## Appendix 1. The Public Sector in China

## Source details

**Canonical URL:** [Appendix 1. The Public Sector in China](https://www.imf.org/-/media/files/publications/wp/2023/english/wpiea2023154-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2023/english/wpiea2023154-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2023/english/wpiea2023154-print-pdf.pdf.json)

---

### Summary findings
- New estimates of China’s government financial balance sheet covering 1997–2019 provide a more comprehensive view of fiscal vulnerabilities than standard debt and deficit metrics.
- China has the highest financial assets in the world, amounting to US$12.5 trillion in 2019.
- Equity holdings of financial and nonfinancial public corporations account for the largest share of those assets and are, at 68 percent of GDP, about three times higher than the average country in the sample.
- Net financial worth:
  - Estimated at 7¼ percent of GDP (general statement).
  - Reported as 7 percent of GDP in 2019 (composition table context).
  - Placed China in the top 15th percentile of the distribution of a 50-country sample.
- Despite large financial assets, net financial worth has been on a steady decline since its peak in 2007, mainly driven by rising local government debt, weak SOE performance, and falling infrastructure returns.
- The bulk of the decline in net financial worth is explained by subnational governments, with likely significant variation across provinces.

### Major drivers and episodes shaping the balance sheet
- Restructuring of public corporations (1997–2007):
  - Produced a strong rebound in asset valuation and a sharp turnaround in net financial worth.
  - China’s restructuring performance was unprecedented: net financial worth was 20 percentage points of GDP above normal levels five years after start, versus the average country where it remained 10 percentage points of GDP below normal levels.
- Investment-led stimulus and proliferation of off-budget financing (2008–2014):
  - Stimulus beginning at the onset of the global financial crisis associated with gradual decline in net financial worth despite rising financial assets, largely reflecting off-budget financing vehicles used by local governments.
- Pre-pandemic deleveraging and SOE expansion (2015–2019):
  - Deleveraging efforts and tightening of off-budget borrowing failed to halt deterioration of net financial worth.
  - Rapid expansion of SOEs without associated profitability improvements represents an emerging vulnerability.

### Methodology and scope
- Financial balance sheet compiled for the general government following IMF’s Government Finance Statistics Manual (GFSM 2014) principles.
- Focus on general government: central and subnational governments, social security funds, and certain government-guided funds.
- Exclusions/inclusions:
  - Excludes SOEs and public financial corporations from the general government perimeter.
  - Includes most LGFVs and public funds such as the Special Construction Fund (SCF) and government-guided funds (GGF).
- Covered financial items only: currency and deposits; monetary gold and SDRs; debt securities; loans; equity and investment fund shares; insurance and pension entitlements; financial derivatives; other accounts receivable/payable.
- Relied on official statistical sources and IMF China Article IV staff reports, complemented with academic estimates to fill data gaps.
- Leveraged IMF’s public sector balance sheets database to benchmark China and conduct an event study comparing China’s public corporation restructuring episode (1997–2007) with global episodes.

### Key statistics and composition (2019 unless otherwise noted)
- General government financial assets: US$12.5 trillion in 2019.
- Government equity holdings of public corporations (financial and nonfinancial): 68 percent of GDP.
  - These equity holdings account for over three-quarters of government financial assets.
  - Benchmark countries’ average equity and investment funds shares: about 22 percent of GDP.
- Currency and deposits holdings: 18 percent of GDP by 2019.
  - These holdings more than doubled in percent of GDP over the last two decades and are on average over 10 percentage points of GDP higher than comparators.
- National social security funds: 2.4 percent of GDP in 2019.
  - Return on equity of the national social security funds ranged from -2.3 to 14 percent over the last two decades.
  - Benchmark countries’ financial assets from insurance, pension, and standardized guarantee schemes: 0.1 percent of GDP on average.
- Financial liabilities (including debt and off-budget borrowing): estimated at 81 percent of GDP in 2019 — roughly the same as the average of benchmark comparators.
- Valuation differences with official sources:
  - Authors’ estimates of financial assets are about 15 percent of GDP lower than Ministry of Finance and CASS estimates, mainly due to differing valuation of SOEs.

### Key estimation challenges and data gaps
- Consolidation challenges:
  - Lack of data on cross-holdings among central and local governments, public financial and nonfinancial corporations.
  - China had over 700,000 public institutions and another 217,000 nonfinancial SOEs in 2019, complicating coverage and consolidation.
- Valuation challenges:
  - GFSM 2014 requires market or fair value measurement, but market prices are unavailable for many assets and liabilities.
  - Valuation of nonfinancial assets (land, infrastructure, natural resources) is particularly problematic due to changing land-use classifications, regional price variation, scarce data, and complex ownership.
  - No sound and verifiable valuation of infrastructure investment given scarce data on investment returns.
- Data gaps prevent estimation for 2020–22: equity valuation data for SOEs not available for 2021–22; 2020 data subject to revision; state holdings in financial institutions not directly available.

### Historical evolution (three policy periods and fiscal impacts)
- Periodization:
  - 1997–2007: restructuring of SOEs and state-owned banks.
  - 2008–2014: scaling-up of public investment at onset of the GFC, relying largely on LGFVs.
  - 2015 onward: attempts to rein in local government off-budget borrowing while relying on local governments for policy objectives.
- Fiscal aggregates and balance-sheet impacts (1997–2007 restructuring):
  - General government expenditure increased by 5.7 percent of GDP (1997–2002).
  - Primary deficit widened by ½ percentage points (1997–2002).
  - Fiscal balance bottomed out at -2.4 percent of GDP in 2002, improved to a small surplus by 2007.
  - Budgetary debt increased by just 4 percent of GDP in the decade since 1997; accounting for off-budget operations, the increase topped 8½ percent of GDP.
  - Net financial worth declined by over 20 percent of GDP before reaching a trough in 2001, then rose to 30 percent of GDP in 2007.
- Restructuring costs and interventions (exact figures):
  - Capital injections to the largest banks (six of the seven commercial banks) amounted to around RMB 4 trillion.
  - Nonperforming loans transferred to asset management companies: RMB 1.4 trillion.
  - Cost of restructuring to the general government (excluding the central bank) estimated at 11 percent of 2005 GDP.
  - RMB 270 billion special government bond direct issuance in 1998 to recapitalize the four largest state-owned banks.
  - Write-ups/creation: government bonds were issued to establish the China Investment Corporation (CIC) in September 2007.

### The scaling-up of domestic public investment (2008–2014)
- Fiscal stimulus program announced November 2008: fiscal package of RMB 4 trillion (about 11 percent of annual GDP), to be spent by 2010.
  - By some estimates, half of the stimulus planned for public infrastructure projects and one-quarter on repairs and maintenance.
  - Investment-to-GDP increased substantially by 5 percentage points of GDP in 2009–10.
- Financing and off-budget activity:
  - In practice, the stimulus involved only a modest cumulative fiscal expansion of 2 percent of GDP as measured by the headline fiscal deficit; about 70 percent of the stimulus was financed off-budget (RMB 2.8 trillion out of RMB 4 trillion).
  - When accounting for off-budget borrowing, government debt rose by 21 percent of GDP during 2008–2013.
- LGFVs and off-budget arrangements:
  - Typical approach: local government transfers ownership of land to an LGFV, which uses the land as collateral to borrow; local governments often provided guarantees and shared debt servicing responsibilities.
  - NAO identified 7,170 LGFVs in 2014; CBIRC publications indicated 11,566 LGFVs; Ministry of Finance expanded NAO list in 2018.
- SOE expansion and performance (2008–13):
  - Asset levels expanded by 45 percentage points of GDP.
  - Over a third of SOEs incurred losses, amounting to a total of 1½ percent of GDP per year.
  - By end-2015, about 10 percent of SOEs were estimated to be nonviable zombie companies.
  - Provincial-level SOEs reported weaker profitability than central SOEs.

### Attempts to reign in LGFV debt while pursuing SOE expansion (2015–2019)
- Policy and recognition:
  - Budget law revised in 2014 to “open the front door and close the back door,” tightening off-budget activity while allowing provincial governments to issue bonds subject to an annual cap.
  - Following audits, the central government recognized about 22 percent of GDP of LGFV debt as general government debt in 2014.
- Bond-swap program:
  - A bond-swap program (RMB 18 trillion or 25 percent of 2015 GDP) replaced high-interest, short-duration debt with provincial government bonds of 2–20 years maturity.
  - National People’s Congress authorized issuance of RMB 1.6 trillion in bonds in 2015: RMB 600 billion for new financing and RMB 1 trillion for refinancing maturing LGFV debt under a three-year debt-swap program.
- Regulatory measures:
  - 2018 New Regulation on asset management aimed to curb financing of ‘non-standard’ assets; regulators later relaxed bond issuance rules and allowed policy banks to refinance local government debt in pilot cities.
- Fiscal outcomes and balance-sheet dynamics (2014–19):
  - Headline fiscal deficits remained in a narrow range of 2.1 to 2.8 percent of GDP during 2014–19.
  - Official government debt remained flat at just below 40 percent of GDP.
  - Broader fiscal deficit (including off-budget borrowing) rose from 7 percent of GDP in 2014 to 12.4 percent of GDP in 2019.
  - Net financial worth trended downward during 2014–19 and by 2019 was estimated at less than a quarter of the 2007 peak.
  - Divergent paths since 2008: central government strengthened position while local governments experienced steady decline, entering negative territory in 2016.
  - Following recognition of 22 percent of GDP in LGFV borrowing to government debt, off-budget local government borrowings continued to rise by 25 percentage of GDP during 2015–19.

### State-owned enterprises (SOEs), financial public corporations, and financial linkages
- SOE counts and structure:
  - 217,000 nonfinancial SOEs in China.
  - 96 central state-owned conglomerates supervised by SASAC; conglomerates have layers of subsidiaries resulting in about 62,000 central SOEs.
  - More than 155,000 local SOEs (without accounting for LGFVs).
- SOE roles:
  - Key to development strategy and contributors to macroeconomic stabilization; providers of social functions including employment stabilization and social services.
- Financial public corporations:
  - Big Four banks are majority state-owned.
  - Policy banks (China Development Bank and two others) are fully state-owned.
  - 22 largest financial institutions used in government equity-holdings estimates account for over 85 percent of the banking system.
- SOE performance and risks (2014–19):
  - Government equity rose by 7 percentage points to 49 percent of GDP in 2019, largely driven by local SOEs.
  - More than 40 percent of SOEs incur losses every year; losses amount to about 2½ percent of GDP per year.
  - Highly leveraged SOEs tend to have lower profitability despite preferential credit access.

### Benchmarking, event-study results, and growth dynamics
- Event study of 47 episodes (26 banking crises and 21 SOE reforms) for 35 countries using an 8-year window (3 years before and 5 years after start) finds:
  - China experienced a rapid deterioration in net financial worth early in episodes followed by a strong recovery; five years after event start China registered growth rates more than 2 percentage points above trend, while the average country was about 1 percentage point below trend.
  - China’s experience shows stronger recovery likely contributed to relatively stronger fiscal and balance-sheet outcomes.

### Conclusions and policy recommendations
- Main conclusions:
  - China had a relatively strong government balance sheet compared to peers prior to the Covid-19 pandemic, reflecting large payoffs from early-2000s SOE restructuring.
  - Gains from earlier reforms have eroded since the GFC as SOE reform momentum slowed and local budget constraints relaxed.
  - The Achilles heel for China lies with local governments and SOEs.
- Policy recommendations and priorities:
  - Implement a comprehensive restructuring agenda and institutional fiscal reforms.
  - Contain LGFV leverage, phase out implicit guarantees, and strengthen corporate and banking restructuring frameworks.
  - Manage state-owned capital to keep state shareholders at arm’s length from SOE management, increase dividend payments to the budget, and phase out SOE social functions.
  - Harden SOE budget constraints by ending implicit guarantees and preferential credit access; carefully allocate losses to owners and creditors.
  - Improve fiscal data and disclosures:
    - Make public-sector balance sheet data more readily available in detail and timing, building on the 2018 government balance sheet release.
    - Publish government’s equity holdings of public corporations with sub-component breakdowns and disclose valuation and consolidation methodology.
    - Improve reporting to better assess vulnerabilities amid off-budget spending, rising exposures abroad, and data lags.

_International Monetary Fund — Appendix 1. The Public Sector in China (from wpiea2023154-print-pdf)_

### Appendix 1. The Public Sector in China .................................................................................

### Appendix 1. The Public Sector in China

### I. Introduction
- Over the last quarter of a century, China has proactively used fiscal policy to weather economic shocks.
- The agile policy responses during this period have been credited with:
  - staving off contagion in the Asian financial crisis (World Bank 1999),
  - preventing a sharp downturn from global spillovers during the Great Recession (Wong 2011; Fardoust, Lin, and Luo 2015),
  - mitigating the economic effects of the Covid-19 pandemic (IMF 2021).
- Quotation included: “上古結繩而治 , 後世聖人易之以書契 , 百官以治 , 萬民以察 , 蓋取諸《夬》”  易傳 ∙ 繫辭下 — “People tied knots in ancient times and later used a ledger to record; officials used those records to govern, while the public hold them accountable with such records—that reflects how policies are made.” Commentaries to the Book of Changes, Confucius, ~500 B.C.
- Text indicates: “On average, China has posted an annual growth rate of” (sentence incomplete in source).

### Appendix structure and major themes (as listed in the source)
- Appendix 2. Methodological Challenges in Estimating Non-Financial Assets
  - Land
  - Infrastructure
  - Natural resources
- Appendix 3. Data Sources and Estimation Methodology
  - A. Financial assets
    - 1. Fiscal deposits and currency holdings
    - 2. National and local social security funds
    - 3. Equity holdings of public corporations
    - 4. State-owned financial institutions
  - B. Financial liabilities
- Appendix 4. Comparison of Local Government Financing Vehicles and State-owned Enterprises—an Illustrative Example
  - An illustrative example
- References

### Implicit analytic focus based on appendix headings
- Estimation challenges for non-financial public assets are organized around Land, Infrastructure, and Natural resources.
- Estimation methodology separates Financial assets and Financial liabilities, with Financial assets further broken down into fiscal deposits and currency holdings; national and local social security funds; equity holdings of public corporations; and state-owned financial institutions.
- Appendix 4 provides an illustrative example comparing Local Government Financing Vehicles and State-owned Enterprises.

_International Monetary Fund — Appendix 1. The Public Sector in China (from wpiea2023154-print-pdf)_

### 8.4 percent since 1997—the second highest in the world—reflecting economic fundamentals and

### 8.4 percent since 1997—the second highest in the world—reflecting economic fundamentals and

### Summary findings
- The paper provides new estimates of China’s government financial balance sheet covering 1997–2019 to offer a more comprehensive view of fiscal vulnerabilities than standard debt and deficit metrics.
- China has the highest financial assets in the world, amounting to $12.5 trillion in 2019.
- Equity holdings of financial and nonfinancial public corporations account for the largest share of those assets and are, at 68 percent of GDP, about three times higher than the average country in the sample.
- Net financial worth is estimated at 7¼ percent of GDP, which places China in the top 15 percentile of the distribution of countries with available data.
- Despite large financial assets, net financial worth has been on a steady decline since its peak in 2007, mainly driven by:
  - rising local government debt;
  - weak SOE performance; and
  - falling infrastructure returns.
- The bulk of the decline in net financial worth is explained by subnational governments, with likely significant variation across provinces.

### Major drivers and episodes shaping the balance sheet
- Three major policy shifts shaped the evolution of the government financial balance sheet:
  - Restructuring of public corporations in the early 2000s produced a strong rebound in asset valuation and a sharp turnaround in net financial worth. Empirically, China’s performance was unprecedented compared to similar episodes worldwide: whereas the net financial worth remains 10 percentage points of GDP below normal levels in the average country five years after the start of a restructuring, it was 20 percent above in the case of China.
  - Investment-led stimulus beginning at the onset of the global financial crisis (GFC) and continuing thereafter is associated with a gradual decline in net financial worth despite rising financial assets, largely reflecting proliferation of off-budget financing vehicles that allowed local governments to circumvent budget constraints.
  - Pre-pandemic deleveraging efforts and tightening of off-budget borrowing failed to halt the deterioration of net financial worth. Rapid expansion of SOEs without associated profitability improvements represent an emerging vulnerability.

### Policy-relevant analysis and implications
- Standard deficit measures for the budgetary government do not capture the fiscal impulse or fiscal risks arising from:
  - investment and social spending of SOEs;
  - credit extended by state-owned commercial and policy banks; and
  - off-budget borrowing by local governments (e.g., LGFVs).
- Solely focusing on stock variables such as debt ignores the accumulation of productive assets; a financial balance sheet approach helps capture both assets and liabilities.
- The broader public sector perimeter is relevant in China because economic development and macroeconomic stabilization frequently rely on entities outside the central and local government budgets.
- Distortions from fiscal policy interventions, while supporting growth, could have damaging long-term effects on China’s aggregate productivity.

### Methodology and scope
- The financial balance sheet is compiled for the general government following IMF’s Government Finance Statistics Manual (GFSM 2014) principles.
- The analysis relies on official statistical sources and IMF China Article IV staff reports, complemented with academic estimates to fill data gaps.
- The paper focuses on the general government (central and subnational governments, social security funds, and certain government-guided funds), but excludes SOEs and public financial corporations from the general government perimeter while including most LGFVs and public funds such as the Special Construction Fund (SCF) and government-guided funds (GGF).
- The study covers financial items only (currency and deposits, monetary gold and SDRs, debt securities, loans, equity and investment fund shares, insurance and pension entitlements, financial derivatives, other accounts receivable/payable) and avoids valuation of nonfinancial assets because their market valuation is highly uncertain.
- The authors leverage the IMF’s database on public sector balance sheets to benchmark China against other countries and conduct an event study comparing China’s public corporation restructuring episode (1997–2007) with similar global episodes.

### Key estimation challenges highlighted
- Consolidation challenges:
  - Lack of data on cross-holdings among central and local governments, public financial and nonfinancial corporations makes meaningful consolidation difficult.
  - Official data provide central and local government debt (excluding LGFV liabilities) and some state equity holdings, but do not provide cross-holdings between public corporations and the government sector.
  - China had over 700,000 public institutions and another 217,000 nonfinancial SOEs in 2019, complicating coverage and consolidation.
- Valuation challenges:
  - Market or fair value measurement is required under GFSM 2014, but market prices are not available for many assets and liabilities.
  - Valuation of nonfinancial assets (land, infrastructure, natural resources) is particularly problematic because:
    - land owned by general government changes across official data releases depending on land use classification;
    - land prices vary significantly across regions and land use rights and relevant data are scarce;
    - there is no sound and verifiable valuation of infrastructure investment given complex ownership and scarce data on investment returns.
- Data gaps prevent estimation of the government balance sheet for 2020–22: specifically, data necessary to estimate valuation of equity holdings in SOEs are not available for 2021–22, and data for 2020 are subject to revision; data on state holding of financial institutions are also not directly available.

### Comparative and literature context
- Earlier compilations of government balance sheets were limited by data availability; renewed efforts followed the emerging market crises of the 1990s and later IMF work has explored central and general government balance sheets across many countries.
- Studies on China’s sovereign and public sector balance sheets emerged over a decade ago as local government debt rose; some earlier studies (covering 2000–2010 or 1998–2012) generally concluded public sector financial positions were healthy.
- This paper extends prior work by:
  - presenting the most comprehensive estimates of China’s government financial balance sheet to date (covering up to pre-Covid-19 period);
  - following GFSM 2014 and estimating market values when data permit, resulting in significant valuation adjustments for SOEs and general government debt relative to previous studies; and
  - benchmarking China empirically against other countries undergoing SOE reforms.

*Source: https://www.imf.org/-/media/files/publications/wp/2023/english/wpiea2023154-print-pdf.pdf*

### Appendix 3). For example, the write-off of nonperforming assets results in a loss of government

### wpiea2023154-print-pdf - Appendix 3). For example, the write-off of nonperforming assets results in a loss of government

### Methodology and fiscal solvency metric
- Financial liabilities scope: official (central and local) government debt plus estimated off-budget borrowings—debt of LGFVs and other leveraged-government funds classified within the perimeter of the general government.
- Net financial worth definition: difference between financial assets and liabilities; shows the extent to which general government liabilities are covered by financial assets and is a measure of fiscal solvency.
- Data caveat: due to data constraints some instruments may be missing on one side (e.g., debt securities, accounts receivables on assets; accounts payable, pension liabilities on liabilities).

### Composition of China’s general government balance sheet (2019) — key facts
- Net financial worth: 7 percent of GDP in 2019.
  - China is at the top 15th percentile of the distribution of a 50-country sample.
- General government financial assets: US$12.5 trillion in 2019 — the highest in the world.
  - Government equity holdings of public corporations (financial and nonfinancial): estimated at 68 percent of GDP and account for over three-quarters of government financial assets.
  - By comparison, average equity and investment funds shares in benchmark countries: about 22 percent of GDP.
- Currency and deposits holdings: 18 percent of GDP by 2019.
  - These holdings more than doubled in percent of GDP over the last two decades and are on average over 10 percentage points of GDP higher than comparators.
  - Implication: room for improving cash management and better recording of related account payables.
- National social security funds: 2.4 percent of GDP in 2019.
  - Return on equity of the national social security funds ranged from -2.3 to 14 percent over the last two decades.
  - Benchmark countries’ financial assets from insurance, pension, and standardized guarantee schemes: 0.1 percent of GDP on average.
- Financial liabilities (including debt and off-budget borrowing): estimated at 81 percent of GDP in 2019 — roughly the same as the average of benchmark comparators.

### Valuation differences with other studies and sources
- Overall difference: the authors’ estimates of financial assets are about 15 percent of GDP lower than the Ministry of Finance and the Chinese Academy of Social Sciences (CASS).
- Main source of discrepancy: valuation of SOEs.
  - Authors use net present value of expected future profits and account for undistributed losses and nonperforming assets in the early 2000s; official sources use book value of government equity holdings.
- Other asset differences:
  - Government’s equity holdings of financial institutions: aligned with Ministry of Finance (covers both bank and nonbank financial institutions); CASS valuation is smaller due to exclusion of nonbank financial institutions.
  - Financial assets of government administration units and social organizations (mostly currency and deposits): similar to Ministry of Finance but lower than CASS (CASS includes gross amount before potential account payables).
- Limitations: dearth of publicly available data prevents full adjustment of bank and nonbank financial corporations’ valuations for portfolio returns and nonperforming assets; hence equity holdings estimates of financial institutions are very close to book value.

### Historical evolution of public wealth (pre-Covid-19) — three policy periods
- Periodization:
  - 1997–2007: restructuring of SOEs and state-owned banks.
  - 2008–2014: scaling-up of public investment at the onset of the GFC, relying largely on LGFVs.
  - 2015 onward: attempts to rein in local government off-budget borrowing while relying on local governments for policy objectives.

### The restructuring of public corporations (1997–2007) — background and outcomes
- SOE sector before reforms:
  - Late 1970s: SOEs accounted for 78 percent of total industrial output and 75 percent of urban employment.
  - By the second half of the 1990s over two-thirds of SOEs were in losses with average return on equity of -4 percent.
- Key reform actions:
  - “Grasping the large, letting go of the small” (zhuada fangxiao) from 1995: consolidate largest SOEs in strategic industries; privatize or close smaller SOEs.
  - Establishment of SASACs in 2003 to improve SOE governance.
  - Phased approach to banking-sector cleanup: initial focus on big four banks, subsequent waves covering rest of financial sector.
  - Financial sector liberalization: private and foreign participation, IPO listings, strengthened regulation (e.g., China Banking Regulatory Commission in 2003).
- Outcomes by 2007:
  - Number of SOEs declined from 262,000 in 1997 to 116,000 a decade later.
  - More than two-thirds of operating industrial SOEs were privatized or shut down between 1998 and 2007.
  - SOE share of industrial value-added declined by more than half; over 30 million workers were laid off.
  - SOE share in urban employment fell to less than 15 percent.
  - Labor productivity of surviving SOEs and privatized firms converged to that of private firms by 2007.
- Banking sector outcomes:
  - Capital ratio of the banking system: 2.5 percent in 2005, climbed to 8.1 percent by 2007.
  - NPL ratio fell to 1.1 percent by end-2010.
  - Government scaled down equity stake in the financial system, though much of banking sector remained partly state-owned.

### Fiscal and balance-sheet impacts of restructuring (1997–2007) — exact figures
- Fiscal aggregates (1997–2002):
  - General government expenditure increased by 5.7 percent of GDP.
  - Primary deficit widened by ½ percentage points (helped by higher revenues).
  - Fiscal balance bottomed out at -2.4 percent of GDP in 2002, improved thereafter reaching a small surplus by 2007.
  - Budgetary debt increased by just 4 percent of GDP in the decade since 1997.
  - Accounting for off-budget operations, the increase in debt topped 8½ percent of GDP.
- Off-budget operations and mobilization:
  - RMB 660 billion cumulative tailored placements of government debt to banks for long-term projects during 1998–2003 (averaging 1 percent of GDP per year).
  - These placements mobilized bank credit and other social capital of a total scale of RMB 3.3 trillion (24 percent of 2003 GDP).
- Balance-sheet trajectory:
  - Net financial worth declined by over 20 percent of GDP before reaching a trough in 2001.
  - Net financial worth rose to 30 percent of GDP in 2007.
- Asset valuation changes driving balance-sheet shifts:
  - SOEs: valuation in 2000 was 11 percent of GDP (about two-thirds of 1997 level); by 2007 the government’s stake value rose to 39 percent of GDP.
  - SOCBs: government equity participation value turned negative in 2000, reached a trough in 2002 with accumulated losses in net financial worth amounting to over 10 percent of GDP.
- Restructuring costs and interventions:
  - Capital injections to the largest banks (six of the seven commercial banks) amounted to around RMB 4 trillion, split among the general government, the central bank, and private/foreign investors.
  - Nonperforming loans transferred to asset management companies: RMB 1.4 trillion.
  - Cost of restructuring to the general government (excluding the central bank) estimated at 11 percent of 2005 GDP.
  - RMB 270 billion special government bond direct issuance in 1998 to recapitalize the four largest state-owned banks.
  - Government bonds were issued to establish the China Investment Corporation (CIC) in September 2007.

*International Monetary Fund — wpiea2023154-print-pdf (Appendix content reproduced as provided)*

### 1. Financial Assets and Liabilities

### 1. Financial Assets and Liabilities

### B. The scaling-up of domestic public investment (2008–2014) — Background
- China announced a fiscal stimulus program in November of 2008 with a fiscal package of RMB 4 trillion (amounting to about 11 percent of annual GDP), to be spent by 2010.
- By some estimates, half of the stimulus was planned for public infrastructure projects and one-quarter on repairs and maintenance.
- Investment-to-GDP increased substantially by 5 percentage points of GDP in 2009–10.
- Monetary policy was eased including multiple reductions of required reserve ratios and cuts in official benchmark interest rates, resulting in a remarkable credit expansion.
- Credit growth disproportionately favored state-owned firms and firms with lower average product of capital, reversing earlier capital reallocation toward private firms.
- In practice, the stimulus involved only a modest cumulative fiscal expansion (as measured by the headline fiscal deficit) of 2 percent of GDP; the rest was mostly financed through off-budget relaxation of local government financial constraints.
- About 70 percent of the stimulus was financed off-budget (RMB2.8 trillion out of a total 4 trillion).
- Initially local governments financed additional spending through bank loans; when faced with big rollovers in 2012 they resorted to nonbank debt financing and increased municipal corporate bond issuance and shadow banking activities during 2012–2015.
- Typical off-budget arrangement: local government transfers ownership of land to an LGFV, which uses the land as collateral to borrow; local governments often provided guarantees on LGFVs’ debt and shared debt servicing responsibilities.
- LGFV prevalence and classification: the National Audit Office identified 7,170 LGFVs in its 2014 report; publications by the China Banking and Insurance Regulatory Commission indicated 11,566 LGFVs; the Ministry of Finance used the NAO list and expanded it in 2018.
- When accounting for off-budget borrowing, government debt rose by 21 percent of GDP during 2008–2013.
- Emerging evidence from individual LGFVs suggests returns on investment declined over time and on average were below interest cost, with the gap widening during 2010–15.
- Inter-governmental fiscal imbalances: about a quarter of provinces had combined local government and LGFV debt exceeding 100 percent of their GDP, while one third had combined debt less than 50 percent of GDP.
- SOE expansion during 2008–13:
  - Asset levels expanded by 45 percentage points of GDP.
  - While SOEs recorded positive net profits at the aggregate level, returns on assets stayed very low and on a declining trend.
  - Over a third of SOEs incurred losses, amounting to a total of 1½ percent of GDP per year.
  - By end-2015, about 10 percent of SOEs were estimated to be nonviable zombie companies.
  - Provincial-level SOEs reported weaker profitability than central SOEs.
  - Rapid SOE expansion may have crowded out private activity and increased overseas exposures.

### Fiscal impact and the government’s balance sheet
- Headline fiscal deficits widened from -0.8 percent of GDP in 2008 to -2.1 percent of GDP in 2013; the largest deficit was -2.7 percent of GDP in 2009.
- Official government debt declined from 17.8 to 16 percent of GDP during 2008–13.
- Broader measures including off-budget LGFV activity show deficits widening from -3.2 percent to -7.6 percent of GDP during 2008–2013, reaching -11.1 percent of GDP at its peak in 2009.
- Government net financial worth:
  - Estimated at 29 percent of GDP at the onset of the GFC.
  - Net financial worth registered a decline during 2008–2013, driven mostly by local governments; the central government strengthened its position while local governments experienced steady decline.
- Financial liabilities and debt measures:
  - The increase in financial liabilities was largely explained by off-budget local government borrowing.
  - The broader measure of debt (including LGFVs) surged by 14 percentage points of GDP during 2008–13, reaching 48 percent of GDP by 2013.
- Asset-side changes:
  - Equity holdings of SOEs went up by 20 percentage points of GDP during 2008–13.
  - The gap between government asset returns (and those on nonfinancial infrastructure assets) and the interest rate on debt widened, with an estimated shortfall of 1½-2 percentage points during 2013–15, partly driven by low profitability among SOEs.
  - Aggregate result: worsening efficiency in allocation of capital and lower aggregate growth rates.

### C. Attempts to reign in LGFV debt while pursuing SOE expansion (2015–2019) — Background
- Policy response: audits of LGFV activities and measures to rein-in off-budget activity; the budget law was revised in 2014 to “open the front door and close the back door,” tightening off-budget activity while allowing provincial governments to issue bonds subject to an annual cap.
- Following audits, the central government recognized about 22 percent of GDP of LGFV debt as general government debt in 2014.

*Source: wpiea2023154-print-pdf - 1. Financial Assets and Liabilities*

### 15. A bond-swap program (RMB 18 trillion or 25 percent of 2015 GDP) has gradually replaced high-

### 15. A bond-swap program (RMB 18 trillion or 25 percent of 2015 GDP) has gradually replaced high-

### Bond-swap program and local government debt instruments
- A bond-swap program (RMB 18 trillion or 25 percent of 2015 GDP) replaced high-interest, short-duration debt (bank loans, LGFV bonds, trusts, and other nonstandard borrowing) with provincial government bonds of 2–20 years maturity.
- Objectives and outcomes:
  - Extended debt maturities.
  - Reduced interest cost.
  - Standardized local government debt instruments.
  - Could have crowded out private investment (Lam, Wei, and van Eden 2017).
- National People’s Congress action in 2015:
  - Authorized issuance of RMB1.6 trillion in bonds in 2015, of which RMB600 billion was for new financing and RMB1 trillion for refinancing maturing LGFV debt under a three-year debt-swap program.

### Regulatory measures to curb shadow banking and LGFV financing
- In 2018, the China Banking and Insurance Regulatory Commission, the China Securities Regulatory Commission, and the State Administration of Foreign Exchange jointly issued a new asset management regulation (the New Regulation) to curb financing of ‘non-standard’ assets with maturity mismatch and rigid repayment problems.
- The New Regulation was expected to affect demand for LGFVs because securities companies and fund companies are key LGFV service providers.
- Regulators later relaxed bond issuance rules and allowed policy banks to refinance local government debt in pilot cities.

### Fiscal management, fiscal risks, and constraints on LGFV activity
- Authorities introduced measures to improve fiscal management and public accountability:
  - Standardize accounting and financial reporting.
  - Improve performance monitoring to enhance fiscal oversight.
  - Ministry of Finance established an early-warning system and risk management guidelines to monitor subnational fiscal risks.
  - Measures include: (i) loss of fiscal authority of local administration while ensuring minimum service levels; (ii) cross-agency oversight to rein in government guarantees; (iii) regulations on local government bond financing against land development and toll-road income.
- Despite reforms, off-budget borrowing persisted due to unresolved structural revenue shortfalls at local level; local governments continued to finance through land sales and specialized financing vehicles.

### LGFV debt and market perceptions
- Since the 2014 debt recognition, LGFV debt continued to rise by 6 percentage points of GDP to reach an estimate of 34 percent of GDP as of end 2018.
- LGFV profitability and interest coverage deteriorated.
- Qiu et al. (2022) find:
  - Strong negative correlation between the credit spread of LGFV bonds and the revenue of local governments.
  - Credit spread at issuance is lower in provinces with higher income per capita and this relationship strengthened after the New Regulation.
- Evidence suggests the implicit local government guarantee perceived by markets did not weaken after the New Regulation.

### SOE reforms, performance, and financial sector linkages
- Government initiatives to reinvigorate SOE reforms aimed to:
  - Reposition the state as capital investor rather than operator; pursue mixed-ownership reforms, professional management, and clearer owner–board responsibilities.
  - Classify SOEs into three categories with specific ownership structures and assessment criteria: (a) strategic SOEs; (b) commercial nonstrategic SOEs; (c) SOEs with social functions.
  - Resolve nonviable SOEs: commitments to cut aggregate SOE losses by 2017 and expedite exit of near 350 subsidiaries of central SOEs and near 4,000 local SOEs.
- Implementation was uneven:
  - Pilot programs on mixed-ownership reforms launched from 2016.
  - Reform of central SOEs advanced slowly due to complex multi-layer subsidiary structures.
  - Despite reforms, expansion of SOEs during 2015–19 led government equity to rise by 7 percentage points to 49 percent of GDP in 2019, largely driven by local SOEs.
- Financial linkages and credit allocation:
  - Credit to SOEs accounted for two-thirds of bank credit and contributed to overall corporate leverage increases.
  - Market perceptions shifted after waves of defaults, fragmenting the bond market; SOEs in certain provinces face higher funding costs.
- SOE financial performance:
  - More than 40 percent of SOEs incur losses every year.
  - Losses amount to about 2½ percent of GDP per year.
  - Highly leveraged SOEs tend to have lower profitability despite preferential credit access and lower interest costs.

### Fiscal impact and government balance sheet dynamics (2014–19)
- Headline fiscal and deficit statistics:
  - Headline fiscal deficits remained in a narrow range of 2.1 to 2.8 percent of GDP during 2014–19.
  - Official government debt remained flat at just below 40 percent of GDP.
  - Broader fiscal deficit (including off-budget borrowing) rose from 7 percent of GDP in 2014 to 12.4 percent of GDP in 2019.
- Government balance sheet:
  - Net financial worth trended downward during 2014–19 but remained positive at the aggregate level.
  - Estimates for 2019 put net financial worth at less than a quarter of the peak level reached in 2007.
  - Divergent paths since 2008: central government strengthened position while local governments experienced steady decline, entering negative territory in 2016.
  - The decline in net financial worth was driven by continued off-budget local government borrowing.
  - Following recognition of 22 percent of GDP in LGFV borrowing to government debt, off-budget local government borrowings continued to rise by 25 percentage of GDP during 2015–19.
  - Local governments increased borrowings via newly created government-guided funds and special construction funds.

### Benchmarking China: event-study results and growth dynamics
- Sample and methodology:
  - Event study of 47 episodes (26 banking crises and 21 SOE reforms) for 35 countries, including China, using an 8-year window (3 years before and 5 years after start of episode).
  - Variables: net financial worth, public debt, overall fiscal balance, and growth.
- Key comparative findings:
  - Net financial worth:
    - Prior to event, net financial worth was more than 5 percentage points of GDP above normal times for both China and average country.
    - China’s net financial worth deteriorated rapidly to more than 15 percentage points of GDP below normal times one year after event start, then recovered swiftly to more than 20 percentage points above normal times five years after start.
    - The average country experienced a more gradual deterioration and remained below normal levels five years after event.
  - Public debt:
    - China’s public debt was already above normal prior to event; average country showed no significant pre-event difference.
    - Debt increases were steeper and more protracted for the average country, with public debt significantly above normal five years after start.
  - Overall fiscal balance:
    - Fiscal deficits widened somewhat earlier in China; trough magnitudes were similar between China and average country.
  - Real GDP growth:
    - Growth fell relative to trend before the event in all cases, with China bottoming out two years earlier than the average country.
    - China experienced a shallower downturn and stronger recovery: five years after event start China registered growth rates more than 2 percentage points above trend, while the average country was about 1 percentage point below trend.
    - Strong recovery likely contributed to relatively stronger fiscal and balance-sheet outcomes in China.

### Conclusions and policy recommendations
- Main conclusions:
  - China had a relatively strong government balance sheet compared to peers prior to the Covid-19 pandemic, reflecting large payoffs from early-2000s SOE restructuring.
  - Gains from earlier reforms have eroded since the GFC as SOE reform momentum slowed and local budget constraints relaxed.
  - The Achilles heel for China lies with local governments and SOEs.
- Policy recommendations and priorities:
  - Implement a comprehensive restructuring agenda and institutional fiscal reforms.
  - Contain LGFV leverage, phase out implicit guarantees, and strengthen corporate and banking restructuring frameworks.
  - Manage state-owned capital to:
    - Keep state shareholders at arm’s length from SOE management.
    - Increase dividend payments to the budget.
    - Phase out SOE social functions.
  - Harden SOE budget constraints by ending implicit guarantees and preferential credit access; carefully allocate losses to owners and creditors.
  - Improve fiscal data and disclosures:
    - Make public-sector balance sheet data more readily available in detail and timing, building on the 2018 government balance sheet release.
    - Publish government’s equity holdings of public corporations (financial and non-financial) with sub-component breakdowns and disclose valuation and consolidation methodology.
    - Improve reporting to better assess vulnerabilities amid off-budget spending, rising exposures abroad, and data lags.

*International Monetary Fund — wpiea2023154-print-pdf*

### Appendix 1. The Public Sector in China

### Appendix 1. The Public Sector in China

### Composition and institutional roles
- The public sector comprises the central government, local (subnational) governments, and public corporations.
- Central and local governments include:
  - Administrative units, including legislature and judicial entities.
  - Government agencies and social organizations (事业单位) such as public schools, hospitals, and research institutes that deliver services and in some instances generate and retain own revenues (for example, public universities).
- Public corporations include:
  - Nonfinancial SOEs and state-owned banks (commercial and policy banks).
  - Nonbank financial institutions (securities firms, insurance companies, and investment funds).
  - National social security funds.
- Main economic functions: service delivery by government units and public corporations’ roles in production, finance, and social stabilization.

### Central government
- The central government has relatively limited spending responsibilities, accounting for less than a quarter of the national budget.
- The 1994 tax reform resulted in the recentralization of revenues with the central government de facto controlling over half of the budget.
- Central transfers to local governments have steadily increased since the early 2000s to address local fiscal gaps.

### Subnational (local) governments
- Four formal levels: provincial, prefectural, county, and township; with an informal village government below township level.
- China’s fiscal system is highly decentralized; most government functions are carried out by subnational governments.
- Subnational responsibilities include:
  - Provision of basic public services such as education, health care, and social security.
  - Sizeable public infrastructure investment.
  - Control of land development including land leasing and (re)assigning land usage rights (important revenue sources).
- Structural fiscal imbalance:
  - Local governments account for about half of general government’s revenue while spending is more than two-thirds.
  - Central transfers are insufficient to fully cover the gap.
- Distribution of functions across tiers:
  - Counties often account for most national expenditure on education and health.

### State-owned enterprises (SOEs) and nonfinancial public corporations
- There are 217,000 non-financial SOEs in China.
- Of these, 96 are central, state-owned conglomerates supervised by the State-Owned Assets Supervision and Administration Commission of the State Council (SASAC).
- Each conglomerate has layers of subsidiaries (sometimes over six levels), resulting in about 62,000 central SOEs.
- In addition, there are more than 155,000 local SOEs (without accounting for LGFVs).
- SOEs are:
  - Key to China’s development strategy and have contributed to macroeconomic stabilization.
  - Providers of social functions including stabilizing employment and providing social services.
- Ownership and valuation issues (summary of later methodological discussion referenced in the appendix):
  - Government share of SOE equity declined from near 100 percent during 1999−2006 to about 78 percent in 2007 using one approach; an alternative measure shows a more gradual decline from 100 to 78 percent during 1997−2007.
  - Official data show SOE equity declined from 60 percent of GDP in 1999 to about 45 percent of GDP in 2007 before gradually recovering; based on a 78 percent ownership share, government-held SOE equity was valued at around 50 percent of GDP in the cited statistics.
  - Adjustments for undistributed losses and nonperforming assets imply equity values below 10 percent of GDP before rebounding to about 40 percent of GDP by the mid-2000s under one adjustment approach.
  - Post-2007 valuation challenges include lack of data on nonperforming assets and divergence between book and market values; the appendix describes using an NPV of expected future net profits over a 30-year horizon with profits assumed to stabilize at 3 percent of GDP by 2022 and a discount factor of 2 percent.

### Financial public corporations and the financial system
- Financial public corporations dominate China’s financial system.
- The Big Four banks are majority state-owned.
- China Development Bank and two policy banks are fully state-owned and provide development-related loans to rural and urban areas.
- Local governments have important ownership stakes (and in many cases control) in lower tier banks.
- Local government-affiliated entities also own nonbank subsidiaries and other intermediaries.
- The state controls a large share of insurance companies, trusts, and securities companies.
- There is evidence that pervasive state control influences credit allocation.

### Intergovernmental structure and aggregate indicators (as presented)
- Total population: 1.39 billion
- National employment: 517.5 million
- Central SOEs (aggregate reported in figure text): 58,000 central SOEs (96 conglomerates managed by SASAC) with assets RMB76.2 trillion
- Administrative units and averages:
  - 22 Provinces and 5 autonomous regions — Average population: 48 million — Employment: 457.4 million
  - 4 Provincial-level municipalities — Average population: 23 million — Employment: 60.1 million
  - 334 Prefectural-level units — Average population: 4.2 million
  - 2,851 County-level units — Average population: 0.5 million
  - 39,888 Town-level units — Average population: 35,000
- Note (from figure): The average number of provincial level firms (which include local SOEs, collectives, joint, private, foreign, and others) is around 600,000 based on registration status.

*International Monetary Fund — Appendix 1. The Public Sector in China*

### 1. Government Equity Holdings of SOEs

### 1. Government Equity Holdings of SOEs

### Valuation and coverage of state-owned financial institutions
- Estimates include the government’s equity holdings in 22 largest financial institutions comprising state-owned commercial banks, policy banks, and national sovereign wealth funds.
- These 22 institutions account for the majority of assets in the financial system or over 85 percent of the banking system.
- Due to data constraints, smaller depository institutions (regional banks and credit cooperatives) and nonbank financial institutions (such as securities brokerages and insurance companies) are not included.
- The State Council announced aggregate value of government holdings as of end-2017 amounting to RMB 9.4 trillion (or 19 percent of GDP); no further breakdown or time series were provided and the valuation method is unclear.

### Estimation approaches for government equity holdings
- Two alternative approaches are used given data constraints:
  - For years prior to mid-2000s (before largest state-owned banks were listed):
    - Use the total owners’ equity of the overall banking system, adjusted by the asset share of state-owned banks.
    - Adjust the size of bank recapitalization by the government in the late 1990s to mid-2000s using other studies (Garcia-Herrero, Gavila, and Santabarbara 2006; Ma 2006; Yang, Zhang, and Tan 2017).
    - Adjust for nonperforming assets using data from China Public Finance Yearbooks.
  - For years after the listing of the largest state-owned banks:
    - Compile government equity holdings using published annual audited financial statements; obtain total equity valuation from CEIC.
    - Consolidate by netting out cross holdings among state-owned financial institutions using notes on top shareholders where available.
    - Example: the audited 2015 financial statement of Industrial and Commercial Bank of China (ICBC) showed Central Huijin Investment held 34.7 percent of its equity; after excluding holdings owned by China Investment Corporation, government equity holdings in ICBC were about 36 percent of total equity, a drop from gross holdings of 90 percent of total equity.
    - Adjusting for cross-holdings to the extent plausible, net government equity holdings are estimated at about 37 percent of total equity in these financial institutions, but they vary widely across individual institutions.

### Sample coverage (institutions included in the 22-sample)
- Five large state-owned banks: Agriculture Bank of China, Bank of China, Bank of Communications, China Construction Bank, Industrial and Commercial Bank of China.
- Several joint-stock banks: Everbright Bank, Industrial Bank, Minsheng Bank, China Merchants Bank, CITIC, Guangdong Development Bank, HuaXia Bank, Pudong Development Bank, Ping An Bank.
- Three policy banks: Agriculture Development Bank of China, China Development Bank, Export-Import Bank of China.
- China Postal Savings, and China Investment Corporation (a national sovereign wealth fund specialized in equity investment).
- Note: China Development Bank has changed its mandate since 2008.

### Financial liabilities and public debt measures
- Data on financial liabilities are based on estimates in IMF Article IV Staff Reports using an ‘augmented’ concept of public to include official government debt and estimates of debt incurred by entities that should have been classified as government according to the Government Finance Statistics Manual (2014).
- The nominal value of debt is reported by the Ministry of Finance and includes explicit debt of central and local governments.
- Official government debt reached 45 percent of GDP in 2020, rising significantly since 2010, notably with the recognition of 22 percent of GDP LGFV debt in 2014 under the revised budget law and related directives.
- As of 2019, off-budget borrowing amounts to 43 percent of GDP on top of official debt.
- Off-budget borrowing estimates include borrowing of nonmarket LGFVs, Government Guided Funds, and Special Construction Funds (SCFs) to the extent data are available.

### Comparison of LGFVs and local SOEs — illustrative example (Yunnan)
- Distinction and overlap:
  - LGFVs:
    - Corporations established by local governments’ finance bureaus to finance specific investment projects—often infrastructure, construction, and real estate.
    - Typically can collateralize bank loans or corporate bonds with land or other assets, either owned by the LGFV itself or pledged by the local governments.
    - Local governments tend to have significant control; historically provided guarantees and sometimes subsidized losses. The 2014 Budget Law prohibited provision of guarantees by local governments.
  - Local SOEs:
    - Owned by local governments and often supervised by local SASAC.
    - Broader mandate than LGFVs, provide a broader set of goods and services mostly on a commercial basis, and have greater access to different financing sources.
    - Can overlap with LGFVs in infrastructure activities.
- Yunnan illustrative case:
  - Yunnan Construction and Investment Holdings Company is a local SOE fully owned by Yunnan provincial-level SASAC (90 percent) and the finance bureau (10 percent). It holds or invests in other SOE shares across sectors (railway, petroleum and gas, banks, securities firms, real estate) and has multiple subsidiaries. It receives credit lines from banks, state subsidies, and other guarantees.
  - Yunnan City Construction Group is considered an LGFV, established by the finance bureau and jointly owned by the local finance bureau and local SASAC. It is the main entity for infrastructure investment in Yunnan and participates in state-authorized investment projects. Main businesses include land development, social housing, real estate, construction, utilities, logistics, travel services.
- Ownership shares illustrated in the Yunnan chart (numbers refer to ownership shares):
  - 6%, 10%, 10%, 90%, 90%, 4%, 51%, 40%, 100% (as annotated in the figure to show cross-holdings among Yunnan Finance Bureau, provincial level MOF, Yunnan provincial SASAC, provincial level SASAC, Yunnan City Construction Group, Yunnan Construction and Investment Holdings Group Company, subsidiaries and other entities).

*Source: IMF; authors’ estimates and cited public sources as presented in the original content.*

### References

### References

### Balance sheet, national accounts, and government net worth
- Allen, M., C. Rosenberg, C. Keller, B. Setser, and N. Roubini. 2002. “A Balance Sheet Approach to Financial Crises.” IMF Working Paper 02/210.  
- Alves, M., De Clerck, S., and Gamboa-Arbelaez, J. 2020. “Public Sector Balance Sheet Database: Overview and Guide for Compilers and Users,” IMF Working Paper No. 20/130.  
- Goldsmith, R.W. 1982. The national balance sheet of the United States, 1953-1980. University of Chicago Press.  
- Goldsmith, R.W. and Lipsey, R.E. 1963. The National Balance Sheet of the United States During the Postwar Period. In Studies in the National Balance Sheet of the United States, Volume 1 (pp. 41-106). Princeton University Press.  
- Dickenson, F. G., and Eakin, F. 1936. “A balance sheet of the nation’s economy,” University of Illinois.  
- Traa, B., and A. Carare. 2007. “A Government’s Net Worth,” Finance and Development 44 (2), pp. 46–49.  
- Revell, J., 1966. The national balance sheet of the United Kingdom. Review of Income and wealth, 12(4), pp.281-305.

### State-owned enterprises (SOEs), reform, and corporate structure
- Bai, C., Lu, J., Tao, Z. 2006. “The Multitask Theory of State Enterprise Reform: Empirical Evidence from China,” American Economic Review, vol. 96, iss. 2, pp. 353–57.  
- Bai, C., C.T. Hsieh, and Z.M. Song. 2016. “The Long Shadow of a Fiscal Expansion,” Brookings Papers on Economic Activity, pp. 129–181.  
- Berkowitz, D., H. Ma, and S. Nishioka. 2017. “Recasting the Iron Rice Bowl: The Reform of China’s State-Owned Enterprises,” Review of Economics and Statistics, October 2017, 99(4), pp.735–47.  
- Hsieh, C.T. and Z.M. Song. 2015. “Grasp the Large, Let Go of the Small: The Transformation of the State Sector in China’, Brookings Papers on Economic Activity, Spring 2015, pp. 295-346.  
- Lam, W.R., and A. Schipke. 2017. “State-Owned Enterprise Reform,” in Modernizing China: Investing in Soft Infrastructure, ed. Lam, Rodlauer, Schipke, 2017, International Monetary Fund, Washington, D.C.  
- Megginson, W.L., Ullah, B., and Wei, Z. 2014. “State Ownership, Soft-budget Constraints, and Cash Holdings: Evidence from China’s Privatized Firms,” Journal of Banking and Finance Vol. 48, pp. 276–291.  
- Molnar, M. and J. Lu. 2019. "State-owned firms behind China’s corporate debt", OECD Economics Department Working Papers, No. 1536.  
- Mano, R. and P. Stokoe. 2017. “Reassessing the Perimeter of Government Accounts in China”, IMF Working Paper No.17/272.

### Banking sector, financial stability, and systemic crises
- Bank for International Settlements. 1999. “Strengthening the Banking System in China: Issues and Experiences,” BIS Policy Papers, no 7, October.  
- Laeven, L, and F. Valencia. 2018. “Systemic Banking Crises Revisited,” IMF Working Paper No.18/206.  
- International Monetary Fund. 2011. “People’s Republic of China Financial System Stability Assessment,” IMF Country Report No.11/321.  
- International Monetary Fund. 2017a. “People’s Republic of China: Financial System Stability Assessment,” IMF Country Report No.17/358.  
- Lou, J. 2000. “China’s Bank Non-Performing Loan Problem: Seriousness and Causes,” International Lawyer 34 (4), pp. 1147–1192.  
- McGuinness, P. and K. Keasey. 2010. “The Listing of Chinese State-owned Banks and Their Path to Banking and Ownership Reform,” The China Quarterly, No.201, pp.125-155, March 2010.  
- Ma, G. 2006. “Sharing China’s Bank Restructuring Bill,” China and World Economy, Vol.14/3, pp. 19–37.  
- Wright, L. and D. Rosen. 2018. “Credit and Credibility: Risks to China’s Economic Resilience”, Center for Strategic and International Studies, October 2018.

### Local government finance, LGFVs, and fiscal risks
- Lam, W.R., J.N., Wei, and H. van Eden. 2017. “Local Government Finances and Fiscal Risks." in Modernizing China: Investing in Soft Infrastructure.  
- Zhang, Y.Y. and S. Barnett. 2014. “Fiscal Vulnerabilities and Risks from Local Government Finance in China,” IMF Working Paper No.14/4.  
- Liang, H. and L. Liu. 2019. “The Mounting LGFV Debt—A ‘Gray Rhino’ in China’s Balance Sheet’, CICC Macro Research, March 2019.  
- Qiu, Z., Sun, X., Wu, L., Jia, D. and Wang, Z., 2022. “The 2018 new asset management regulation and LGFV bonds in China,” Economic and Political Studies, pp.1-25.  
- Chen, Z., He, Z. and Liu, C., 2020. “The financing of local government in China: Stimulus loan wanes and shadow banking waxes,” Journal of Financial Economics, 137(1), pp.42-71.

### Productivity, growth, trade, and firm dynamics
- Hsieh, C.T. and Klenow, P.J. 2009. “Misallocation and manufacturing TFP in China and India,” The Quarterly journal of economics, 124(4), pp.1403-1448.  
- Brandt, L., Van Biesebroeck, J., Wang, L., and Y. Zhang. 2017. “WTO Accession and Performance of Chinese Manufacturing Firms,” American Economic Review, Vol. 107(9), pp. 2784–2820.  
- Tombe, T. and X. Zhu. 2019. “Trade, Migration, and Productivity: A Quantitative Analysis of China,” American Economic Review, 109, 1843–1872.  
- Zhu, X. 2012. “Understanding China’s Growth: Past, Present, and Future,” Journal of Economic Perspectives, Vol. 26, No. 4 (Fall), pp. 103–124.  
- Song, Z., K. Storesletten, and F. Zilibotti. 2011. “Growing Like China,” American Economic Review, Vol.101, No.1, pp.196–233.  
- Cerdeiro, D. and C. Ruane, 2022. “China's Declining Business Dynamism,” IMF Working Paper No. 22/32.

### Misallocation, credit allocation, and corporate debt
- Cong, L.W., Gao, H., Ponticelli, J. and Yang, X., 2019. “Credit allocation under economic stimulus: Evidence from China,” The Review of Financial Studies, 32(9), pp.3412-3460.  
- Firth, M., Lin, C., Lu, P., and Wong, S. M. L. 2009. “Inside the Black Box: Bank Credit Allocation in China’s Private Sector,” Journal of Banking & Finance, Vol. 33, pp. 1144–1155.  
- Huang, Y., Pagano, M., and Panizza, U. 2020. “Local crowding-out in China,” The Journal of Finance, 75(6), 2855-2898.  
- Wei, X., Chen, Y., Zhou, M., and Zhou, Y. 2016. “SOE Preference and Credit Misallocation: A Model and Some Evidence from China,” Economic Letters No. 138, pp. 38–41.  
- Maliszewski, W., et al. 2016. “Resolving China’s Corporate Debt Problem,” IMF Working Paper, No.16/203.

### Institutional, policy, and methodological contributions
- International Monetary Fund. 2018a. “Managing Public Wealth,” Fiscal Monitor (October).  
- International Monetary Fund. 2020. “State-Owned Enterprises: The Other Government,” Fiscal Monitor (April), Chapter 3.  
- Lam, W.R., M. Rodlauer, A. Schipke. 2017. “Modernizing China: Investing in Soft Infrastructure”, International Monetary Fund.  
- Mano, R. and P. Stokoe. 2017. “Reassessing the Perimeter of Government Accounts in China”, IMF Working Paper No.17/272.  
- Alves, M., De Clerck, S., and Gamboa-Arbelaez, J. 2020. “Public Sector Balance Sheet Database: Overview and Guide for Compilers and Users,” IMF Working Paper No. 20/130.

*Source: wpiea2023154-print-pdf - References*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2023/english/wpiea2023154-print-pdf.pdf_