 - 1.64 - 1.63 - (5.62) (5.61) (4.95) (5.05) - Goodness-of-fit and diagnostic statistics: - R-bar2: 0.68 0.67 0.62 0.63 - DW: 2.07 2.10 2.16 2.00 - White test: 0.52 0.53 0.34 0.65 - JB test: 0.47 0.46 0.68 0.58 - LB-Q test: 0.28 0.27 0.16 0.18 ### Notes on Annex Table 7 (definitions and tests) - MG = monthly gold imports; PG = Price of gold in rupees; CPI = Consumer price index; WPI = Wholesale price index. - PG_R = Price of gold in rupees deflated by CPI (column 4) and WPI (column 5). - BSES = Bombay Stock Exchange Index (BSE-30); EXCH = Exchange rate (Rupees per US dollar). - Prefix DL stands for log difference; DDRATE = Variation in deposit rate; DDRATER = Variation in real (CPI-adjusted) deposit rate; DLEXCH_R = exchange rate depreciation less CPI inflation (col. 4) and WPI inflation (col. 5). - Figures in parentheses are t-statistics. - White test = significance level (p-value) for White test for the null of homoskedasticity of residuals. - JB test = significance level (p-value) for Jarque-Bera test for the null of normality of residuals. - LB-Q test = significance level (p-value) of Box-Pierce-Ljung Q-statistic for the null of no residual autocorrelation for 6 lags. *Source: Authors' estimates.* --- _Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2014/_wp1468.pdf_