## _wp1404

## Source details

**Canonical URL:** [_wp1404](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2014/_wp1404.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2014/_wp1404.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2014/_wp1404.pdf.json)

---

### I. Introduction: purpose and scope
- Objective: estimate size of off-budget fiscal activity in China since the early 2000s by constructing a time series of “augmented” fiscal data (i.e., adding off-budget fiscal activity to general government statistics).
- Caveats:
  - Augmented fiscal data are a complement, not a replacement, for standard general government fiscal data.
  - Calculations rely on numerous assumptions and judgments; estimates are subject to uncertainty.
  - The augmented measures are not comprehensive: excluded quasi-fiscal activity includes liabilities of regular SOEs and policy banks, and the series measures gross debt only (excludes government assets and SOE net worth/profits).
  - Improved official data (e.g., forthcoming audit of local government debt) would enhance accuracy.

### II. Rationale, approach, and bottom-line for “augmented” debt and deficits
- Rationale:
  - Infrastructure investment became local governments' main growth strategy and preferred countercyclical tool, financed largely off-budget via land sales and Local Government Financing Vehicle (LGFV) borrowings.
  - Key questions addressed: (1) actual level of fiscal vulnerabilities (government debt stock), and (2) actual size of fiscal stimulus supporting growth.
- Approach and definitions:
  - Augmented government boundary expands to include off-budget and LGFV activity; LGFVs are distinguished from other SOEs because they are largely set up, owned, and operated by local governments for fiscal-type activities with shared debt-service responsibilities.
  - Augmented debt: captures borrowing by LGFVs through market financing channels; excludes liabilities of regular SOEs and contingent liabilities (NPLs, policy bank loans, pension liabilities); measures gross debt only.
  - Augmented net borrowing: adds market financing of LGFVs (banks, bonds, trusts) to government net lending/borrowing; captures debt-creating flows and corresponds closely to change in augmented debt.
  - Augmented fiscal deficit: adds market financing of LGFVs and financing from selling land usage rights (net of costs such as resettlement and compensation) to the government deficit; land sales treated as a financing item akin to privatization.
- Bottom-line (IMF staff estimates):
  - Augmented government debt has risen to around 45 percent of GDP in 2012.
  - For 2012, augmented net borrowing was around 8 percent of GDP.
  - For 2012, the augmented fiscal deficit was on the order of 10 percent of GDP.

### III. Local fiscal structure and drivers of off-budget financing
- Institutional/fiscal context:
  - After the 1994 intergovernmental fiscal reform, the central government’s share of total fiscal revenue increased from less than 30 percent to around 50 percent in 2012.
  - No significant change in expenditure assignments occurred; higher-level governments can offload obligations to lower levels.
  - Local governments are responsible for much of infrastructure investment, service delivery, and social spending—together accounting for about 85 percent of total expenditure.
  - Local governments have limited own-revenue resources and little discretion over tax rates, making them dependent on central government transfers that mainly cover current spending, leaving less margin for infrastructure investment.
- Borrowing constraints and LGFV emergence:
  - Local governments are largely prohibited from on-budget borrowing; they resorted to off-budget mechanisms, commonly through LGFVs borrowing from banks, trust companies, or bond markets.
  - LGFVs often finance projects with limited revenue streams; many projects lack sufficient revenues to meet future debt servicing needs.
  - LGFVs are legally distinct from local SOEs and are established specifically for infrastructure spending; they often receive public credit enhancement and implicit guarantees.

### IV. Land financing model and its magnitudes
- Role of land sales:
  - Local governments’ share of land sales proceeds increased from 40 to 95 percent following the 1994 fiscal reform.
  - Land sales proceeds usually accrue to government-managed funds (GMFs) and became a major source of revenue used to finance infrastructure investment.
  - Direct taxes from land account for about 10 percent of total fiscal revenue; indirect taxes from construction and real estate can amount to over 50 percent of total fiscal revenues in some cities.
  - Land sales proceeds account for about two-thirds of GMF funds in value terms.
- Risks and regional variation:
  - Land-sales dependence creates risk via potential real estate over-supply and market correction; a correction could trigger falling land-sale proceeds → cut spending → contraction.
  - Many local governments injected land or property as capital to LGFVs or as collateral; a correction would reduce collateral values and hamper LGFVs’ borrowing and debt-service capacity.
  - Land sales are more important in relatively developed and fast emerging middle-income provinces; lower income regions had less income from land sales.
- Measurement and data treatment:
  - Gross land sales data: CEIC; example—2011 gross land sales explained Y 3.3 trillion (7 percent of GDP) of the Y 3.8 trillion in GMFs’ revenue.
  - Net land sales for 2010 and 2011 derived from Ministry of Finance data:
    - In 2010, net proceeds were 50 percent of the gross.
    - In 2011, net proceeds were 33 percent of the gross.
    - For other years, net proceeds are assumed to be 42 percent (the average of 2010 and 2011) to construct a time series.
  - Land sales data for 2005–09 appear incomplete; series extrapolated using Soufun growth up to 2009 and official data for 2010–12.
- Off-budget financing composition:
  - Bank loans accounted for around 80 percent of total local government debt by 2010, equivalent to Y 12.7 trillion:
    - Y 10.7 trillion borrowed by provincial, city, and county governments (NAO, 2011).
    - Y 0.8 trillion by township governments (Yu and Wei, 2012).
    - Y 1.2 trillion for roads and highways not captured by NAO (2011).
  - Trust loans to infrastructure projects amounted to Y 1.39 trillion by September 2012; this fell to just below Y 0.9 trillion by the first quarter of 2010 (no data before 2010; earlier flows estimated using bank loan growth rates).
  - LGFVs accounted for over one-quarter of corporate bond issuance in 2012 (WIND database).
  - Government cash deposits totaled around 4 percent of GDP at end-2012 and are included as financing.

### V. Augmented fiscal deficit, net borrowing, and dynamics
- Conceptual distinctions:
  - Augmented net borrowing:
    - Closer to GFSM2001 net lending/borrowing.
    - Adds LGFV borrowing from commercial banks, trust companies, and the corporate bond market to the general government deficit.
    - Constructed from financing (below-the-line) data; requires assumptions due to data gaps.
  - Augmented fiscal deficit:
    - Treats net proceeds from land sales as a financing item (i.e., like privatization proceeds), making the augmented fiscal deficit larger than the change in augmented net borrowing.
    - Better measures the impact of fiscal policy on demand because of the treatment of net proceeds from land sales.
  - Augmented net borrowing and fiscal deficit are calculated identically except for treatment of net proceeds from land sales.
- Historical magnitudes and patterns:
  - From 2007 to 2009, the augmented fiscal deficit increased by around 10 percent of GDP.
  - The augmented fiscal deficit was unwound rapidly and by 2011 had been reduced by around 8 percent of GDP from its peak.
  - In 2012, the augmented fiscal deficit increased by around 4 percent of GDP relative to 2011.
  - Augmented fiscal debt rises to around 45 percent of GDP in 2012.
  - Augmented fiscal deficit averaged about 10 percent of GDP since 2009.
  - Augmented net borrowing averaged about 7 percent of GDP since 2009.
  - In 2011, nominal GDP grew by around 18 percent, which by itself would have reduced the augmented fiscal debt ratio by around 6 percent of GDP; augmented net borrowing was around 3 percent of GDP in 2011, so the debt ratio declined in 2011.

### VI. Debt sustainability, stress tests, and scenario findings
- Baseline and medium-term assumptions:
  - Baseline: The favorable interest-growth differential (though less favorable than recent history) is sufficient to put the augmented fiscal debt on a downward trajectory by 2018.
  - In the baseline scenario, China’s augmented fiscal position is sustainable under the assumption of normalization of the augmented fiscal deficit over the medium term (the augmented deficit gradually declines by 3 percent of GDP through the medium term).
- Permanent shocks:
  - Real interest rate shock: "real interest rate is at baseline plus 4 percentage points" — sufficient to put augmented debt on an upward trajectory for much of the projection period, but "by 2018, the augmented debt would be on a slightly declining path and still at a manageable level."
  - Real GDP growth shock: "real GDP growth is at baseline minus 4 percentage points" — produces virtually identical debt dynamics to the interest shock.
- Primary deficit shock:
  - "The augmented primary deficit is around 2 percent of GDP higher than the baseline" — increases the debt stock, but "by 2018, still would not be on a firmly rising path."
- Combined shocks:
  - Combining a 4 percentage point worsening in the interest-growth differential with a larger primary deficit could raise augmented fiscal debt to "55 percent of GDP by 2018" in the extreme combined scenario described.
  - Logic: slower growth could prompt counter-cyclical fiscal response (a rise in the augmented fiscal deficit).
- Debt-shock (contingent liabilities) scenarios:
  - A sudden 10 percent of GDP increase in the debt ratio in 2013 would raise the debt ratio but "the debt dynamics would still be sustainable in the baseline."
  - It would take "a sudden rise in debt of around 50 percent of GDP, using the baseline assumptions, to put debt on a rising path in absence of additional fiscal adjustment."
- Land-sales shock:
  - If "net proceeds from land sales fall by 50 percent" and "borrowing rises by the same amount so that government spending does not fall," "debt rises moderately and stabilizes at around 50 percent of GDP by 2018."

### VII. Interest-growth differential, financing costs, and LGFV repayment capacity
- Interest-growth differential:
  - The interest-growth differential is a key factor keeping China's augmented debt ratio in check; a reversal is a principal risk to the debt outlook.
  - Policy implication: "In the long run, therefore, the fiscal position would have to gradually strengthen to offset the expected deterioration in the interest-growth differential."
- LGFV financing costs and maturities:
  - "LGFV borrowing costs are generally around double that of the central government."
  - "Government bond yields average around 3 percent in nominal terms."
  - "Bank lending rates to LGFVs have tended to be at slightly above the benchmark lending rate (by around 1–2 percentage points)."
  - Typical LGFV debt "would be paying around 6–8 percent interest rate, or more than double that of the central government."
  - Corporate bond rates for LGFVs "range from 3 percent to 6 percent depending on the maturities and credit ratings."
  - Trust loan rates "can range from as low as 6½ percent to above 10 percent."
- Repayment capacity and maturity risk:
  - Over 50 percent of subnational government debt was expected to mature by the end of 2012 (NAO (2011)).
  - Most infrastructure projects "were not expected to generate significant cash flow for 10 years or even longer."
  - Evidence suggests "a majority of the maturing debt was either rolled over or repaid by new borrowing."
  - NAO (2013) update: "gross debt issuance was equivalent to more than 50 percent of 2010 local government debt stock in the past two years, but the net debt stock increased only by 13 percent"; "about 40 percent of 2010 debt stock was either repaid or reclassified as private debt."
  - LGFV profitability and repayment capacity:
    - "Operating profits accounted for only about 4 percent of total LGFV profits."
    - "Township and city LGFVs were, in aggregate, making operating losses."
    - "Government subsidies account for about 90 percent of total profits."
    - The maturity mismatch and poor profitability imply "a majority of the maturing debt will most likely be serviced from debt roll-over or new borrowings."

### VIII. Coverage limits, contingent liabilities, and government assets
- Key exclusions from augmented measures:
  - Exclude liabilities of regular SOEs and state-owned financial entities; exclude contingent liabilities (banking sector NPLs, policy bank loans, pension liabilities).
  - "We exclude debt of the Ministry of Railways from the augmented debt stock. Their debt amounted to about 5 percent of GDP in 2012." The Ministry was corporatized in 2013 and its debt moved to a newly created SOE; SOE debt is excluded for consistency.
  - Social security is not consolidated into general government accounts and is excluded. National Pension Fund held "assets of 4.3 percent of GDP at end-2012."
- Contingent liabilities and asset magnitudes:
  - Potential contingent liabilities include nonfinancial SOE debt (excluding LGFVs), policy banks’ liabilities, fiscal costs of recapitalizing banks, liabilities of state-owned asset management companies.
  - Yang and others (CASS, 2012) estimated "total 2010 contingent liabilities at just above 100 percent of GDP." Components included:
    - debt of nonfinancial SOEs (excluding LGFVs): Y 35.6 trillion
    - policy banks’ outstanding financial debt: Y 5.2 trillion
    - old NPLs assumed by the asset management companies: Y 4.2 trillion
    - nonperforming loans (NPLs) in the current banking sector: Y 0.4 trillion
    - liabilities from social security funds: Y 3.5 trillion
  - Government assets estimates (CASS):
    - Y 60 trillion "(147 percent of GDP)" of equity of state-owned, for-profit nonfinancial enterprises.
    - Y 8 trillion "(20 percent of GDP)" of equity of state-owned for-profit financial enterprises.
    - Another close to Y 8 trillion of equity of state-owned nonprofit enterprises.
    - Land reserve held by land banks would add another Y 5 trillion or more.
    - An expanded estimate of around "350 percent of GDP" if including natural resources, international reserves, and social security fund assets.
  - Cautions: asset estimates are uncertain and assets can be illiquid, especially in downturns.
- Bank exposure to LGFV loans (selected figures from Ma (2012) / Table 3 as presented):
  - China Development Bank: "5,500 66.70.25"
  - Industrial Commercial Bank of China: "9318.71.09"
  - Agriculture Bank of China: "530.110.230.17"
  - Bank of China: "531.58.551.1"
  - China Construction Bank: "5809.440.19"
  - Bank of Communications: "308.312.670"
  - China Minsheng Banking Group: "172.115.06"
  - Shenzhen Development Bank: "6614.720.57"
  - China Everbright Bank: "94.911.50"
  - Shanghai Pudong Development Bank: "1058.84"
  - China Citic Bank: "170"
  - China Merchants Bank: "124.68.6"
  - Note: numbers presented in source (amounts by the second quarter of 2011); table headers indicate LGFV Loans (CYN billion), LGFV/Total Loans (In percent), and NPL (In percent).

### IX. Policy responses, reforms, and conclusions
- Observed and recommended policy actions:
  - Authorities reclassified LGFVs into categories based on revenue-generating capacity and dependence on government subsidies:
    - LGFVs that do not generate income can no longer serve as financing vehicles and local governments must finance spending on-budget; many such LGFVs cannot pay off existing debt and typically need to roll over debt and pledge more collateral.
    - LGFVs considered commercially operated remain eligible to borrow from the market.
  - The China Bank Regulatory Commission (CBRC) has scrutinized loans to LGFVs more closely.
  - Recommended and ongoing reforms include:
    - Putting in place a better framework to manage and monitor local government borrowing.
    - Fiscal reforms to intergovernmental relations.
    - Selective relaxation of borrowing constraints.
    - Quantity and quality control on investment.
    - Development of a fiscal risk management system.
- Policy-relevant takeaways:
  - Augmented fiscal data reveal that China's fiscal policy was considerably more counter-cyclical than suggested by headline general government data; the 2009–10 increase in the augmented fiscal deficit provided a considerable lift to activity and was unwound quickly.
  - The favorable interest-growth differential has been central to keeping augmented debt dynamics sustainable; deterioration of that differential is the principal medium- to long-term fiscal risk.
  - China still has fiscal buffers and government assets that could, in principle, be mobilized, but contingent liabilities, LGFV financing costs, maturity mismatches, weak LGFV profitability, and potential illiquidity of assets are key vulnerabilities that warrant monitoring and, over time, gradual fiscal strengthening.
- Outlook:
  - Debt sustainability analysis and stress tests indicate the augmented fiscal debt is still at a manageable level, assuming modest consolidation in the augmented fiscal deficit over the medium term.
  - The higher augmented debt and deficits imply China has somewhat less fiscal space than suggested by headline data and is more vulnerable to a macroeconomic shock.
  - There is still room to use fiscal policy to support demand while following a path of gradual adjustment of the augmented fiscal deficit.

*Prepared from the supplied content of the referenced IMF chapter/section.*

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

### _wp1404 - References

### I. Introduction: purpose and scope
- Objective: estimate size of off-budget fiscal activity in China since the early 2000s by constructing a time series of “augmented” fiscal data (i.e., adding off-budget fiscal activity to general government statistics).
- Caveats:
  - Augmented fiscal data are a complement, not a replacement, for standard general government fiscal data.
  - Calculations rely on numerous assumptions and judgments; estimates are subject to uncertainty.
  - The augmented measures are not comprehensive: excluded quasi-fiscal activity includes liabilities of regular SOEs and policy banks, and the series measures gross debt only (excludes government assets and SOE net worth/profits).
  - Improved official data (e.g., forthcoming audit of local government debt) would enhance accuracy.

### II. Box 1 — Rationale, approach, and bottom-line for “augmented” debt and deficits
- Rationale:
  - Infrastructure investment became local governments' main growth strategy and preferred countercyclical tool, financed largely off-budget via land sales and Local Government Financing Vehicle (LGFV) borrowings.
  - Key questions: (1) actual level of fiscal vulnerabilities (government debt stock), and (2) actual size of fiscal stimulus supporting growth.
- Approach: develop an “augmented” concept expanding the government perimeter to include off-budget and LGFV activity; LGFVs are distinguished from other SOEs because they are largely set up, owned, and operated by local governments for fiscal-type activities with shared debt-service responsibilities.
- Definition of augmented measures:
  - Augmented debt: captures borrowing by LGFVs through market financing channels; excludes liabilities of regular SOEs and contingent liabilities (NPLs, policy bank loans, pension liabilities); measures gross debt only.
  - Augmented net borrowing: adds market financing of LGFVs (banks, bonds, trusts) to government net lending/borrowing; captures debt-creating flows and corresponds closely to change in augmented debt.
  - Augmented fiscal deficit: adds market financing of LGFVs and financing from selling land usage rights (net of costs such as resettlement and compensation) to the government deficit; land sales treated as a financing item akin to privatization and do not contribute to debt accumulation.
- Bottom-line estimates (IMF staff):
  - Augmented government debt has risen to around 45 percent of GDP in 2012.
  - For 2012, augmented net borrowing was around 8 percent of GDP.
  - For 2012, the augmented fiscal deficit was on the order of 10 percent of GDP.

### III. Local fiscal structure and drivers of off-budget financing
- Institutional/fiscal context:
  - After the 1994 intergovernmental fiscal reform, the central government’s share of total fiscal revenue increased from less than 30 percent to around 50 percent in 2012.
  - No significant change in expenditure assignments occurred; higher-level governments can offload obligations to lower levels.
  - Local governments are responsible for much of infrastructure investment, service delivery, and social spending—together accounting for about 85 percent of total expenditure.
  - Local governments have limited own-revenue resources and little discretion over tax rates, making them dependent on central government transfers that mainly cover current spending, leaving less margin for infrastructure investment.
- Borrowing constraints and LGFV emergence:
  - Local governments are largely prohibited from on-budget borrowing; they resorted to off-budget mechanisms.
  - Typical mechanism: local government creates a company (LGFV) that borrows from banks, trust companies, or bond markets to finance infrastructure.
  - LGFVs often finance projects with limited revenue streams; many projects lack sufficient revenues to meet future debt servicing needs.
  - LGFVs are distinguishable from local SOEs in Chinese statistics and are established specifically for infrastructure spending.

### IV. Role of the 2009 stimulus and implications
- During the 2009 stimulus, investment (including infrastructure) accelerated significantly to support growth and advance urbanization.
- Urbanization increased infrastructure needs; with local governments unable to borrow on-budget, LGFVs were a natural vehicle for financing.
- Off-budget financing (LGFVs and land sales) played a central role in the stimulus but is not fully visible in headline fiscal statistics.

### V. Conceptual boundaries and consolidation choices
- Augmented government boundary: includes central government, local government, government-managed funds, and LGFVs; excludes SOEs and state-owned financial entities.
- Rationale for exclusion of SOEs: LGFVs are largely created and operated by local governments for fiscal purposes, with government sharing of debt service and subsidies for losses; this distinguishes them from broader SOEs.
- Consolidation considerations:
  - Consolidated augmented fiscal debt can be constructed as the sum of central, local, and LGFV debt because cross-holdings are minimal.
  - Wider concepts (nonfinancial public sector, financial public sector) are more difficult to consolidate due to intra-public-sector holdings that require netting.

### VI. Methodological notes and data limitations
- Augmented fiscal measures rely on IMF staff estimates where LGFV data are incomplete.
- Augmented fiscal deficit includes land sales net of resettlement and compensation costs (treated as financing akin to privatization).
- Some items remain excluded from augmented measures: contingent liabilities (banking sector NPLs, policy bank loans), pension liabilities, liabilities of regular SOEs, and state-owned financial entities.
- The augmented approach measures gross debt and does not subtract government assets or account for SOE net worth and profits.

*Prepared from the supplied content of the referenced IMF chapter/section.*

### Box 2. Examples of LGFVs

### _wp1404 - Box 2. Examples of LGFVs

### Fushun Development Investment Corporation (FSDIC)
- Established: June 8, 2002.
- Initial capital injected by Fushun city government: Y 150 million.
- Additional paid-in capital transferred four years after establishment: a land usage right (date not specified).
- 2009 SASAC-approved transfers to FSDIC as additional paid-in capital:
  - Three solely state-owned companies transferred all equity to FSDIC and became subsidiaries:
    - Fushun City Water Supply Company;
    - Economic Development Zone Investment and Development Co., Ltd.;
    - Fushun City, State-owned Guarantee Center.
  - Amount transferred as additional paid-in capital in 2009: Y 720 million.
- Responsibilities:
  - Funding, operation, and construction of the majority of Fushun city-wide infrastructure investment.
  - Example projects: Fushun City sewage collection pipe network project; City-west flood protection project; the Core District road network construction project; infrastructure development projects in the Economic Development Zone; social housing projects.
- Income composition:
  - Mainly subsidies from the city government and revenue generated through its subsidiaries (largely from Fushun City Water Supply Company).
  - Land sales have become an important source of funds in more recent years due to rising land prices from construction and real estate growth.
- Financing and assets:
  - Issues corporate bonds using land usage rights as collateral to support its credit ratings and reduce borrowing costs.
  - Relatively more liquid assets account for a majority of total assets, including land, public buildings, roads, and other infrastructures.
  - Much of the land is pledged as collateral; other assets (buildings and roads) may have limited liquidity and weak potential to generate profit.

### Haicheng Urban Development Investment Corporation (HUDIC)
- Established: May 2001.
- Initial capital injected by Haicheng city government: Y 105 million.
- 2009 transfers from government to company: land usage rights, buildings, and other fixed assets amounting to Y 2.8 billion.
- Subsidiaries (all state-owned):
  - HaiCheng Heating Company;
  - Urban Infrastructure Company;
  - a water company.
- Responsibilities:
  - Investing in, constructing, and managing various urban infrastructure projects.
  - Land development (typically cleaning and then selling land received from the local government).
  - Led projects such as road and bridge construction.
- Income composition:
  - Revenue from subsidiaries;
  - Agency fees from infrastructure projects initiated by the local government;
  - Government subsidies.
- Financial position:
  - Operating income flow has been weak, making HUDIC highly reliant on government subsidies.

### Context: Augmented Fiscal Debt and Deficit (selected excerpts)
- IMF staff estimate of augmented fiscal debt: around 45 percent of GDP in 2012.
- Range of other estimates for augmented fiscal debt: from just below 30 percent to above 70 percent of GDP.
- Estimates of local government debt: most estimates range between 25 and 35 percent of GDP.
- Example differing coverage: Yu and Wei (DRC) include townships in addition to provinces, cities, and counties; their estimate of local government debt is Y 2 trillion (5 percent of 2012 GDP), higher than the NAO (2011) estimate which excluded townships.
- Market and analyst differences:
  - Bank of America’s 2009 estimate includes LGFVs’ noninfrastructure loans and is higher than Citi’s estimate which only captures infrastructure loans.
  - Many estimates also include contingent liabilities such as NPLs held by asset management companies, policy bank loans, and pension fund liabilities.
- Implication: Augmented fiscal debt being higher than general government debt implies augmented deficits must also be higher.

### Concepts and Measurement Notes
- Two concepts introduced:
  - Augmented net borrowing:
    - Closer to GFSM2001 definition of net lending/borrowing.
    - Corresponds to financing needs of the augmented government and closely matches changes in augmented fiscal debt.
    - Adds off-budget quasi-fiscal activity, mainly infrastructure, to the general government data.
    - Constructed based on financing (below-the-line) data and requires numerous assumptions due to data gaps.
    - Calculated by adding LGFV borrowing from commercial banks, trust companies, and the corporate bond market to the general government deficit.
  - Augmented fiscal deficit:
    - Treats the net proceeds from land sales as a financing item (i.e., like privatization proceeds), making the augmented fiscal deficit larger than the change in augmented net lending.
    - Better measures the impact of fiscal policy on demand because of the treatment of net proceeds from land sales.
- Note on financing model for infrastructure based on land: Box 3 (referenced) explains how this financing model is used in China (content not reproduced here).
- Calculation specifics:
  - Augmented net borrowing and fiscal deficit concepts are similar and calculated identically except for treatment of net proceeds from land sales.

*Source: _wp1404 - Box 2. Examples of LGFVs (excerpt).*

### Box 3. Land Financing Model

### Box 3. Land Financing Model

### Incentives and role of land sales
- Local governments’ share of land sales proceeds increased from 40 to 95 percent following the 1994 fiscal reform (Peterson, 2007).
- Land sales became a major source of revenue for local governments as urbanization advanced, with the proceeds usually accruing to government-managed funds (GMFs).
- Land sales proceeds were used to finance infrastructure investment, supporting urbanization and catalyzing other investment; this created a land sales → investment → growth incentive for local officials.
- Direct taxes from land (urban usage, agriculture occupancy, and deed) account for about 10 percent of total fiscal revenue.
- Indirect taxes such as sales and corporate income taxes generated from construction and real estate companies amount to over 50 percent of total fiscal revenues in some cities.
- Land sales proceeds account for about two-thirds of GMF funds in value terms.
- With only about 50 percent of households living in urban areas, China’s urbanization process is likely to continue and thus support land-related activity going forward.

### Risks and regional variation
- Lands sales create risk via potential real estate over-supply and market correction; a correction could trigger a negative feedback loop: falling land-sale proceeds → cut spending while construction activity and fiscal revenue decline → further contraction.
- Many local governments injected land or property as capital to LGFVs or as collateral; a correction would reduce collateral values and hamper LGFVs’ ability to borrow or to sell land to service maturing debt.
- The relative importance of land sales varies by region: more important in relatively developed and fast emerging middle-income provinces; lower income regions had less income from land sales.
- Regional differences may have contributed to widening regional inequality.

### Market financing and LGFVs
- On-budget borrowing by local governments is limited and tightly regulated. Aggregate amounts were Y 200 billion a year in 2009–11 and Y 250 billion in 2012; these are included in general government debt.
- As of end-2010, NAO (2011) reported that state-council approved direct borrowing was around 7 percent of GDP.
- In response to the global financial crisis and stimulus needs, local governments used LGFVs and government-related entities to borrow from policy banks, commercial banks, trust companies, and the corporate bond market.
- LGFVs are legally distinct entities, often receiving public credit enhancement; some LGFV debts were implicitly guaranteed by local governments.
- Collateral for LGFV borrowing often included land or other assets, owned by the LGFV, pledged by the local government, or provided by another entity.
- The analysis includes borrowing by LGFVs and excludes SOE debt; the simplifying assumption is that all LGFV debt warrants inclusion in augmented net borrowing, while SOE debt does not.

### Off-budget financing—composition and magnitudes
- Bank loans accounted for around 80 percent of total local government debt by 2010, equivalent to Y 12.7 trillion:
  - Y 10.7 trillion borrowed by provincial, city, and county governments (NAO, 2011).
  - Y 0.8 trillion by township governments (Yu and Wei, 2012).
  - Y 1.2 trillion for roads and highways not captured by NAO (2011).
- Corporate bonds: LGFVs accounted for over one-quarter of issuance in 2012 (WIND database); corporate bond market has grown significantly.
- Trust loans to infrastructure projects amounted to Y 1.39 trillion by September 2012 (China Trustee Association); this fell to just below Y 0.9 trillion by the first quarter of 2010. No data before 2010; earlier flows are estimated using bank loan growth rates.
- Changes in government cash deposits are included as financing; deposits totaled around 4 percent of GDP at end-2012.
- Data gaps prevent decomposition of net bank loan issuance into gross issuance, repayment through land sales proceeds, and debt rollover.
- NAO (2013) survey suggests debt rose by 13 percent since 2010; used to estimate debt stock at end-2012.

### Augmented fiscal deficit: methodology and land sales treatment
- Augmented fiscal deficit calculation: augmented net borrowing + net proceeds from land sales (portion actually used to finance infrastructure and other spending).
- Net proceeds from asset (land) sales are treated as financing, not revenue.
- Gross land sales data: CEIC; example—2011 gross land sales explained Y 3.3 trillion (7 percent of GDP) of the Y 3.8 trillion in GMFs’ revenue.
- Net land sales for 2010 and 2011 derived from Ministry of Finance data (gross proceeds minus acquisition costs, compensation to farmers, and land development costs):
  - In 2010, net proceeds were 50 percent of the gross.
  - In 2011, net proceeds were 33 percent of the gross.
  - For other years, net proceeds are assumed to be 42 percent (the average of 2010 and 2011) to construct a time series.
- Land sales data for 2005–09 appear incomplete; series extrapolated using Soufun growth up to 2009 and official data for 2010–12.

### Above-the-line robustness check
- Above-the-line estimate constructs augmented fiscal revenue and augmented fiscal expenditure.
- Augmented fiscal revenue ≈ general government revenue, plus revenue (excluding land sales) from central and local GMFs.
  - Reported GMF revenue in 2011 was Y 4.1 trillion; only Y 0.8 trillion after subtracting land sales.
  - Of this, Y 0.3 trillion from the central government and Y 0.5 million from local governments (staff assumption based on subtraction reported).
- Augmented fiscal expenditure is adjusted for GMF and LGFV spending; LGFV spending not directly available, so infrastructure investment is used as proxy.
  - Infrastructure investment is calculated from fixed asset investment components (primary industry; electricity and heating; gas; water; railway transport; highway transport; waterway transport; air transport; water conservancy and environment management). Breakdown only available since 2004; earlier years assumed to grow at same rate as total fixed asset investment.
  - Based on 2011 data, assumptions for share of infrastructure spending: 21 percent of local government budget spending; 14 percent of central government budget spending; 36 percent of local GMF spending; 47 percent of central GMF spending. Residual infrastructure investment is attributed to LGFVs.
- Above-the-line estimates broadly consistent with financing approach, similar prior to 2008; both show sharp pickup in 2009. Above-the-line suggests a larger augmented fiscal deficit in recent years because it includes all infrastructure spending (including execution by SOEs or private companies) and excludes LGFV own revenue due to lack of data.

### Interpreting the estimates and recent magnitudes
- The augmented fiscal deficit is consistently larger than augmented net borrowing; difference arises from treating net proceeds from land sales (around 2 percent of GDP in recent years) as financing rather than revenue.
- Using augmented data:
  - From 2007 to 2009, the augmented fiscal deficit increased by around 10 percent of GDP.
  - The augmented fiscal deficit was unwound rapidly and by 2011 had been reduced by around 8 percent of GDP from its peak.
  - In 2012, the augmented fiscal deficit increased by around 4 percent of GDP relative to 2011.
- Augmented fiscal debt rises to around 45 percent of GDP in 2012.
  - Augmented fiscal deficit averaged about 10 percent of GDP since 2009.
  - Augmented net borrowing averaged about 7 percent of GDP since 2009.
  - In 2011, nominal GDP grew by around 18 percent, which by itself would have reduced the augmented fiscal debt ratio by around 6 percent of GDP; augmented net borrowing was around 3 percent of GDP in 2011, so the debt ratio declined in 2011.
- Cross-country perspective: China’s augmented fiscal debt to GDP ratio is comparable to other emerging markets and well below most advanced economies, but augmented fiscal deficit in 2012 is larger than general government deficit in most other emerging and advanced economies.
- If augmented fiscal deficit returns to around 4 percent of GDP (roughly the average of 2000–07), China would be about the middle of the sample of emerging and advanced economies.
- Estimates are subject to considerable uncertainty due to data gaps (LGFV revenues/debt composition; incomplete historical land sale data; LGFV spending breakdowns).

### Debt sustainability (summary of baseline)
- In the baseline scenario, China’s augmented fiscal position is sustainable under the assumption of normalization of the augmented fiscal deficit over the medium term (the augmented deficit gradually declines by 3 percent of GDP through the medium term).

*Source: Box 3. Land Financing Model, _wp1404 - Box 3. Land Financing Model*

### 2018. Combined with the still favorable interest-growth differential (though less favorable than

### _wp1404 - 2018. Combined with the still favorable interest-growth differential (though less favorable than

### Debt sustainability and stress-test findings
- Baseline: The favorable interest-growth differential (though less favorable than recent history) is sufficient to put the augmented fiscal debt on a downward trajectory by 2018.
- A "3 percent of GDP adjustment" is described as ambitious but less than the rapid unwinding after the 2009–10 stimulus; even with this adjustment, the augmented fiscal deficit would remain larger than in 2011.
- Permanent shocks considered:
  - Real interest rate shock: "real interest rate is at baseline plus 4 percentage points" — this shock is sufficient to put augmented debt on an upward trajectory for much of the projection period, but "by 2018, the augmented debt would be on a slightly declining path and still at a manageable level."
  - Real GDP growth shock: "real GDP growth is at baseline minus 4 percentage points" — produces virtually identical debt dynamics to the interest shock, underscoring that the interest-growth differential drives debt dynamics.
- Primary deficit shock:
  - "The augmented primary deficit is around 2 percent of GDP higher than the baseline" (noted alternatively as "primary balance is at baseline minus one-half standard deviation") — increases the debt stock, but "by 2018, still would not be on a firmly rising path."
- Combined shocks:
  - Combining a 4 percentage point worsening in the interest-growth differential with a larger primary deficit could raise augmented fiscal debt to "55 percent of GDP by 2018" in the extreme combined scenario described.
  - The analysis notes logic for such combination: slower growth could prompt counter-cyclical fiscal response (a rise in the augmented fiscal deficit).
- Debt-shock (contingent liabilities) scenarios:
  - A sudden 10 percent of GDP increase in the debt ratio in 2013 would raise the debt ratio but "the debt dynamics would still be sustainable in the baseline."
  - It would take "a sudden rise in debt of around 50 percent of GDP, using the baseline assumptions, to put debt on a rising path in absence of additional fiscal adjustment."
- Land-sales shock:
  - If "net proceeds from land sales fall by 50 percent" and "borrowing rises by the same amount so that government spending does not fall," "debt rises moderately and stabilizes at around 50 percent of GDP by 2018."

### Interest-rate growth differential and financing costs
- The interest-growth differential is a key factor keeping China's augmented debt ratio in check; a reversal is a principal risk to the debt outlook.
- LGFV financing is increasingly from more expensive, market-based sources, raising effective interest costs relative to central government debt.
- Relative borrowing costs and rates described:
  - "LGFV borrowing costs are generally around double that of the central government."
  - "Government bond yields average around 3 percent in nominal terms."
  - "Bank lending rates to LGFVs have tended to be at slightly above the benchmark lending rate (by around 1–2 percentage points)."
  - Typical LGFV debt "would be paying around 6–8 percent interest rate, or more than double that of the central government."
  - Corporate bond rates for LGFVs "range from 3 percent to 6 percent depending on the maturities and credit ratings."
  - Trust loan rates "can range from as low as 6½ percent to above 10 percent."
- Financial sector reform could raise average borrowing costs (IMF, 2013a), and the interest-growth differential is "likely to become considerably less favorable over time" as (real) interest rates rise and trend growth slows.
- Policy implication: "In the long run, therefore, the fiscal position would have to gradually strengthen to offset the expected deterioration in the interest-growth differential."

### Gross financing needs and LGFV debt servicing
- Over 50 percent of subnational government debt was expected to mature by the end of 2012 (NAO (2011)).
- Most infrastructure projects "were not expected to generate significant cash flow for 10 years or even longer."
- Sources for servicing maturing debt: land sales, operating profits (such as highway fees), local fiscal revenues, inter-governmental transfers; evidence suggests "a majority of the maturing debt was either rolled over or repaid by new borrowing."
- NAO (2013) update: "gross debt issuance was equivalent to more than 50 percent of 2010 local government debt stock in the past two years, but the net debt stock increased only by 13 percent"; "about 40 percent of 2010 debt stock was either repaid or reclassified as private debt."
- LGFV profitability and repayment capacity (from Ma (2012), Table 2 summary):
  - "Operating profits accounted for only about 4 percent of total LGFV profits."
  - "Township and city LGFVs were, in aggregate, making operating losses."
  - "Government subsidies account for about 90 percent of total profits."
  - The maturity mismatch and poor profitability imply "a majority of the maturing debt will most likely be serviced from debt roll-over or new borrowings."
- Maturity structure (NAO (2011) / IMF staff estimates): substantial shares of corporate bonds issued to LGFVs with varying tenors; many maturing within 1–5 years (chart not reproduced here).

### Coverage and other considerations (contingent liabilities and assets)
- Exclusions from augmented data:
  - "We exclude debt of the Ministry of Railways from the augmented debt stock. Their debt amounted to about 5 percent of GDP in 2012." The Ministry was corporatized in 2013 and its debt moved to a newly created SOE; SOE debt is excluded from augmented series (so Ministry of Railway debt excluded for entire time series for consistency).
  - Social security is not consolidated into general government accounts and is excluded. Inclusion would increase augmented revenue and spending and somewhat reduce the deficit.
  - National Pension Fund held "assets of 4.3 percent of GDP at end-2012."
  - Legacy costs in the pension system were "80–132percent of GDP in 2008" and the actuarial deficit was "around 95 percent of 2001 GDP" (estimates, albeit dated).
- Contingent liabilities:
  - Contingent liabilities are excluded from augmented debt but are an important source of risk.
  - Potential contingent liabilities include: nonfinancial SOE debt (excluding LGFVs), policy banks' liabilities, fiscal costs of recapitalizing banks (losses related to NPLs), liabilities of state-owned asset management companies.
  - Yang and others (CASS, 2012) estimated "total 2010 contingent liabilities at just above 100 percent of GDP." Their components include:
    - debt of nonfinancial SOEs (excluding LGFVs): Y 35.6 trillion
    - policy banks’ outstanding financial debt: Y 5.2 trillion
    - old NPLs assumed by the asset management companies: Y 4.2 trillion
    - nonperforming loans (NPLs) in the current banking sector: Y 0.4 trillion
    - liabilities from social security funds: Y 3.5 trillion
  - Note: In calculating contingent liabilities of the augmented government, potential costs related to NPLs from LGFVs "should be excluded" to avoid double counting because LGFV borrowing is already included in augmented fiscal debt.
- Government assets (Yang and others, CASS (2012) estimates):
  - Government assets amount to "nearly 200 percent of GDP" in one estimate.
  - Specific holdings: Y 60 trillion "(147 percent of GDP)" worth of equity of state-owned, for-profit nonfinancial enterprises; Y 8 trillion "(20 percent of GDP)" of equity of state-owned for-profit financial enterprises; another close to Y 8 trillion of equity of state-owned nonprofit enterprises.
  - "Land reserve held by land banks would add another Y 5 trillion or more."
  - An expanded estimate of around "350 percent of GDP" if including value of natural resources, international reserves held by the central bank, and social security fund assets.
  - Cautions: estimates are subject to uncertainty and assets can be less liquid than anticipated, especially when being sold in an economic downturn.

### Bank exposure to LGFV loans (selected figures from Ma (2012) / Table 3)
- China Development Bank: "5,500 66.70.25"
- Industrial Commercial Bank of China: "9318.71.09"
- Agriculture Bank of China: "530.110.230.17"
- Bank of China: "531.58.551.1"
- China Construction Bank: "5809.440.19"
- Bank of Communications: "308.312.670"
- China Minsheng Banking Group: "172.115.06"
- Shenzhen Development Bank: "6614.720.57"
- China Everbright Bank: "94.911.50"
- Shanghai Pudong Development Bank: "1058.84"
- China Citic Bank: "170"
- China Merchants Bank: "124.68.6"
- Note: numbers as presented in source (amounts by the second quarter of 2011); table headers in source indicate LGFV Loans (CYN billion), LGFV/Total Loans (In percent), and NPL (In percent).

### Conclusion (policy-relevant takeaways)
- Augmented fiscal data reveal that China's fiscal policy was considerably more counter-cyclical than suggested by headline general government data; the 2009–10 increase in the augmented fiscal deficit provided a considerable lift to activity and was unwound quickly as the economy recovered.
- The favorable interest-growth differential has been central to keeping augmented debt dynamics sustainable; a deterioration of that differential is the principal medium- to long-term fiscal risk.
- China still has fiscal buffers and government assets that could, in principle, be mobilized, but contingent liabilities, LGFV financing costs, maturity mismatches, weak LGFV profitability, and potential illiquidity of assets are key vulnerabilities that warrant monitoring and, over time, gradual fiscal strengthening.

*Source: IMF staff estimates and compiled material from the provided content unit.*

### 2012. Meanwhile, the augmented fiscal deficit is also considerably larger than the headline

### _wp1404 - 2012. Meanwhile, the augmented fiscal deficit is also considerably larger than the headline

### Augmented fiscal deficit and debt
- The augmented fiscal deficit is considerably larger than the headline government deficit.
- Augmented fiscal debt had risen to around 45 percent of GDP in 2012.
- This augmented fiscal debt is around double general government debt but broadly in line with other estimates using an expanded definition of government.
- The augmented fiscal debt estimate covers only part of the government’s balance sheet: it excludes the government’s considerable holdings of financial and nonfinancial assets, contingent liabilities, and liabilities of SOEs and the public financial sector.

### Debt sustainability and vulnerabilities
- Debt sustainability analysis and stress tests indicate the augmented fiscal debt is still at a manageable level, assuming modest consolidation in the augmented fiscal deficit over the medium term.
- The higher augmented debt and deficits imply China has somewhat less fiscal space than suggested by headline data and is more vulnerable to a macroeconomic shock.
- There is still room to use fiscal policy to support demand while following a path of gradual adjustment of the augmented fiscal deficit.

### Underlying challenges in local government finances
- Mismatch between local government expenditure responsibilities and revenue sources.
- Tight restrictions on direct borrowing have led local governments to seek off-budget financing, most notably via Local Government Financing Vehicles (LGFVs), to finance infrastructure and support ongoing urbanization.
- Risks associated with local government practices include:
  - Reliance on land sales, which distorts the real estate market and could exacerbate a cyclical downturn.
  - Rollover risk from the maturity structure of existing borrowing.
  - Potential for the interest-growth differential to become less favorable, partly because local governments face higher interest costs than the central government.

### Policy responses and institutional reforms
- Authorities reclassified LGFVs into categories based on revenue-generating capacity and dependence on government subsidies:
  - LGFVs that do not generate income can no longer serve as financing vehicles and local governments must finance spending on-budget; many such LGFVs cannot pay off existing debt and typically need to roll over debt and pledge more collateral.
  - LGFVs considered commercially operated remain eligible to borrow from the market.
- The China Bank Regulatory Commission (CBRC) has scrutinized loans to LGFVs more closely.
- Recommended and ongoing reforms include:
  - Putting in place a better framework to manage and monitor local government borrowing.
  - Fiscal reforms to intergovernmental relations.
  - Selective relaxation of borrowing constraints.
  - Quantity and quality control on investment.
  - Development of a fiscal risk management system.
- These reforms aim to prevent further buildup of risks while ensuring adequate financing for priority social and infrastructure spending.

### Implications and outlook
- On balance, augmented fiscal data suggest China’s fiscal position is weaker than headline data indicate but remains within sustainability thresholds.
- Continued reform and stronger management of local government finances are needed to address fundamental institutional weaknesses and to contain fiscal risks.

*Source: IMF staff estimates and analysis as presented in the provided content.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2014/_wp1404.pdf_