## _wp16183 - Introduction

## Source details

**Canonical URL:** [_wp16183 - Introduction](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp16183.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp16183.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp16183.pdf.json)

---

### I. Overview and Context
- China achieved GDP growth averaging about 10 percent and per capita income more than tripled over the past two decades.
- Pre-global financial crisis (GFC) growth relied heavily on exports; after WTO entry in 2001 China became the world’s manufacturing hub and the largest exporter by 2008.
- After the GFC, external demand collapsed and China transitioned from an export-based to investment-driven economy, with the investment ratio surging to 45 percent.
- Current account surplus peaked at 10 percent of GDP in 2007 and came down to around 2–3 percent in recent years.
- Credit intensity (new lending needed for an additional unit of output) has more than doubled in the post-GFC period.
- Income inequality rose substantially: Gini index increased from 0.3 in the 1980s to 0.53 in 2013.
- Environmental pollution, including carbon emissions and local air pollution (PM2.5), has worsened significantly.
- The paper identifies an inflection point: continuing the old growth model risks the middle-income trap or a financial crisis; excessive investment and resource misallocation lower efficiency, raise credit intensity, and expand debt to unsustainable levels.

### II. Framework of Rebalancing
- Definition: Rebalancing contains four key elements—external, internal, environmental, and distributional rebalancing—which are closely interlinked.
- External/internal elements:
  - External rebalancing: role of external versus domestic demand.
  - Internal rebalancing (demand side): shifting from investment to consumption.
  - Internal rebalancing (supply side): transitioning from industry to services.
  - Internal rebalancing (input side): reducing credit intensity and improving allocative efficiency.
- Environmental and income distribution:
  - Environmental rebalancing: reduce energy and carbon emission intensity of output and local air pollution (PM2.5).
  - Income distribution rebalancing: increase labor income share of GDP and reduce income inequality.
- Indicators used to track rebalancing:
  - External: current account balance as percent of GDP; contribution of net exports to GDP growth.
  - Internal (demand): nominal share of consumption and investment in GDP; relative contribution to GDP growth in real terms.
  - Internal (supply): nominal and real shares of industry versus services; industrial employment share.
  - Credit: credit intensity; credit to GDP ratio; difference in return on assets (ROA) between state-owned firms and private firms.
  - Environment: energy intensity = electricity consumption per unit of output; carbon emission intensity = carbon emission per unit of output; PM2.5 for local air pollution.
  - Income distribution: Gini index; urban-rural income gap; labor income share of GDP.
- Key interlinkages and stylized facts:
  - High national saving rate (household saving dominant) underlies external and internal imbalances; corporate saving surged in early 2000s but normalized later.
  - When savings exceed domestic investment needs, excess savings flow out producing current account surplus; post-GFC, domestic absorption relied on high investment ratio.
  - Household saving rose from 5 percent in the mid-1970s to 40 percent in recent years, with trend breaks in late 1970s and late 1990s.
  - Drivers of high household saving: demographic change (one-child policy, falling fertility) and precautionary motives (healthcare reform in early 1990s increasing out-of-pocket risk; SOE reform in late 1990s reducing job security).
  - Successful rebalancing requires lower savings and a commensurate fall in investment; given habit persistence, savings decline likely gradual, implying gradual investment adjustment and focus on improving investment efficiency.
  - China’s capital stock per worker is less than 30 percent of U.S. level—capital deepening can continue but at a slower pace; post-crisis capital deepening outpaced income convergence.
  - Regression analysis: changes in the investment ratio alone can explain about 70 percent of variation of credit intensity historically.
  - Service sector is 85 percent less energy intensive than industry; a one percentage point rise of the service sector share in GDP will reduce electricity growth by 2 percentage points for the same output growth.
  - Transition to services increases labor income share in GDP due to higher labor intensity.

### III. Progress on Rebalancing (assessment)
- Overall assessment:
  - External rebalancing advanced well.
  - Internal rebalancing mixed—substantial supply-side progress, moderate demand-side progress, limited credit-side progress.
  - Mixed progress on income equality and environment: energy intensity down, labor’s share up, but income inequality and local air pollution remain very high.
- External:
  - Current account surplus fell from 10 percent of GDP in 2007 to around 2–3 percent recently.
  - Contribution of net exports to growth fluctuated around zero (from 2 percentage points of GDP annually in the pre-GFC peak).
- Internal (demand):
  - Since 2012, demand-side rebalancing from investment to consumption progressed moderately, with acceleration in 2015 and 2016:H1.
  - Consumption contributed about two thirds of overall GDP growth in recent period.
  - Investment ratio remained elevated at 43 percent of GDP while private consumption is 38 percent of GDP.
- Internal (credit):
  - Credit intensity of output doubled since pre-GFC and continued to rise.
  - Credit misallocation driven by financing weak firms, especially SOEs in overcapacity sectors (construction, steel).
  - Since 2016, early signs show lending shifting from overcapacity sectors to the “new economy,” but overall misallocation remains significant.
  - Indicators show rising leverage and higher percent of debt with ICR<1 and higher leverage in stressed sectors between 2010 and 2014.
- Internal (supply):
  - De-industrialization began at about US$ 9,000 (1990 international prices); industrial output share peaked in 2011.
  - Since 2012, nominal industrial share has steadily declined; real share decline muted so far—consistent with international experience where price effects lead nominal declines before real reallocation.
  - Industrial employment share peaked in 2012 and has been steadily declining.
- Efficiency:
  - Incremental capital output ratio (ICOR) rose post-crisis relative to pre-crisis average, indicating falling investment efficiency.
  - Widening gap in ROA: SOE ROA vs. private firm ROA shows deterioration in allocation efficiency.

### IV. Baseline prospects for rebalancing (projections to 2021)
- Assumptions underlying baseline:
  - Substantial progress in pro-consumption and services reform.
  - Lack of decisive SOE reforms and slow hardening of budget constraints.
  - Continued improvement in external balancing.
- Demographics and saving:
  - China will experience rapid ageing in the next 15 years, with the old-age dependence ratio forecast to double from its current level by 2030.
  - Cross-country evidence: one percentage point increase in the old-age dependence ratio lowers the saving rate by 0.4 to 1 percentage point.
  - Government health care spending rising from current 1.5 percent of GDP to 2.1 percent by 2020 (staff projection referenced).
  - Household saving projected to fall from 24 percent of GDP to about 21 percent of GDP by 2021.
  - Note: the recent lifting of the one-child policy may induce the savings rate to fall faster.
- National saving and investment:
  - National saving projected to fall to 41 percent of GDP by 2021.
  - Gross fixed asset formation projected to fall gradually to below 40 percent of GDP by 2021.
  - Current account surplus projected to remain low and decline further to less than 1 percent of GDP in the medium-term.
- Consumption and labor income:
  - Private consumption share in output projected to rise from around 38 percent in 2015 to 43 percent in 2021.
  - Services are more labor intensive: transition expected to boost labor income share in GDP.
- Credit dynamics:
  - Credit intensity expected to fall modestly but remain high.
  - Private credit to GDP ratio (excluding LGFVs) expected to reach 200 percent by 2021, up from about 160 percent in 2015 (assuming loan write-offs in the amount of 5 percent GDP in the next five years).
- Structural change toward services:
  - Nominal share of services in GDP projected to rise from current 50 percent to 55 percent by 2021.
  - By 2021, industrial employment share forecast to decline to 27 percent; service employment share forecast to rise to 51 percent.
- Environmental intensity:
  - With projected increase in service sector share, energy intensity of the economy expected to fall by 15 percent from the current level in 2021.

### V. Proactive scenario: faster internal rebalancing (policy-driven)
- Key policy actions and anticipated effects:
  - Faster NPL write-offs and corporate restructuring (exit of nonviable firms) and stronger hardening of budget constraints could stabilize private credit at some 20 percentage points of GDP lower than baseline in the medium-term.
  - Growth would be higher reflecting higher Total Factor Productivity growth.
  - Larger increase in government health spending would further reduce household saving and boost consumption.
  - Faster deregulation of the service sector could facilitate service sector productivity growth and accelerate resource reallocation from industry to services.
- Climate policy to meet Paris pledges:
  - Paris pledge: lower CO2 intensity of GDP by 60‒65 percent by 2030 from the 2005 level.
  - Structural and technological change alone expected to lower CO2 intensity by 45 percent by 2030.
  - Parry and others (2016) suggest a progressive carbon tax increasing from $5 per ton in 2017 to $70 per ton in 2030 could reach the Paris target by 2030.
  - Such a carbon tax would generate additional fiscal revenue of about 2 percent of GDP and facilitate supply-side rebalancing by shifting resources away from heavy polluting industries.

### VI. Risks: scenarios of concern
- Slowdown in investment without commensurate fall in saving:
  - If saving falls less than expected while investment adjusts, China could continue running a large current account surplus in the medium term, affecting global saving-investment imbalances and putting downward pressure on global interest rates.
- Faster decline in saving with continued high investment:
  - If saving declines faster (e.g., due to lifting of one-child policy or faster social safety net strengthening) while investment remains high, China could become a current account deficit country in the medium term and import foreign capital—raising exposure to balance of payments crises as investment returns fall.
- Premature deindustrialization:
  - Transitioning to services too fast may stall convergence: industrial productivity remains much lower than advanced economies and premature reallocation could produce sharper growth slowdown.
- Continued debt overhang:
  - Continued build-up of credit and financial risks could lead to a disruptive adjustment if budget constraints are not forcefully hardened.
  - Buffers (FX reserves and government debt) are large now but likely to diminish over time if reforms lag.

### VII. Policy implications and recommendations
- Exchange rate policy:
  - Move to an effectively floating exchange rate regime to prevent future misalignments; after significant appreciation since 2005, China’s exchange rate has broadly reached an equilibrium.
- Internal rebalancing—three priority fronts:
  - Higher health care spending:
    - Empirical evidence: a one RMB increase in government health care spending tends to reduce household saving by two RMB.
    - To achieve projected decline in saving rate, health care spending needs to increase from 1.7 percent of GDP in 2015 to 2.3 percent of GDP by 2021.
  - Deregulate the service sector:
    - Reform needed to improve service-sector productivity to sustain TFP growth during transition.
    - Duval and others (2015) estimate deregulation will increase productivity level by at least 20 percent in a five-year horizon.
    - Higher service productivity would lower relative price of services and promote consumption (Nabar and Yan, 2013).
  - Hardening budget constraints:
    - Budget constraints of SOEs need to be hardened; nonviable firms should exit rather than survive on rolled-over loans.
    - Currently, 40 percent of industrial assets are managed by SOEs, with asset returns some 7 percentage points lower than private counterparts.
    - Narrowing the gap in return on assets to its pre-crisis level of 2 percentage points could provide a significant boost to industrial growth.
- Income distribution and fiscal policy:
  - More redistributive fiscal policy needed: adopt a more progressive tax structure, increase transfers, and strengthen the social safety net.
  - Increase spending on education to prevent income inequality translating into unequal educational access.
  - Current social spending: China spends around 4 percent of GDP on social protection, which is less than half the ratio for upper-middle income economies—indicating ample room for higher social spending.
- Outcome with decisive implementation:
  - With decisive implementation of these policies, China can sustain strong growth and achieve transition to a greener, more inclusive, and more consumption- and service-driven economy.

*Source — IMF staff (excerpts from the PDF chapter provided).*

### Introduction ...........................................................................................................

### _wp16183 - Introduction

### I. Overview and Context
- China achieved GDP growth averaging about 10 percent and per capita income more than tripled over the past two decades.
- Pre-global financial crisis (GFC) growth relied heavily on exports; after WTO entry in 2001 China became the world’s manufacturing hub and the largest exporter by 2008.
- After the GFC, external demand collapsed and China transitioned from an export-based to investment-driven economy, with the investment ratio surging to 45 percent.
- Current account surplus peaked at 10 percent of GDP in 2007 and came down to around 2–3 percent in recent years.
- Credit intensity (new lending needed for an additional unit of output) has more than doubled in the post-GFC period.
- Income inequality rose substantially: Gini index increased from 0.3 in the 1980s to 0.53 in 2013.
- Environmental pollution, including carbon emissions and local air pollution (PM2.5), has worsened significantly.
- The paper identifies an inflection point: continuing the old growth model risks the middle-income trap or a financial crisis; excessive investment and resource misallocation lower efficiency, raise credit intensity, and expand debt to unsustainable levels.

### II. Framework of Rebalancing
- Definition: Rebalancing contains four key elements—external, internal, environmental, and distributional rebalancing—which are closely interlinked.
- External/internal:
  - External rebalancing: role of external versus domestic demand.
  - Internal rebalancing: shifting from investment to consumption (demand side); transitioning from industry to services (supply side); reducing credit intensity and improving allocative efficiency (input side).
- Environmental/income distribution:
  - Environmental rebalancing: reduce energy and carbon emission intensity of output and local air pollution (PM2.5).
  - Income distribution rebalancing: increase labor income share of GDP and reduce income inequality.
- Indicators:
  - External: current account balance as percent of GDP; alternatively contribution of net exports to GDP growth.
  - Internal (demand): nominal share of consumption and investment in GDP; relative contribution to GDP growth in real terms.
  - Internal (supply): nominal and real shares of industry versus services; industrial employment share.
  - Credit: credit intensity; credit to GDP ratio; difference in return on assets (ROA) between state-owned firms and private firms (asset allocation efficiency).
  - Environment: (1) energy intensity = electricity consumption per unit of output; (2) carbon emission intensity = carbon emission per unit of output; (3) PM2.5 for local air pollution.
  - Income distribution: Gini index; urban-rural income gap; labor income share of GDP.
- Interlinkages:
  - High national saving rate (household saving dominant) underlies external and internal imbalances. Corporate saving surged in early 2000s but normalized later.
  - When savings exceed domestic investment needs, excess savings flow out producing current account surplus; post-GFC, domestic absorption relied on high investment ratio.
  - Household saving rose from 5 percent in the mid-1970s to 40 percent in recent years, with trend breaks in late 1970s and late 1990s.
  - Drivers of high household saving: demographic change (one-child policy, falling fertility) and precautionary motives (healthcare reform in early 1990s increasing out-of-pocket risk; SOE reform in late 1990s reducing job security).
  - Successful rebalancing requires lower savings and a commensurate fall in investment; given habit persistence, savings decline likely gradual, implying gradual investment adjustment and focus on improving investment efficiency.
  - China’s capital stock per worker is less than 30 percent of U.S. level—capital deepening can continue but at a slower pace; post-crisis capital deepening outpaced income convergence.
  - Regression analysis: changes in the investment ratio alone can explain about 70 percent of variation of credit intensity historically.
  - Service sector is 85 percent less energy intensive than industry; a one percentage point rise of the service sector share in GDP will reduce electricity growth by 2 percentage points for the same output growth.
  - Transition to services increases labor income share in GDP due to higher labor intensity.

### III. Progress on Rebalancing
- Overall: External rebalancing advanced well; internal rebalancing mixed—substantial supply-side progress, moderate demand-side progress, limited credit-side progress; mixed progress on income equality and environment (energy intensity down, labor’s share up, but income inequality and local air pollution remain very high).
- External:
  - Current account surplus fell from 10 percent of GDP in 2007 to around 2–3 percent recently.
  - Contribution of net exports to growth fluctuated around zero (from 2 percentage points of GDP annually in the pre-GFC peak).
- Internal (demand):
  - Since 2012, demand-side rebalancing from investment to consumption progressed moderately, with acceleration in 2015 and 2016:H1.
  - Consumption contributed about two thirds of overall GDP growth in recent period.
  - Investment ratio remained elevated at 43 percent of GDP while private consumption is 38 percent of GDP.
- Internal (credit):
  - Credit intensity of output doubled since pre-GFC and continued to rise.
  - Credit misallocation driven by financing weak firms, especially SOEs in overcapacity sectors (construction, steel).
  - Since 2016, early signs show lending shifting from overcapacity sectors to the “new economy,” but overall misallocation remains significant.
  - Indicators show rising leverage and higher percent of debt with ICR<1 and higher leverage in stressed sectors between 2010 and 2014.
- Internal (supply):
  - De-industrialization began at about US$ 9,000 (1990 international prices); industrial output share peaked in 2011.
  - Since 2012, nominal industrial share has steadily declined; real share decline muted so far—consistent with international experience where price effects lead nominal declines before real reallocation.
  - Industrial employment share peaked in 2012 and has been steadily declining.
- Efficiency:
  - Incremental capital output ratio (ICOR) rose post-crisis relative to pre-crisis average, indicating falling investment efficiency.
  - Widening gap in ROA: SOE ROA vs. private firm ROA shows deterioration in allocation efficiency.

### IV. Risks and Implications (as presented)
- Continuing the investment/export-led model risks:
  - Dragging China into the middle-income trap.
  - Triggering a financial crisis via rising debt and falling efficiency.
- Rebalancing necessities:
  - Lower household saving rate and gradual reduction in investment ratio.
  - Improved investment efficiency and better credit allocation (reduce financing of weak SOEs and overcapacity sectors).
  - Further shift toward consumption and services to reduce energy and carbon intensity and raise labor income share.
  - Policy implementation is crucial: government commitments in the twelfth and thirteenth five-year plans aim for a greener, more inclusive, more consumer- and service-based, and less credit-driven economy; a comprehensive plan covering interlinked aspects is needed.

*Source: _wp16183 - Introduction (IMF staff text provided in source content).*

### 24.      Rebalancing progress on the environment and inequality has been mixed. Progress has

### 24.      Rebalancing progress on the environment and inequality has been mixed. Progress has

### Mixed progress: environment and inequality
- Carbon emission and electricity intensity of GDP have fallen, but PM 2.5 indexes in cities remain very high.
- Income inequality has risen sharply: Gini index increased from 0.3 in the 1980s to 0.53 in 2013—described as among the highest in the world.
- Labor income has recently gained a larger share of GDP, but the redistributive role of fiscal policy remains limited (small difference between gross and net Gini).

### Baseline prospects for rebalancing
- Assumptions:
  - Substantial progress in pro-consumption and services reform.
  - Lack of decisive SOE reforms and slow hardening of budget constraints.
  - Continued improvement in external balancing.
- Demographics and saving:
  - China will experience rapid ageing in the next 15 years, with the old-age dependence ratio forecast to double from its current level by 2030.
  - Cross-country evidence: one percentage point increase in the old-age dependence ratio lowers the saving rate by 0.4 to 1 percentage point.
  - Government health care spending rising from current 1.5 percent of GDP to 2.1 percent by 2020 (staff projection referenced).
  - Household saving projected to fall from 24 percent of GDP to about 21 percent of GDP by 2021.
  - Note: the recent lifting of the one-child policy may induce the savings rate to fall faster.
- National saving and investment:
  - National saving projected to fall to 41 percent of GDP by 2021.
  - Gross fixed asset formation projected to fall gradually to below 40 percent of GDP by 2021.
  - Current account surplus projected to remain low and decline further to less than 1 percent of GDP in the medium-term.
- Consumption and labor income:
  - Private consumption share in output projected to rise from around 38 percent in 2015 to 43 percent in 2021.
  - Services are more labor intensive: transition expected to boost labor income share in GDP.
- Credit dynamics:
  - Credit intensity expected to fall modestly but remain high.
  - Private credit to GDP ratio (excluding LGFVs) expected to reach 200 percent by 2021, up from about 160 percent in 2015 (assuming loan write-offs in the amount of 5 percent GDP in the next five years).
- Structural change toward services:
  - Nominal share of services in GDP projected to rise from current 50 percent to 55 percent by 2021.
  - By 2021, industrial employment share forecast to decline to 27 percent; service employment share forecast to rise to 51 percent.
- Environmental intensity:
  - With projected increase in service sector share, energy intensity of the economy expected to fall by 15 percent from the current level in 2021.

### Proactive scenario: faster internal rebalancing
- Key policy actions and effects:
  - Faster NPL write-offs and corporate restructuring (exit of nonviable firms) and stronger hardening of budget constraints could stabilize private credit at some 20 percentage points of GDP lower than baseline in the medium-term.
  - Growth would be higher reflecting higher Total Factor Productivity growth.
  - Larger increase in government health spending would further reduce household saving and boost consumption.
  - Faster deregulation of the service sector could facilitate service sector productivity growth and accelerate resource reallocation from industry to services.
- Climate policy to meet Paris pledges:
  - Paris pledge: lower CO2 intensity of GDP by 60‒65 percent by 2030 from the 2005 level.
  - Structural and technological change alone expected to lower CO2 intensity by 45 percent by 2030.
  - Parry and others (2016) suggest a progressive carbon tax increasing from $5 per ton in 2017 to $70 per ton in 2030 could reach the Paris target by 2030.
  - Such a carbon tax would generate additional fiscal revenue of about 2 percent of GDP and facilitate supply-side rebalancing by shifting resources away from heavy polluting industries.

### Risks: rebalancing going astray (scenarios of concern)
- Slowdown in investment without commensurate fall in saving:
  - If saving falls less than expected while investment adjusts, China could continue running a large current account surplus in the medium term, affecting global saving-investment imbalances and putting downward pressure on global interest rates.
- Faster decline in saving with continued high investment:
  - If saving declines faster (e.g., due to lifting of one-child policy or faster social safety net strengthening) while investment remains high, China could become a current account deficit country in the medium term and import foreign capital—raising exposure to balance of payments crises as investment returns fall.
- Premature deindustrialization:
  - Transitioning to services too fast may stall convergence: industrial productivity remains much lower than advanced economies and premature reallocation could produce sharper growth slowdown.
- Continued debt overhang:
  - Continued build-up of credit and financial risks could lead to a disruptive adjustment if budget constraints are not forcefully hardened.
  - Buffers (FX reserves and government debt) are large now but likely to diminish over time if reforms lag.

### Policy implications and recommendations
- Exchange rate policy:
  - Move to an effectively floating exchange rate regime to prevent future misalignments; after significant appreciation since 2005, China’s exchange rate has broadly reached an equilibrium.
- Internal rebalancing requires policy efforts on three fronts:
  - Higher health care spending:
    - Empirical evidence: a one RMB increase in government health care spending tends to reduce household saving by two RMB.
    - To achieve projected decline in saving rate, health care spending needs to increase from 1.7 percent of GDP in 2015 to 2.3 percent of GDP by 2021.
  - Deregulate the service sector:
    - Reform needed to improve service-sector productivity to sustain TFP growth during transition.
    - Duval and others (2015) estimate deregulation will increase productivity level by at least 20 percent in a five-year horizon.
    - Higher service productivity would lower relative price of services and promote consumption (Nabar and Yan, 2013).
  - Hardening budget constraints:
    - Budget constraints of SOEs need to be hardened; nonviable firms should exit rather than survive on rolled-over loans.
    - Currently, 40 percent of industrial assets are managed by SOEs, with asset returns some 7 percentage points lower than private counterparts.
    - Narrowing the gap in return on assets to its pre-crisis level of 2 percentage points could provide a significant boost to industrial growth.
- Income distribution and fiscal policy:
  - More redistributive fiscal policy needed: adopt a more progressive tax structure, increase transfers, and strengthen the social safety net.
  - Increase spending on education to prevent income inequality translating into unequal educational access.
  - Current social spending: China spends around 4 percent of GDP on social protection, which is less than half the ratio for upper-middle income economies—indicating ample room for higher social spending.
- Outcome with decisive implementation:
  - With decisive implementation of these policies, China can sustain strong growth and achieve transition to a greener, more inclusive, and more consumption- and service-driven economy.

*Italic: Source — IMF staff (excerpts from the PDF chapter provided).*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp16183.pdf_