## _wp1568

## Source details

**Canonical URL:** [_wp1568](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp1568.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp1568.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp1568.pdf.json)

---

### I. Introduction — overview and motivation
- China grew at an average rate of almost 10 percent per year since the late 1970s, raising 660 million people out of poverty.
- Per capita income increased from $320 in 1980 to about $5,500 in 2012.
- The population living on less than $1.25 a day declined from 85 percent of the population in 1980 to 11 percent by 2012.
- Pre-tax market income Gini coefficient: 0.28 in 1980, 0.44 in 2000, and 0.52 by 2013.
- Regional variation in inequality is significant.
- Growth incidence (1980–2012): cumulative increase of 331 percent for the lowest-income quintile and 1,042 percent for the highest-income quintile.
- Income shares (1980 vs. 2012): top quintile captured 47 percent of total income (up from 38 percent in 1980); lowest quintile accounted for 4.7 percent (down from 8.7 percent).
- Fiscal redistribution (market minus net Gini): averaged 1.7 Gini-index points in the 1990s; averaged -1.1 Gini-index points over 2000–2013.
- Distribution of taxes (2012): taxes accounted for 10.8 percent of annual income among the bottom decile households (13.3 percent among the bottom five percent) and 8.7 percent among the top decile households.
- Paper objective: identify proximate determinants of income inequality in China (1980–2013) with focus on distributional effects of fiscal policy; address endogeneity using international tourist arrivals as an IV for real GDP per capita; complement single-country time-series with a BRIC+ panel (Brazil, Russia, India, China, plus 30 other emerging market economies) using IV-GMM.

### II. Comparative context — China and the world
- World (unweighted) market Gini coefficient: increased from 0.41 in the 1980s to 0.43 by 2007 and to 0.45 in 2013.
- World (unweighted) net Gini coefficient: increased from 0.36 in the 1980s to 0.38 by 2013.
- China: market and net Gini increased from averages of 0.30 and 0.29 in the 1980s to 0.52 and 0.53 by 2013.
- Income share of the top 1 percent in China: 2.8 percent in 1980 and 4.9 percent in 2009.
- Global fiscal redistribution (gross minus net Gini): declined from 7.1 Gini-index points in the 1980s to 6.3 Gini-index points in the 1990s, recovered to 6.9 Gini-index points in the 2000s.
- China’s fiscal redistribution: average of 1 Gini-index points in the 1980s, 1.7 Gini-index points in the 1990s, and -1.1 Gini-index points over 2000–13.

### III. Literature synthesis — determinants of inequality
- Kuznets hypothesis: inverted U-shaped relationship between income inequality and development.
- Empirical evidence mixed across studies.
- Determinants emphasized in the literature:
  - Inflation tends to increase inequality.
  - Globalization/trade openness: ambiguous theoretical effects; mixed empirical findings.
  - FDI: often associated with higher inequality; possible inverted-U relation.
  - Financial development: may lower inequality over long term but worsen it at early stages.
  - Demographics and urbanization: affect inequality via age composition and rural-urban migration.
  - Education: higher attainment generally linked to greater equality, subject to distribution and quality.
  - Fiscal policy: level and progressivity of taxes and targeting of expenditure influence net income inequality.

### IV. China’s fiscal policy synopsis — structure and trends
- Fiscal decentralization: provincial governments now account for 53 percent of general government revenues and 85 percent of expenditures.
- Tax-to-GDP ratio: doubled from less than 10 percent in the early 1990s to 19 percent in 2013; OECD average ≈ 35 percent.
- Tax composition: more than 50 percent of revenues from indirect taxes; personal income taxes amount to 6 percent of total tax revenues (and 1.1 percent of GDP); indirect taxes ≈ 10 percent of GDP.
- Personal income tax: top statutory rate 45 percent; effective number of personal income taxpayers is less than 3 percent of the working population.
- Government spending: rose from 18 percent of GDP in 1990 to 29 percent in 2013; still below OECD average of 45 percent.
- Off-budget local government spending: about 15 percent of GDP.
- Composition of spending: increase concentrated in infrastructure investment and public administration; social spending ≈ 6 percent of GDP.
- Social programs: expanded minimum subsistence allowance (dibao) and a new rural pension scheme with limited coverage and benefit levels (rural pension income example: 12,000 yuan annual for rural households vs. 33,000 yuan for urban households; retirement insurance coverage: 34.5 percent rural vs. 87 percent urban, from China Household Finance Survey 2010).
- Subnational fiscal relations: subnational governments responsible for over half of spending but have limited revenue-raising capacity and unequal per capita allocations.

### V. Data description and measurement
- Sample: annual data 1980–2013 covering China and a panel of 33 BRIC+ countries.
- Main dependent variable: net Gini coefficients (after taxes and transfers) from the Standardized World Income Inequality Database (SWIID).
- SWIID provides market and net Gini on annual basis using multiple-imputation; subject to measurement uncertainty.
- China-specific explanatory variables: real GDP per capita (instrumented by international tourist arrivals), tax revenues as percent of GDP, government spending as percent of GDP.
- Panel additional controls: trade openness, financial development (domestic credit to private sector as percent of GDP), index of human capital per person, urbanization, old-age dependency.
- Key China descriptive statistics:
  - Net Gini coefficient: Obs. 34; Mean 40.33; Median 40.3; St. Dev. 9.52; Minimum 27.3; Maximum 53.6.
  - Real GDP per capita: Obs. 34; Mean 4,440.33; Median 229.0; St. Dev. 3,640.98; Minimum 6.6; Maximum 13,164.4.
  - International tourist arrivals (Millions): Obs. 34; Mean 27.8; Median 23.3; St. Dev. 18.2; Minimum 5.7; Maximum 57.7.
  - Government spending (percent of GDP): Obs. 34; Mean 19.4; Median 18.4; St. Dev. 5.2; Minimum 10.7; Maximum 29.1.
  - Tax revenue (percent of GDP): Obs. 34; Mean 14.6; Median 14.6; St. Dev. 3.2; Minimum 9.3; Maximum 22.2.
- Panel (BRIC+) selected statistics:
  - Net Gini coefficient: Obs. 1,012; Mean 40.04; St. Dev. 9.26; Min 19.2; Max 62.6.
  - Real GDP per capita (USD): Obs. 1,073; Mean 4,492.62; St. Dev. 4,443.42; Min 21.7; Max 15,533.8.
  - International tourist arrivals (Millions): Obs. 792; Mean 6.8; St. Dev. 3.3; Min 0.0; Max 57.7.
  - Government spending (percent of GDP): Obs. 665; Mean 28.1; St. Dev. 6.7; Min 0.0; Max 55.4.
  - Tax revenue (percent of GDP): Obs. 563; Mean 18.5; St. Dev. 7.9; Min 3.9; Max 50.0.
  - Trade openness (percent of GDP): Obs. 963; Mean 62.1; St. Dev. 53.4; Min 11.5; Max 321.1.
  - Domestic credit to private sector (percent of GDP): Obs. 1,031; Mean 39.8; St. Dev. 32.3; Min 1.2; Max 167.5.
  - Index of human capital per person: Obs. 920; Mean 2.3; St. Dev. 0.5; Min 1.3; Max 3.3.
  - Urban population (percent of total): Obs. 1,156; Mean 57.5; St. Dev. 17.6; Min 18.3; Max 95.0.

### VI. Empirical strategy
- China time-series IV approach: two-stage least squares and GMM using international tourist arrivals as instrument for real GDP per capita.
- Panel (BRIC+) dynamic IV-GMM: includes lagged dependent variable to capture persistence; controls for country and time effects; accounts for serial correlation and endogeneity.
- Specification highlights:
  - China model: net Gini, real GDP per capita (and squared), government spending (% GDP), tax revenue (% GDP).
  - BRIC+ model adds trade openness, financial development, human capital, urbanization.
- IV rationale: international tourist arrivals correlated with GDP per capita and plausibly exogenous to inequality in China; caution in BRIC+ where tourism may be economically significant.
- Panel diagnostics: Wooldridge-Drukker test detects first-order serial correlation; GMM applied to address dynamic panel bias and endogeneity.

### VII. Main empirical findings
- Kuznets (inverted U-shaped) relationship between income inequality and per capita income supported:
  - China IV: Log(Real GDP per Capita) positive and significant; (Log(Real GDP per Capita))^2 negative and significant.
  - BRIC+ IV-GMM: Log(Real GDP per Capita) positive and significant; squared term negative and significant across specifications.
- China-specific fiscal effects (IV two-stage least squares):
  - Log(Government spending as percent of GDP): coefficient 0.152 (statistically significant at 5 percent) — government spending associated with worsening income inequality in China.
  - Log(Tax revenue as percent of GDP): coefficient -0.085 (statistically significant at 10 percent) — taxation associated with improving income distribution in China.
  - Adjusted R2 reported as 0.959 and 0.963 in model variants.
  - HAC score Chi 25.058* and 6.197**; HAC regression F27.084*** and 39.810*** as shown in table.
  - Sample period for China IV: 1974-2013.
- BRIC+ panel fiscal effects (IV-GMM):
  - Log(Government spending as percent of GDP): small negative coefficients (e.g., -0.009, -0.001) but statistically insignificant.
  - Log(Tax revenue as percent of GDP): small negative coefficients (e.g., -0.004, -0.013) but statistically insignificant.
  - High persistence in net Gini: coefficient on Log(Net Gini Coefficient) t-1 about 0.919***, 0.874***, 0.894*** across models.
  - Trade openness and human capital show signs of improving distribution but not statistically significant; financial development shows worsening sign but not statistically significant; urbanization sign points to worsening distribution but not statistically significant.
  - BRIC+ sample period: 1980-2013; Wald X2 statistics highly significant across specifications (10352.73***, 3429.81***, 3408.46***).
- Robustness:
  - China results robust to GMM IV estimator (results available upon request); collinearity diagnostics indicate VIF 1.24 for government spending and tax revenue.
  - Panel baseline IV-GMM findings robust to inclusion of control variables.

### VIII. Fiscal structure and tax progressivity (additional detail)
- China’s tax-to-GDP ratio almost doubled over the past two decades to 19 percent, but it remains significantly below the OECD average of about 35 percent.
- China collects more than half of its revenues from indirect taxes.
- Personal income taxes amount to 6 percent of total tax revenues.
- Indirect taxes on goods and services account for over 50 percent of total tax revenues.
- The personal income tax has a top rate of 45 percent, but broad tax brackets and a generous allowance schedule diminish effective progressivity.
- The effective number of personal income tax payers is less than 3 percent of the working population.

### IX. Government spending and benefit incidence (additional detail)
- Government spending grew from 18 percent of GDP in 1990 to 29 percent in 2013, still below the OECD average of 45 percent.
- Social protection and healthcare account for about 6 percent of GDP in China (compared to an average of 15 percent in OECD countries and 9 percent in upper-middle income countries).
- Excluding social protection and healthcare, China’s non-redistributive government spending is comparable to that in OECD countries.
- Incidence of public services and transfers favors high-income groups in urban areas:
  - The top quartile of households received about 80 percent of pension spending, compared with only 2 percent for the bottom quartile.
- Subnational governments are responsible for more than half of total spending but have limited revenue-raising capacity and substantial differences in per capita allocations for basic public services.
- Rural programs have limited coverage and provide a low level of income relative to urban workers.
- Rural households account for over 95 percent of poor households in China.

### X. Policy implications and recommendations
- Fiscal policy has contributed to China’s changing income distribution but has not offset other forces widening inequality.
- Policy recommendations:
  - Broaden the tax base and make the tax system more progressive; shift from indirect to direct taxation to narrow income inequality.
    - Strengthen tax administration and broaden personal income tax, including capital gains, to increase effective taxation of the rich.
    - Impose VAT on services that tend to be consumed more by the rich.
    - Lower high labor taxation that hurts low- and middle-income brackets more than the rich.
    - Increase direct taxes on capital and wealth, especially through more effective land and property taxation.
    - The planned extension of a recurrent property tax from pilot implementation to the rest of the country is identified as a step to generate additional revenues and improve progressivity.
  - Improve progressivity through well-targeted spending programs to expand access to education, healthcare, and social services, particularly in rural areas.
    - Realign expenditure assignments with revenue sources across all layers of government given the decentralized fiscal system.
    - Expand the social safety net including means-tested income support to the poor and unemployment insurance.
    - Reduce untargeted energy subsidies that tend to benefit the rich more than the poor.
  - Reform the pension system to improve redistributive outcomes while ensuring sustainability:
    - Redesign eligibility criteria for the basic retirement pension.
    - Implement structural and parametric changes such as pooling of provincial-level pension funds and adjusting the retirement age and replacement rates.
- Fiscal sustainability considerations:
  - Redistributive measures need to be consistent with fiscal sustainability, particularly given the fiscal cost of rapid population aging.
  - As of 2011, post-tax subsidies for petroleum products, electricity, natural gas and coal account for over 3.8 percent of GDP in China—almost four times the amount of healthcare spending.
  - Fiscal policy is one component of an inclusive growth strategy that must be complemented by a comprehensive range of structural reforms.

### XI. Summary conclusions
- Income inequality in China rose sharply alongside rapid growth; the Kuznets inverted U-shape is supported empirically for China and the BRIC+ panel.
- In China, government spending as currently allocated is associated with worsening inequality, while taxation shows a redistributive effect.
- In the BRIC+ panel, government spending and taxation exhibit redistributive signs but are statistically insignificant in the baseline IV-GMM estimates.
- Appropriate fiscal redistribution can reduce net income inequality and contribute to balanced and sustainable growth.

*Source: _wp1568 - References — IMF working paper content (excerpts provided).*

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

### _wp1568 - References ...............................................................................................................

### I. Introduction — overview and motivation
- China grew at an average rate of almost 10 percent per year since the late 1970s, raising 660 million people out of poverty.
- Per capita income increased from $320 in 1980 to about $5,500 in 2012.
- The population living on less than $1.25 a day declined from 85 percent of the population in 1980 to 11 percent by 2012.
- Pre-tax market income Gini coefficient: 0.28 in 1980, 0.44 in 2000, and 0.52 by 2013.
- Regional variation in inequality is significant.
- Growth incidence (1980–2012): cumulative increase of 331 percent for the lowest-income quintile and 1,042 percent for the highest-income quintile.
- Income shares (1980 vs. 2012): top quintile captured 47 percent of total income (up from 38 percent in 1980); lowest quintile accounted for 4.7 percent (down from 8.7 percent).
- Fiscal redistribution (market minus net Gini): averaged 1.7 Gini-index points in the 1990s; averaged -1.1 Gini-index points over 2000–2013.
- Distribution of taxes (2012): taxes accounted for 10.8 percent of annual income among the bottom decile households (13.3 percent among the bottom five percent) and 8.7 percent among the top decile households.
- Paper objective: identify proximate determinants of income inequality in China (1980–2013) with focus on distributional effects of fiscal policy; address endogeneity using international tourist arrivals as an IV for real GDP per capita; complement single-country time-series with a BRIC+ panel (Brazil, Russia, India, China, plus 30 other emerging market economies) using IV-GMM.

### II. Comparative context — China and the world
- World (unweighted) market Gini coefficient: increased from 0.41 in the 1980s to 0.43 by 2007 and to 0.45 in 2013.
- World (unweighted) net Gini coefficient: increased from 0.36 in the 1980s to 0.38 by 2013.
- China: market and net Gini increased from averages of 0.30 and 0.29 in the 1980s to 0.52 and 0.53 by 2013.
- Income share of the top 1 percent in China: 2.8 percent in 1980 and 4.9 percent in 2009.
- Global fiscal redistribution (gross minus net Gini): declined from 7.1 Gini-index points in the 1980s to 6.3 Gini-index points in the 1990s, recovered to 6.9 Gini-index points in the 2000s.
- China’s fiscal redistribution: average of 1 Gini-index points in the 1980s, 1.7 Gini-index points in the 1990s, and -1.1 Gini-index points over 2000–13.

### III. Literature synthesis — determinants of inequality
- Kuznets hypothesis: inverted U-shaped relationship between income inequality and development.
- Mixed empirical support: some studies find positive correlation between growth and inequality; others find no support for Kuznets.
- Key determinants highlighted in literature:
  - Inflation tends to increase inequality (Datt and Ravallion, 1998; Ferreira, Leite, and Litchfield, 2007).
  - Globalization/trade openness: theory yields ambiguous results; empirical findings mixed (Dollar and Kraay, 2004; Barro, 2000; Milanovic, 2005).
  - FDI: often associated with higher inequality; some evidence of inverted-U relation.
  - Financial development: often lowers inequality over long term but may worsen it at early stages.
  - Demographics and urbanization: changes in age composition and rural-urban migration affect inequality.
  - Education: higher attainment generally linked to greater equality but effects depend on distribution and quality of education.
  - Fiscal policy: level and progressivity of taxes and targeting of expenditure influence net income inequality (Feenberg and Poterba, 1993; Auten and Carroll, 1999; Bastagli, Coady, and Gupta, 2012).

### IV. China’s fiscal policy synopsis — structure and trends
- Fiscal decentralization: provincial governments now account for 53 percent of general government revenues and 85 percent of expenditures.
- Tax-to-GDP ratio: doubled from less than 10 percent in the early 1990s to 19 percent in 2013; OECD average ≈ 35 percent.
- Tax composition: more than 50 percent of revenues from indirect taxes; personal income taxes amount to 6 percent of total tax revenues (and 1.1 percent of GDP); indirect taxes ≈ 10 percent of GDP.
- Personal income tax: top statutory rate 45 percent, but wide brackets and generous allowances result in low effective progressivity; effective number of personal income taxpayers is less than 3 percent of the working population.
- Government spending: rose from 18 percent of GDP in 1990 to 29 percent in 2013; still below OECD average of 45 percent.
- Off-budget local government spending: about 15 percent of GDP.
- Composition of spending: increase concentrated in infrastructure investment and public administration; social spending ≈ 6 percent of GDP.
- Social programs: expansion of minimum subsistence allowance (dibao) and a new rural pension scheme, but coverage and benefit levels remain limited (rural pension income examples: 12,000 yuan annual for rural households vs. 33,000 yuan for urban households; retirement insurance coverage: 34.5 percent rural vs. 87 percent urban, from China Household Finance Survey 2010).
- Subnational fiscal relations: subnational governments responsible for over half of spending but have limited revenue-raising capacity and unequal per capita allocations.

### V. Data description and measurement
- Sample: annual data 1980–2013 covering China and a panel of 33 BRIC+ countries.
- Main dependent variable: net Gini coefficients (after taxes and transfers) from the Standardized World Income Inequality Database (SWIID).
- SWIID characteristics: provides market and net Gini on annual basis using multiple-imputation to standardize sources; preferred for comparability over time and across countries, but subject to measurement uncertainty (confidence intervals illustrated in Figure 4).
- China-specific explanatory variables: real GDP per capita (instrumented by international tourist arrivals), tax revenues as percent of GDP, government spending as percent of GDP.
- Panel (BRIC+) additional controls: trade openness, financial development (domestic credit to private sector as percent of GDP), index of human capital per person, urbanization, old-age dependency.
- Data sources: IMF Government Finance Statistics, International Financial Statistics, World Economic Outlook, World Bank WDI, National Bureau of Statistics of China.
- Unit-root testing: ADF and Ng-Perron tests applied; variables stationary after logarithmic transformation (unit-root results available upon request).

Key descriptive statistics (China, sample sizes and statistics as reported):
- Net Gini coefficient: Obs. 34; Mean 40.33; Median 40.3; St. Dev. 9.52; Minimum 27.3; Maximum 53.6.
- Real GDP per capita: Obs. 34; Mean 4,440.33; Median 229.0; St. Dev. 3,640.98; Minimum 6.6; Maximum 13,164.4.
- International tourist arrivals (Millions): Obs. 34; Mean 27.8; Median 23.3; St. Dev. 18.2; Minimum 5.7; Maximum 57.7.
- Government spending (percent of GDP): Obs. 34; Mean 19.4; Median 18.4; St. Dev. 5.2; Minimum 10.7; Maximum 29.1.
- Tax revenue (percent of GDP): Obs. 34; Mean 14.6; Median 14.6; St. Dev. 3.2; Minimum 9.3; Maximum 22.2.

Panel (BRIC+) descriptive snapshots (selected means):
- Net Gini coefficient: Obs. 1,012; Mean 40.04; St. Dev. 9.26; Min 19.2; Max 62.6.
- Real GDP per capita (USD): Obs. 1,073; Mean 4,492.62; St. Dev. 4,443.42; Min 21.7; Max 15,533.8.
- International tourist arrivals (Millions): Obs. 792; Mean 6.8; St. Dev. 3.3; Min 0.0; Max 57.7.
- Government spending (percent of GDP): Obs. 665; Mean 28.1; St. Dev. 6.7; Min 0.0; Max 55.4.
- Tax revenue (percent of GDP): Obs. 563; Mean 18.5; St. Dev. 7.9; Min 3.9; Max 50.0.
- Trade openness (percent of GDP): Obs. 963; Mean 62.1; St. Dev. 53.4; Min 11.5; Max 321.1.
- Domestic credit to private sector (percent of GDP): Obs. 1,031; Mean 39.8; St. Dev. 32.3; Min 1.2; Max 167.5.
- Index of human capital per person: Obs. 920; Mean 2.3; St. Dev. 0.5; Min 1.3; Max 3.3.
- Urban population (percent of total): Obs. 1,156; Mean 57.5; St. Dev. 17.6; Min 18.3; Max 95.0.

### VI. Empirical strategy
- China time-series IV approach: two-stage least squares and GMM using international tourist arrivals as instrument for real GDP per capita to address reverse causality between inequality and growth.
- Panel (BRIC+) dynamic IV-GMM: includes lagged dependent variable to capture persistence; controls for country and time effects; accounts for serial correlation and endogeneity.
- Specification highlights:
  - China model variables: net Gini, real GDP per capita (and squared), government spending (% GDP), tax revenue (% GDP).
  - BRIC+ model adds trade openness, financial development, human capital, urbanization.
- IV choice rationale: international tourist arrivals correlated with GDP per capita and plausibly exogenous to inequality in China (robust IV tests reported); caution noted where tourism may be economically significant in some BRIC+ countries.
- Panel diagnostics: Wooldridge-Drukker test detects first-order serial correlation; GMM applied to address dynamic panel bias and endogeneity.

### VII. Main empirical findings
- Existence of an inverted U-shaped relationship (Kuznets curve) between income inequality and per capita income:
  - China IV results: Log(Real GDP per Capita) positive and significant; (Log(Real GDP per Capita))^2 negative and significant.
  - BRIC+ IV-GMM results: Log(Real GDP per Capita) positive and significant; squared term negative and significant across specifications.
- China-specific fiscal effects (IV two-stage least squares):
  - Log(Government spending as percent of GDP): coefficient 0.152 (statistically significant at 5 percent) — government spending associated with worsening income inequality in China.
  - Log(Tax revenue as percent of GDP): coefficient -0.085 (statistically significant at 10 percent) — taxation associated with improving income distribution in China.
  - Adjusted R2 reported as 0.959 and 0.963 in model variants; HAC regression F-statistics and score chi statistics reported (HAC score Chi 25.058* and 6.197**; HAC regression F27.084*** and 39.810*** as shown in table).
  - Sample period for China IV: 1974-2013.
- BRIC+ panel fiscal effects (IV-GMM):
  - Log(Government spending as percent of GDP): small negative coefficients (e.g., -0.009, -0.001) but statistically insignificant.
  - Log(Tax revenue as percent of GDP): small negative coefficients (e.g., -0.004, -0.013) but statistically insignificant.
  - High persistence in net Gini: coefficient on Log(Net Gini Coefficient) t-1 about 0.919***, 0.874***, 0.894*** across models.
  - Trade openness and human capital show signs of improving distribution but not statistically significant; financial development shows worsening sign but not statistically significant; urbanization sign points to worsening distribution but not statistically significant.
  - BRIC+ sample period: 1980-2013; Wald X2 statistics highly significant across specifications (10352.73***, 3429.81***, 3408.46***).
- Robustness:
  - China results robust to GMM IV estimator (results available upon request); collinearity diagnostics indicate VIF 1.24 for government spending and tax revenue.
  - Panel baseline IV-GMM findings robust to inclusion of control variables.

### VIII. Policy implications and conclusions
- Fiscal policy has contributed to China’s changing income distribution but has not offset other forces widening inequality.
- Key policy recommendations drawn from empirical findings:
  - Broaden the tax base and make the tax system more progressive; shift from indirect to direct taxation to narrow income inequality.
  - Improve progressivity through well-targeted spending programs to expand access to education, healthcare, and social services, particularly in rural areas (rural households account for over 95 percent of poor households in China).
  - Re-design fiscal policy in China and BRIC+ countries to foster inclusive growth while preserving fiscal sustainability and minimizing distortions and efficiency losses.
- Summary conclusions:
  - Income inequality in China rose sharply alongside rapid growth; the Kuznets inverted U-shape is supported empirically for China and the BRIC+ panel.
  - In China, government spending as currently allocated is associated with worsening inequality, while taxation shows a redistributive effect.
  - In the BRIC+ panel, government spending and taxation exhibit redistributive signs but are statistically insignificant in the baseline IV-GMM estimates.
  - Appropriate fiscal redistribution can reduce net income inequality and contribute to balanced and sustainable growth.

*Source: _wp1568 - References — IMF working paper content (excerpts provided).*

### 1.1 Gini-index points during the period 2000–13.

### 1.1 Gini-index points during the period 2000–13.

### Fiscal structure and tax progressivity
- China’s tax-to-GDP ratio almost doubled over the past two decades to 19 percent, but it remains significantly below the OECD average of about 35 percent.
- China collects more than half of its revenues from indirect taxes.
- Personal income taxes amount to 6 percent of total tax revenues.
- Indirect taxes on goods and services account for over 50 percent of total tax revenues.
- The personal income tax has a top rate of 45 percent, but broad tax brackets and a generous allowance schedule diminish effective progressivity.
- The effective number of personal income tax payers is less than 3 percent of the working population, indicating a high degree of informality and tax avoidance.

### Government spending and benefit incidence
- Government spending grew from 18 percent of GDP in 1990 to 29 percent in 2013, still below the OECD average of 45 percent.
- Social protection and healthcare account for about 6 percent of GDP in China (compared to an average of 15 percent in OECD countries and 9 percent in upper-middle income countries).
- Excluding social protection and healthcare, China’s non-redistributive government spending is comparable to that in OECD countries.
- Incidence of public services and transfers favors high-income groups in urban areas.
  - The top quartile of households received about 80 percent of pension spending, compared with only 2 percent for the bottom quartile.
- Subnational governments are responsible for more than half of total spending but have limited revenue-raising capacity and substantial differences in per capita allocations for basic public services.
- Rural programs (expanded minimum subsistence allowance and a new rural pension scheme) have limited coverage and provide a low level of income compared to urban workers.
- Rural households account for over 95 percent of poor households in China.

### Empirical findings on inequality and fiscal policy
- An increase in real GDP per capita—instrumented by the number of international tourist arrivals—leads to an increase in the net Gini coefficient, while its square term lowers income inequality, confirming an inverted U-shaped relationship between income inequality and economic growth.
- Government spending and taxation have opposing effects on income inequality in the panel analysis:
  - Government spending appears to worsen inequality.
  - Taxation improves the distribution of household income.
- Time-series analysis for China and panel results are broadly consistent, with one exception: in the China time-series both government spending and taxation have the desired redistributive effect, albeit statistically insignificant.
- Overall, the redistributive impact of fiscal policy in China appears stronger than in the BRIC+ panel but the “net” effect is still not enough to compensate for adverse impacts of other influential factors identified in the analysis.

### Policy recommendations to improve redistributive impact
- Broaden the tax system in a more progressive way to help narrow income inequality.
  - Strengthen tax administration and broaden personal income tax, including capital gains, to increase effective taxation of the rich.
  - Impose VAT on services that tend to be consumed more by the rich.
  - Lower high labor taxation that hurts low- and middle-income brackets more than the rich.
  - Increase direct taxes on capital and wealth, especially through more effective land and property taxation.
  - The planned extension of a recurrent property tax from pilot implementation to the rest of the country is identified as a step to generate additional revenues and improve progressivity.
- On the expenditure side, improve progressivity through well-targeted programs that increase access for the poor, particularly in rural areas.
  - Realign expenditure assignments with revenue sources across all layers of government given the decentralized fiscal system.
  - Expand the social safety net including means-tested income support to the poor and unemployment insurance.
  - Reduce untargeted energy subsidies that tend to benefit the rich more than the poor.
- Reform the pension system to improve redistributive outcomes while ensuring sustainability:
  - Redesign eligibility criteria for the basic retirement pension.
  - Implement structural and parametric changes such as pooling of provincial-level pension funds and adjusting the retirement age and replacement rates to ensure long-term sustainability in view of the rise in China’s old-age dependency ratio.

### Fiscal sustainability considerations
- Redistributive measures need to be consistent with fiscal sustainability, particularly given the fiscal cost of rapid population aging.
- As of 2011, post-tax subsidies for petroleum products, electricity, natural gas and coal account for over 3.8 percent of GDP in China—almost four times the amount of healthcare spending.
- Fiscal policy is one component of an inclusive growth strategy that must be complemented by a comprehensive range of structural reforms to sustain economic growth and broaden access to opportunities across society.

*Source: _wp1568 - 1.1 Gini-index points during the period 2000–13.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp1568.pdf_