## _wp1270 - References

## Source details

**Canonical URL:** [_wp1270 - References](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp1270.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp1270.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp1270.pdf.json)

---

### I. Introduction: recent patterns in investment in India
- Corporate investment was 14 percent of GDP before the crisis and slowed to 10 percent of GDP since 2009.
- Growth of investment (gross fixed capital formation):
  - slowed to about –0.6 (quarter on quarter) on average in the first three quarters in 2011;
  - averaged nearly 3 percent between 2000–2007.
- High-frequency capital goods production data suggest corporate investment may be weakening further (machinery and equipment production declining).
- Scenario example:
  - if the quarterly growth rate of corporate investment stays at –¼ percent (the average growth in investment in 2011), lower demand would reduce medium-term GDP growth to around 7 percent.
- Potential drivers of weak corporate investment:
  - Macroeconomic uncertainty (high and volatile inflation, weaker global outlook).
  - Monetary tightening since early 2010.
  - Structural factors: unfavorable business environment, weakening governance, slower government project approvals, high costs of doing business.

### II. Macroeconomic analysis: model, key coefficients, and contributions
- Macro model (annual data, FY1992–2011) for corporate investment:
  - CI_t = a1 X_{t–1} + a2 VIX_t + a3 CI_{t–1} + ε_t
  - X includes volatility of the inflation rate (CPI), the inflation rate (CPI), real GDP growth, the real interest rate, and the world’s real GDP growth.
  - VIX used to control for global stock volatility.
  - Volatility of the inflation rate calculated by the standard deviation of monthly inflation rates (month on month, seasonally adjusted) in the year.
  - Real interest rate measure: average of nominal prime lending rates of ICICI Bank and IDBI Bank deflated by CPI inflation.
- Main macro findings:
  - The coefficient on the volatility of inflation is consistently negative and statistically significant in most specifications.
  - Coefficients on the real interest rate and global stock volatility are negative and statistically significant.
  - Other explanatory variables generally have expected signs but are often statistically insignificant.
- Selected coefficient estimates (as reported in Table 1):
  - lag of volatility of inflation rate: -2.79* (standard error 1.32), -2.62* (1.31), -1.59 (1.08).
  - lag of inflation rate: -0.11 (0.16), -0.051 (0.17), -0.21* (0.11).
  - lag of real GDP growth: 0.10 (0.22), 0.14 (0.17), -0.03 (no SE in table snippet).
  - lag of real GDP growth ex agriculture: 0.28 (0.32).
  - lag of real interest rate: -0.51** (0.22), -0.45* (0.23), -0.34* (0.17).
  - lag of world's real GDP growth: 0.36 (0.31), 0.37 (0.31).
  - global stock volatility: -0.11*** (0.035) in one specification.
  - lag of corporate investment (in % of GDP): 0.65*** (0.19), 0.55** (0.23), 0.85*** (0.14).
  - Total contribution to corporate investment between end-FY07 and end-FY11 (using coeffs in specification (3)): -1.3 (% of GDP).
- Quarterly analysis (gross fixed capital formation as dependent variable):
  - I_t = a1 X_{t–1} + a2 X_{t–2} + a3 VIX_t + a4 I_{t–1} + a5 I_{t–2} + ε_t
  - Standard macro variables largely explain investment historically, but since late 2010 actual investment has fallen short of predicted levels, implying additional structural factors (for example, business environment) not captured by standard macro variables.

### III. Firm-level micro panel analysis: determinants of corporate investment
- Micro model:
  - i_{j,t}/k_{j,t} = a1 profitability_{j,t–1} + a2 liquidity_{j,t–1} + a3 leverage_{j,t–1} + a4 i_{j,t–1}/k_{j,t–1} + controls + ε_{j,t}
  - Profitability measured by Tobin’s q; liquidity measured by ratio of liquid assets to capital k; leverage measured by ratio of debt to total assets.
  - Firm stock price volatility and time dummies included as controls.
  - Estimation: Arellano–Bond Dynamic Panel GMM after first-differencing.
- Data and sample:
  - CMIE Prowess dataset; sample restricted to nonfinancial companies.
  - Sample after selection: 12,306 observations (2,291 companies) over 1990–2011.
- Main micro findings:
  - Profitability and liquidity coefficients are positive and generally significant.
  - Leverage coefficient is negative and significant.
  - Interaction of liquidity with manufacturing dummy, firm size dummy, or exporter dummy is statistically insignificant in reported specifications, implying the positive liquidity effect does not differ significantly by these characteristics in those specifications.
  - Positive liquidity coefficients imply remaining frictions in financial access; in perfect markets liquidity would not affect investment.
  - Raising profitability is critical to stimulate investment in India.
- Policy-relevant implication from micro evidence:
  - Improving the business environment, upgrading the financial system, and developing infrastructure could stimulate corporate investment by increasing profitability and easing financial constraints.

### IV. Business environment, competitiveness, and within-India variability
- Doing Business (World Bank, 2011):
  - India ranked 132nd of 183 countries in ease of doing business in 2011 (up from 139th in 2010).
  - 2011 ranks for specific indicators: starting a business 166th, registering property 97th, enforcing contracts 182nd (with costs amounting to nearly 40 percent of claims).
- OECD PMR indicators: barriers to entrepreneurship are restrictive in India relative to advanced and emerging-economy standards.
- Global Competitiveness (World Economic Forum):
  - India’s headline global competitiveness fell to 56th place in 2011 from 49th in 2009.
  - Institutions sub-index ranking declined from 34th in 2006 to 69th in 2011.
- Within-India variability:
  - Doing Business indicators vary significantly across Indian cities (e.g., costs of registering property; costs to export in US$ per container).
  - OECD PMR indicator shows noticeable differences across states.
  - Some indicators (e.g., corporate tax rate, employment regulation) vary less across cities/states.
- Implication: substantial heterogeneity suggests cities/states can learn from best performers; reforming regulations, improving institutions, and developing infrastructure could enhance corporate profitability.

### V. Firm-level regressions of business environment on profitability (Table 4 summary)
- Regression specification:
  - profitability_{j,t} = a1 doing business indicators_{j,t} + a2 infrastructure proxy_{j,t} + controls + ε_{j,t}
  - Controls: time dummies and other controls (leverage and liquidity); estimation by first differences and fixed effects.
  - Infrastructure proxy primarily: phone density (percentage of telephone connections, including cell phones).
- Selected coefficients (across first-difference and fixed-effect estimators; some columns restricted to exporters):
  - costs of starting business: examples include -0.0030 (0.0028), -0.0035 (0.0024), -0.029** (0.013), -0.034*** (0.011), -0.33*** (0.094), -0.30*** (0.096).
  - costs of registering property: examples include 0.0099 (0.013), -0.023** (0.011), -0.21** (0.10).
  - costs of enforcing contracts: examples include -0.0087* (0.0051), -0.12*** (0.040), -0.080** (0.037).
  - time to export and costs to export: some negative and significant estimates for exporters (e.g., -0.0084* (0.0046)).
  - phone density: several negative coefficients (e.g., -0.0022 (0.0017), -0.00063 (0.0016), -0.033** (0.014), -0.077* (0.040)), suggesting phone density may be a poor proxy for broader infrastructure improvements in this context.
- Interpretation:
  - Results generally support that higher business costs reduce profitability.
  - For exporters, costs to export show a negative and significant coefficient in one specification (10 percent level), suggesting transport/infrastructure improvements may benefit exporters.
  - Mixed evidence on infrastructure proxies implies further work is needed to clarify infrastructure’s role for profitability.

### VI. Robustness checks and alternative-data results (Appendix A)
- Robustness using Thomson Reuters Worldscope replicates main results.
- Table 2 (robustness: selected coefficient patterns):
  - Profitability (q) coefficients reported: 0.024*** (0.0025), 0.040*** (0.0033), 0.028*** (0.0040), 0.060*** (0.015), 0.040*** (0.0068).
  - Alternative q: 0.0013 (0.0012).
  - Liquidity coefficients: 0.016*** (0.0020), 0.035*** (0.0023), 0.050*** (0.0042), 0.087*** (0.028), 0.055*** (0.014), 0.062*** (0.014).
  - Leverage coefficients: -0.035*** (0.0087), -0.21*** (0.014), -0.30*** (0.027), -0.28*** (0.095), -0.16** (0.073), -0.26*** (0.068).
  - Stock price volatility examples: -0.00090*** (0.00032), -0.00074** (0.00035).
  - Lag of i/k examples: 0.42*** (0.011), 0.22*** (0.0091), -0.23*** (0.012), 0.38*** (0.029), 0.35*** (0.023), 0.36*** (0.022).
  - Number of observations across specifications (as presented): 12306123068909668689098909 (as presented).
- Table 3 (heterogeneity by sector, size, exporter status):
  - Profitability (q) and liquidity coefficients remain generally positive and significant across many subsamples (examples: profitability 0.048*** (0.0091) in manufacturing full sample; liquidity 0.064*** (0.017) in services).
  - Interaction dummies generally insignificant: dummy manufacturing * liquidity -0.014 (0.037); dummy small * liquidity -0.00067 (0.016); dummy exporters * liquidity 0.012 (0.023).
  - Leverage generally negative across subsamples (examples: -0.19** (0.089), -0.70*** (0.19)).
  - Lag of i/k consistently positive and significant across many subsamples (examples: 0.41*** (0.027), 0.36*** (0.024)).
  - Number of observations across subsamples (as presented): 700114698470560333068909743414758909 (as presented).

### VII. Corporate financial indicators and trends (Figure 2 summary)
- Profitability:
  - Median Tobin’s q and Median Return on Assets of the nonfinancial sector in India have not returned to pre-crisis levels (FY2001–FY2010 series shown).
- Liquidity:
  - Median Current Ratio and Interest Coverage Ratio of the nonfinancial sector in India are somewhat low relative to comparators (India, Brazil, China, Russia, ASEAN4).
  - Current ratio defined as liquid assets divided by short-term liabilities.
  - Interest coverage ratio defined as interest payments divided by earnings before interest and taxes.
- Leverage:
  - Median Debt to Assets and Median Debt to Cash Flow ratios are on the high side and increasing; the increase is more pronounced relative to cash inflows.
- Figure source: IMF Corporate Vulnerability database.

### VIII. Policy implications, quantitative impact, and conclusions
- Macro and structural factors both affect corporate investment.
- Macro findings and quantitative contribution:
  - High and volatile inflation, and heightened global uncertainty, have quantitatively contributed negatively to corporate investment (estimated net contribution around –1.3 percent of GDP between end-FY2007 and end-FY2011).
  - Monetary easing since the global financial crisis supported corporate investment; monetary tightening since early 2010 may have started hurting investment at the margin.
  - Policy implication: lowering and stabilizing inflation is critical for sustained investment growth.
- Micro-level findings and policy implications:
  - Improving the business environment is essential to stimulate corporate investment.
  - Priority areas: cutting various costs of doing business, improving financial access, developing infrastructure (especially transport), strengthening governance, speeding project approvals.
  - Substantial variability across Indian regions implies potential for intra-country learning.
  - Other important factors highlighted (not exhaustively tested): stable provision of electricity; ease of land acquisition; less restrictive regulations in product and labor markets; simpler administrative procedures; higher quality education and skills.
- Estimated impact of reducing costs of doing business (as reported):
  - Reducing the average of each cost of doing business to the lowest among Indian cities surveyed in 2009 is estimated to:
    - Boost aggregate demand by ¼ to 1½ percent of GDP (Prowess and Worldscope-based ranges reported).
    - Raise corporate investment by 3 to 13½ percent.
  - Alternative-data estimate (Thomson Reuters Worldscope) of aggregate demand impact from reducing each cost to the lowest among Indian cities: "½ to 1½ percent of GDP".
- Table 5 (reported estimates — "Estimated Aggregate Impact of Reducing Costs of Doing Business"):
  - Change in aggregate corporate investment (in percent)
    - 3.1    —13.5 2.2    —9.0
    - costs of starting business 0.6   —    1.8 0.6   —    1.6
    - costs of registering property 0.3   —    1.9 0.1   —    0.5
    - costs of enforcing contracts 1.0   —4.0 0.9   —3.6
    - costs to export 1.2   —    5.8 0.7   —    3.2
  - Direct demand impact on GDP (in percent of GDP)
    - 0.3    —1.5 0.2    —1.0
    - costs of starting business 0.1   —    0.2 0.1   —    0.2
    - costs of registering property 0.03 —0.2 0.0    —0.1
    - costs of enforcing contracts 0.1   —0.4 0.1   —0.4
    - costs to export 0.1   —    0.6 0.1   —    0.3
  - Reducing the average of each cost to: Lowest / Third Lowest (columns implied in source table)
- Summary policy priorities:
  - Reduce volatility and level of inflation to improve investor sentiment.
  - Consider investment impact when designing monetary policy (monetary tightening can hurt investment).
  - Improve business environment: reduce costs of doing business, strengthen governance, speed project approvals.
  - Upgrade the financial system to reduce liquidity constraints.
  - Develop infrastructure, especially transport, to ease supply-side constraints and support exporters.
- Robustness:
  - Appendix A: Regression results using Thomson Reuters Worldscope confirm similarity to main results (Tables A.1 and A.2).
  - Semi-aggregate (city/state) data sources include World Bank Doing Business surveys (2005, 2007, 2009), phone density (CEIC), number of bank offices per square kilometer (Reserve Bank of India and Ministry of Statistics and Programme Implementation), and ratio of electricity supply to demand (CEIC).

*Source: IMF Working Paper (_wp1270) — IV. POLICY ISSUES AND CONCLUSIONS; Appendices A and B (excerpt provided).*

### References .............................................................................................................

### _wp1270 - References

### I. Introduction: recent patterns in investment in India
- Before the global financial crisis, strong corporate investment drove investment growth; corporate investment was 14 percent of GDP before the crisis and slowed to 10 percent of GDP since 2009.
- Growth of investment (gross fixed capital formation):
  - slowed to about –0.6 (quarter on quarter) on average in the first three quarters in 2011;
  - averaged nearly 3 percent between 2000–2007.
- High-frequency capital goods production data suggest corporate investment may be weakening further (machinery and equipment production declining).
- Scenario example: if the quarterly growth rate of corporate investment stays at –¼ percent (the average growth in investment in 2011), lower demand would reduce medium-term GDP growth to around 7 percent.
- Potential drivers of weak corporate investment discussed:
  - Macroeconomic uncertainty (high and volatile inflation, weaker global outlook).
  - Monetary tightening since early 2010.
  - Structural factors: unfavorable business environment, weakening governance, slower government project approvals, high costs of doing business.

### II. Analysis of macroeconomic data: model, key coefficients, and contributions
- Estimated macro model (annual data, FY1992–2011) for corporate investment:
  - CI_t = a1 X_{t–1} + a2 VIX_t + a3 CI_{t–1} + ε_t
  - X includes volatility of the inflation rate (CPI), the inflation rate (CPI), real GDP growth, the real interest rate, and the world’s real GDP growth.
  - VIX used to control for global stock volatility.
  - Volatility of the inflation rate calculated by the standard deviation of monthly inflation rates (month on month, seasonally adjusted) in the year.
  - Real interest rate measure: average of nominal prime lending rates of ICICI Bank and IDBI Bank deflated by CPI inflation.
- Main macro findings (Table 1 and text):
  - The coefficient on the volatility of inflation is consistently negative and statistically significant in most specifications.
  - Coefficients on the real interest rate and global stock volatility are negative and statistically significant.
  - Other explanatory variables generally have expected signs but are often statistically insignificant.
- Selected coefficient estimates and contribution (as reported in Table 1):
  - lag of volatility of inflation rate: -2.79* (standard error 1.32), -2.62* (1.31), -1.59 (1.08) across specifications.
  - lag of inflation rate: -0.11 (0.16), -0.051 (0.17), -0.21* (0.11).
  - lag of real GDP growth: 0.10 (0.22), 0.14 (0.17), -0.03 (no SE in table snippet).
  - lag of real GDP growth ex agriculture: 0.28 (0.32) [single entry].
  - lag of real interest rate: -0.51** (0.22), -0.45* (0.23), -0.34* (0.17).
  - lag of world's real GDP growth: 0.36 (0.31), 0.37 (0.31).
  - global stock volatility: -0.11*** (0.035) in one specification.
  - lag of corporate investment (in % of GDP): 0.65*** (0.19), 0.55** (0.23), 0.85*** (0.14).
  - Total contribution to corporate investment between end-FY07 and end-FY11 (using coeffs in specification (3)): -1.3 (% of GDP).
- Quarterly analysis (using gross fixed capital formation as dependent variable due to lack of quarterly corporate investment series):
  - Model: I_t = a1 X_{t–1} + a2 X_{t–2} + a3 VIX_t + a4 I_{t–1} + a5 I_{t–2} + ε_t
  - I is gross fixed capital formation (in percent of GDP); X includes quarterly measures of volatility of inflation (CPI), inflation (CPI), real GDP growth, real interest rate, and world real GDP growth; VIX included.
  - Result: standard macro variables largely explain investment, but since late 2010 actual investment has fallen short of levels predicted by these variables, suggesting additional factors (for example, business environment) not captured by standard macro variables.

### III. Analysis of firm-level micro panel data: determinants of corporate investment
- Micro model estimated (firm j, year t):
  - i_{j,t}/k_{j,t} = a1 profitability_{j,t–1} + a2 liquidity_{j,t–1} + a3 leverage_{j,t–1} + a4 i_{j,t–1}/k_{j,t–1} + control vars + ε_{j,t}
  - Profitability measured by Tobin’s q.
  - Liquidity measured by ratio of liquid assets to capital k (alternative liquidity measures, e.g., current ratio, do not change results).
  - Leverage measured by ratio of debt to total assets.
  - Firm stock price volatility and time dummies included as controls.
  - Estimation method: Arellano–Bond Dynamic Panel GMM Estimator (Dynamic GMM) after first-differencing.
- Data source and sample:
  - Prowess data set provided by the Centre for Monitoring Indian Economy (CMIE) covering listed and unlisted companies.
  - Sample after selection: 12,306 observations (2,291 companies) over 1990–2011; sample restricted to nonfinancial companies.
- Main micro findings:
  - Profitability, liquidity, and leverage are key determinants of corporate investment.
    - Coefficients on profitability and liquidity are positive and generally significant.
    - Coefficient on leverage is negative and significant.
  - Interaction tests:
    - Interaction of liquidity with manufacturing dummy, firm size dummy, or exporter dummy is statistically insignificant (columns referenced: 3rd, 6th, 9th), implying the positive liquidity effect does not differ significantly by these firm characteristics in the reported specifications.
  - Interpretation:
    - Positive liquidity coefficients imply remaining frictions in financial access; in perfect financial markets liquidity would not affect investment.
    - Raising profitability is critical to stimulate investment in India.
  - Policy-relevant implication from micro evidence:
    - Improving the business environment, upgrading the financial system, and developing infrastructure could stimulate corporate investment by increasing profitability and easing financial constraints.

### IV. Synthesis and policy implications (as presented)
- Both macroeconomic and structural factors affect corporate investment in India.
- Macroeconomic factors (high and volatile inflation, global uncertainty, higher real interest rates) have quantitatively contributed negatively to corporate investment (estimated net contribution around –1.3 percent of GDP between end-FY2007 and end-FY2011).
- Macro variables explain much of historical corporate investment, but recent shortfalls relative to macro-based predictions since late 2010 point to additional structural factors, notably the business environment.
- Policy directions emphasized by the analysis:
  - Reduce volatility and level of inflation to improve investor sentiment.
  - Consider the investment impact of monetary policy tightening.
  - Improve business environment (reduce costs of doing business, strengthen governance, speed project approvals).
  - Upgrade the financial system to reduce liquidity constraints.
  - Develop infrastructure to ease supply-side constraints.

*Source: _wp1270 - References (excerpt provided).*

### Appendix A conducts a robustness check, using an alternative data set (Thomson Reuters Worldscope).

### _wp1270 - Appendix A conducts a robustness check, using an alternative data set (Thomson Reuters Worldscope)

### Robustness check and estimation of investment function (Table 2)
- Estimation methods reported: OLS, Fixed effects, First difference, FD+IV (dynamic GMM).
- Profitability (q) coefficients:
  - 0.024*** (0.0025)
  - 0.040*** (0.0033)
  - 0.028*** (0.0040)
  - 0.060*** (0.015)
  - 0.040*** (0.0068)
- Alternative q:
  - 0.0013 (0.0012)
- Liquidity coefficients:
  - 0.016*** (0.0020)
  - 0.035*** (0.0023)
  - 0.050*** (0.0042)
  - 0.087*** (0.028)
  - 0.055*** (0.014)
  - 0.062*** (0.014)
- Leverage coefficients:
  - -0.035*** (0.0087)
  - -0.21*** (0.014)
  - -0.30*** (0.027)
  - -0.28*** (0.095)
  - -0.16** (0.073)
  - -0.26*** (0.068)
- Stock price volatility:
  - -0.00090*** (0.00032)
  - -0.00074** (0.00035)
  - -0.000029 (0.00033)
  - 0.00071 (0.00050)
  - -0.0053 (0.0035)
  - -0.0052* (0.0031)
- Lag of i/k coefficients:
  - 0.42*** (0.011)
  - 0.22*** (0.0091)
  - -0.23*** (0.012)
  - 0.38*** (0.029)
  - 0.35*** (0.023)
  - 0.36*** (0.022)
- Number of observations across specifications: 12306123068909668689098909 (as presented).

Notes from Table 2:
- Source: CMIE Prowess.
- Significance notation: *** p<0.01, ** p<0.05, * p<0.1. Robust standard errors in parentheses.
- Instruments for FD+IV/dynamic GMM: 2 and 3 period lags of profitability, liquidity, leverage, and the dependent variable.
- Alternative q defined as (market value of equities + total debt - current assets)/replacement costs of capital (from Nabar and Syed, 2011).

### Heterogeneity by sector, size, and export status (Table 3)
- Profitability (q) coefficients across subsamples:
  - Manufacturing full sample + dummy manufacturing * liquidity: 0.048*** (0.0091)
  - Services (first column): 0.019* (0.011)
  - Large firms: 0.039*** (0.0066)
  - Small firms: 0.031*** (0.0084)
  - Exporters: 0.039*** (0.012)
  - Nonexporters: 0.038*** (0.0065)
  - Full sample + dummy small * liquidity: 0.031*** (0.0070)
  - Full sample + dummy exporters * liquidity: 0.079*** (0.023)
  - Other reported: 0.040*** (0.0067)
- Liquidity coefficients across subsamples:
  - Manufacturing full sample: 0.031 (0.020)
  - Services: 0.064*** (0.017)
  - Large firms: 0.056** (0.026)
  - Small firms: 0.057*** (0.019)
  - Exporters: 0.037** (0.016)
  - Nonexporters: 0.050*** (0.016)
  - Full sample + dummy small * liquidity: 0.065*** (0.017)
  - Full sample + dummy exporters * liquidity: 0.087*** (0.021)
  - Additional: 0.059*** (0.022)
- Interaction dummies:
  - dummy manufacturing * liquidity: -0.014 (0.037)
  - dummy small * liquidity: -0.00067 (0.016)
  - dummy exporters * liquidity: 0.012 (0.023)
- Leverage across subsamples (selected):
  - -0.19** (0.089)
  - -0.080 (0.14)
  - -0.16** (0.078)
  - -0.19** (0.078)
  - -0.32** (0.13)
  - -0.20** (0.081)
  - -0.061 (0.077)
  - -0.70*** (0.19)
  - -0.19** (0.074)
- Stock price volatility across subsamples: coefficients generally small and statistically insignificant in many specifications (examples: -0.0045 (0.0035), -0.0036 (0.0045), -0.0057 (0.0040)).
- Lag of i/k across subsamples: consistently positive and significant in many specifications (examples: 0.41*** (0.027), 0.18*** (0.048), 0.36*** (0.024), 0.39*** (0.030), 0.23*** (0.037), 0.35*** (0.023), 0.37*** (0.025), 0.25*** (0.042), 0.35*** (0.023)).
- Number of observations across subsamples (as presented): 700114698470560333068909743414758909 (as presented).

Notes from Table 3:
- Source: CMIE Prowess.
- Robust standard errors in parentheses.
- Instruments: 2 and 3 period lags of profitability, liquidity, leverage, and the dependent variable. For the services regression (second column), only 2 period lags are used to reduce the number of instruments relative to the number of companies.

### Corporate financial indicators and trends (Figure 2 summary)
- Profitability:
  - Median Tobin’s q and Median Return on Assets of the nonfinancial sector in India have not returned to pre-crisis levels (FY2001–FY2010 series shown).
- Liquidity:
  - Median Current Ratio and Interest Coverage Ratio of the nonfinancial sector in India are somewhat low relative to comparators (India, Brazil, China, Russia, ASEAN4).
  - Current ratio defined as liquid assets divided by short-term liabilities.
  - Interest coverage ratio defined as interest payments divided by earnings before interest and taxes.
- Leverage:
  - Median Debt to Assets and Median Debt to Cash Flow ratios are on the high side and increasing; the increase is more pronounced relative to cash inflows.

(Source indicated for the figures: IMF Corporate Vulnerability database.)

### Business environment, competitiveness, and variability within India
- World Bank doing business (2011):
  - India ranked 132nd of 183 countries in ease of doing business in 2011 (up from 139th in 2010).
  - In 2011, India’s ranks for specific indicators: starting a business 166th, registering property 97th, enforcing contracts 182nd (with costs amounting to nearly 40 percent of claims).
- OECD Product Market Regulation (PMR) indicators:
  - PMR shows barriers to entrepreneurship are restrictive in India relative to advanced and emerging economy standards (OECD PMR, 2008 charts referenced).
- Global Competitiveness (World Economic Forum):
  - India’s headline global competitiveness fell to 56th place in 2011 from 49th in 2009.
  - The institutions sub-index ranking declined from 34th in 2006 to 69th in 2011.
- Within-India variability:
  - Doing Business indicators vary significantly across Indian cities (costs of registering property; costs to export in US$ per container).
  - OECD Product Market Regulation indicator shows noticeable differences across states.
  - Some indicators (e.g., corporate tax rate, employment regulation) vary less across cities/states.
- Implication: substantial heterogeneity suggests cities/states can learn from best performers; reforming regulations, improving institutions, and developing infrastructure could enhance corporate profitability.

### Firm-level panel analysis of business environment on profitability (Table 4)
- Regression specification:
  - profitability_j,t = a1 doing business indicators_j,t + a2 infrastructure proxy_j,t + control vars + ε_j,t
  - Controls: time dummies and other controls (leverage and liquidity). Estimation by first differences and fixed effects to address unobservable firm-specific factors.
  - Infrastructure proxy used: phone density (percentage of telephone connections, including cell phone connections). Alternative proxies tested (number of bank offices per square kilometer, ratio of electricity supply to demand) did not consistently yield expected positive signs.
- Selected coefficients (reported across first-difference and fixed-effect estimators; several columns restricted to exporters):
  - costs of starting business:
    - -0.0030 (0.0028)
    - -0.0035 (0.0024)
    - -0.029** (0.013)
    - -0.034*** (0.011)
    - -0.022 (0.020)
    - -0.023 (0.021)
    - -0.33*** (0.094)
    - -0.30*** (0.096)
  - costs of registering property:
    - 0.0099 (0.013)
    - -0.023** (0.011)
    - -0.0042 (0.034)
    - 0.0019 (0.030)
    - -0.094 (0.094)
    - -0.21** (0.10)
    - 0.220 (0.25)
    - 0.24 (0.25)
  - costs of enforcing contracts:
    - -0.0087* (0.0051)
    - -0.0053 (0.0041)
    - -0.028 (0.018)
    - -0.015 (0.017)
    - -0.12*** (0.040)
    - -0.080** (0.037)
    - -0.18 (0.14)
    - -0.21 (0.15)
  - time to export:
    - -0.0090 (0.0072)
    - -0.0047 (0.0069)
    - -0.053 (0.052)
    - -0.060 (0.060)
  - costs to export:
    - -0.00097 (0.00063)
    - -0.00051 (0.00068)
    - -0.0084* (0.0046)
    - -0.0079 (0.0060)
  - phone density:
    - -0.0022 (0.0017)
    - -0.00063 (0.0016)
    - -0.0083* (0.0048)
    - -0.0028 (0.0053)
    - -0.033** (0.014)
    - -0.027* (0.015)
    - -0.077* (0.040)
    - -0.088* (0.046)
- Number of observations across Table 4 columns as presented: 217335529701888204931479321688.
- Notes:
  - Source: CMIE Prowess.
  - Significance notation: *** p<0.01, ** p<0.05, * p<0.1. Robust standard errors in parentheses.
  - Alternative q definition noted as in Table 2.

Interpretation of Table 4:
- Results generally support hypothesis that higher business costs reduce profitability: costs of starting business, registering property, and enforcing contracts have expected negative signs and are significant in many specifications.
- For exporters, costs to export show a negative and significant coefficient in one specification (significant at the 10 percent level), suggesting transport/infrastructure improvements may benefit exporters.
- Phone density shows a negative sign in many specifications, suggesting it may be a poor proxy for broader infrastructure improvements in this context.

### Policy implications and quantitative impact (summary)
- Improving the business environment could substantially boost corporate investment and aggregate demand:
  - Reducing the average of each cost of doing business to the lowest among Indian cities surveyed in 2009 is estimated to:
    - Boost aggregate demand by ¼ to 1½ percent of GDP.
    - Raise corporate investment by 3 to 13½ percent.
  - Of the business costs considered, lowering the average costs to export is estimated to be the most effective for increasing GDP (text truncated at this point).

### Additional contextual findings and literature
- Evidence from prior studies leveraging within-country variability is summarized (Box 1), including Besley and Burgess (2004), Purfield (2006), Kochhar et al. (2006), Topalova (2008), Conway, Herd, and Chalaux (2008), Topalova and Khandelwal (2011).
- While there is some evidence that developing infrastructure could boost profitability, the paper notes mixed results depending on the infrastructure proxy and calls for further work to reach clearer conclusions about the relevance of infrastructure for corporate profitability.

*Source: CMIE Prowess.*

### 0.1 to 0.6 percent. Reducing the average of each of the other costs to the lowest could raise

### IV. POLICY ISSUES AND CONCLUSIONS

### Estimated impact of reducing costs of doing business
- Reducing the average of each of the other costs to the lowest could raise GDP by "0.03 to 0.4 percent" each.
- Using an alternative data set (Thomson Reuters Worldscope), reducing the average of each cost of doing business to the lowest among Indian cities is estimated to raise aggregate demand by "½ to 1½ percent of GDP", a similar range to that estimated using Prowess (Table 5).

- Table 5 (as reported in source) — "Estimated Aggregate Impact of Reducing Costs of Doing Business":
  - Change in aggregate corporate investment (in percent)
    - 3.1    —13.5 2.2    —9.0
    - costs of starting business 0.6   —    1.8 0.6   —    1.6
    - costs of registering property 0.3   —    1.9 0.1   —    0.5
    - costs of enforcing contracts 1.0   —4.0 0.9   —3.6
    - costs to export 1.2   —    5.8 0.7   —    3.2
  - Direct demand impact on GDP (in percent of GDP)
    - 0.3    —1.5 0.2    —1.0
    - costs of starting business 0.1   —    0.2 0.1   —    0.2
    - costs of registering property 0.03 —0.2 0.0    —0.1
    - costs of enforcing contracts 0.1   —0.4 0.1   —0.4
    - costs to export 0.1   —    0.6 0.1   —    0.3
  - Reducing the average of each cost to: Lowest / Third Lowest (columns implied in source table)

### Macroeconomic findings and policy implications
- Findings:
  - High and volatile inflation, and heightened global uncertainty, may have dampened corporate investment.
  - Monetary easing since the global financial crisis provided important support for corporate investment.
  - Monetary tightening since early 2010 may have started hurting corporate investment at the margin.
- Policy implication:
  - Lowering and stabilizing inflation is critical for sustained investment growth.

### Micro-level findings and policy implications (business environment)
- Findings from micro panel data:
  - Improving the business environment is essential to stimulate corporate investment.
  - India has considerable room to improve its business environment; priority areas include cutting various costs of doing business and improving financial access.
  - Some evidence suggests developing infrastructure, especially transport, could support corporate investment.
  - Substantial variability across Indian regions implies potential for intra-country learning.
  - Many business-environment-related factors not tested in the paper may also be important (only factors that vary enough within India and have multi-year data were examined).
- Examples of other important factors (not exhaustively tested in the paper but highlighted):
  - Stable provision of electricity
  - Ease of land acquisition
  - Less restrictive regulations in product and labor markets
  - Simpler administrative procedures
  - Higher quality education and skills
- Comparative note:
  - In many of these areas, India falls behind other emerging economies.

### Robustness and regression results (appendix highlights)
- Appendix A: Regression results using Thomson Reuters Worldscope confirm similarity to main results (Tables A.1 and A.2).
- Based on coefficients in Tables A.1 and A.2, the alternative-data estimate of aggregate demand impact from reducing each cost to the lowest among Indian cities is "½ to 1½ percent of GDP".
- Table A.1 (selected coefficient patterns as reported):
  - profitability q: 0.015*** 0.033*** 0.025*** 0.047*** 0.032*** (with standard errors in parentheses in source)
  - alternative q: 0.0041*** (0.0011)
  - liquidity: 0.018*** 0.040*** 0.045*** 0.067** 0.044** 0.041* (standard errors in source)
  - leverage: -0.082*** -0.29*** -0.34*** -0.28* -0.25*** -0.26*** (standard errors in source)
  - stock price volatility: 0.00055** 0.00070 0.00062 0.0012 0.0063** 0.0077** (standard errors in source)
  - lag of i/k: 0.42*** 0.12*** -0.27*** 0.26*** 0.30*** 0.30*** (standard errors in source)
  - Num of observations: 5516 5516 4013 1287 840 314031 (as reported in source table)
- Table A.2 (selected reported coefficients for profitability regressions):
  - costs of starting business: -0.015** -0.013*** -0.012 -0.0021 -0.053 -0.043 -0.020 -0.060 (standard errors in source)
  - costs of registering property: -0.049** -0.056*** -0.033 -0.036 -0.30** -0.37** -0.11 -0.051 (standard errors in source)
  - costs of enforcing contracts: -0.0057 -0.0057 -0.058* -0.059** -0.12** -0.14** -0.38** -0.28 (standard errors in source)
  - costs to export: -0.0019* -0.0018* -0.014** -0.0099 (standard errors in source)
  - phone density: -0.00070 -0.00065 -0.015 -0.016** -0.038* -0.029 -0.13** -0.099 (standard errors in source)
  - Num of observations (as reported): 1182 2048 3887 5511 4919 4437 5720
- Significance notation used in source: *** p<0.01, ** p<0.05, * p<0.1. Instruments and alternative q definitions are as described in source tables.

### Semi-aggregate data used (appendix highlights)
- Data sources and variables used at city- or state-level:
  - Doing business indicators: World Bank Doing Business surveys (2005, 2007, 2009). Main indicators include number of administrative procedures; time needed to finish procedures; costs. The survey covered 9, 12, and 17 Indian cities in 2005, 2007, and 2009, respectively.
  - Phone density: CEIC; defined as the percentage of telephone connections, including cell phone connections; state-level variable.
  - Number of bank offices per square kilometer: calculated from Reserve Bank of India data on bank offices and Ministry of Statistics and Programme Implementation data on state areas.
  - Ratio of electricity supply to demand: CEIC; state-level variable.
- Table B.1 (main Doing Business indicators reported in source):
  - Starting business: Procedures (number) ✓✓; Time (days) ✓✓✓; Cost (% of income per capita) ✓✓✓
  - Dealing with construction permits: Procedures (number) ✓✓; Time (days) ✓✓; Cost (% of income per capita) ✓✓
  - Employment regulation: Rigidity of employment index ✓✓; Cost of firing (weeks of wages) ✓✓
  - Registering property: Procedures (number) ✓✓; Time (days) ✓✓✓; Cost (% of property value) ✓✓✓
  - Paying taxes: Payments (number) ✓✓; Time (hours) ✓✓; Total tax rate (% of profit) ✓✓
  - Trading across borders: Documents for export (number) ✓✓; Time to export (days) ✓✓; Cost to export (US$ per container) ✓✓
  - Enforcing contract: Procedures (number) ✓✓; Time (days) ✓✓✓; Cost (% of claim) ✓✓✓
  - Closing business: Time (years) ✓✓✓; Cost (% of estate) ✓✓✓
- Table B.2 (cities covered in Doing Business survey as reported): Ahmedabad, Bengaluru, Bhubaneshwar, Chandigarh, Chennai, Gurgaon, Guwahati, Hyderabad, Indore, Jaipur, Kochi, Kolkata, Lucknow, Ludhiana, Mumbai, New Delhi, Noida, Patna, Ranchi — coverage by year indicated in source table; Num of cities covered 9 12 17.

_Italic: Source: IMF Working Paper (content unit: IV. POLICY ISSUES AND CONCLUSIONS; Appendices A and B) as provided in the supplied PDF content._

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp1270.pdf_