Information about Asia and the Pacific Asia y el Pacífico
Journal Issue
Share
Article

People’s Republic of China: Staff Report for the 2017 Article IV Consultation

Author(s):
International Monetary Fund. Asia and Pacific Dept
Published Date:
August 2017
Share
  • ShareShare
Information about Asia and the Pacific Asia y el Pacífico
Show Summary Details

Context: Stimulus Keeps Growth Strong but Vulnerabilities Remain High

1. While growth has remained strong, vulnerabilities continue to accumulate, pointing to the need to accelerate reform efforts already in train. Growth has been bolstered by a supportive macro-policy mix, strengthening external demand as well as progress in domestic reforms. Strong growth and tighter enforcement of capital flow management measures (CFMs) have also helped stem external pressures. Reforms have advanced across a wide domain, including reducing overcapacity, strengthening local government borrowing frameworks, and addressing financial sector risks. However, reform progress needs to accelerate to secure medium-term stability and address the risk that the current trajectory of the economy could eventually lead to a sharp adjustment.

2. Though annual growth slowed in 2016, the momentum of the Chinese economy accelerated over the course of the year and into early 2017. GDP growth in 2016 reached 6.7 percent, down from 6.9 percent in 2015 and in line with the authorities’ target of 6.5-7 percent. However, after a largely constant deceleration in quarterly output since early 2010, underlying momentum stabilized in the second half of 2016.

  • On the demand side, consumption firmed amid a still-tight labor market and accounted for nearly two-thirds of total growth, the highest share since 2000. Investment also remained strong, supported by continued fast growth in public infrastructure and the first acceleration in real estate investment in five years.
  • On the supply side, the service sector remained the key driver, reflecting growth in the new economy (e.g. information technology) and a recovery in real estate services. But the change in momentum came from industry which stabilized after a 5-year deceleration, due in part to a sharp recovery in prices of key commodities.

    Real GDP Stabilizes While Nominal Accelerates

    (In percent, year-on-year)

    Source: Haver Analytics.

Growth then accelerated into the first quarter of 2017 when real output rose 6.9 percent (yoy), faster than any quarter in 2016. While domestic demand remained strong and real import growth reached double digits, the contribution of net exports turned positive, largely reflecting the recovering global economy.

3. Stronger domestic demand helped further reduce China’s external imbalance, though it remains moderately stronger compared to the level consistent with medium-term fundamentals. In 2016, the current account surplus fell by almost 1 percentage point to 1.7 percent of GDP. The falling surplus was driven by a sharp recovery in goods imports and continued strength in tourism outflows (though due to data limitations, tourism imports may be overstated by roughly ½ percent of GDP since 2014, reflecting misclassified capital outflows). With this fall in the surplus, and taking into account potential overstatement of tourism and other data uncertainties, staff assesses that the 2016 current account was ½-2½ percent of GDP stronger than the level implied by medium-term fundamentals and desirable policies, versus 1–3 percent of GDP in 2015. The reduction of the current account gap reflects in part the effect of fiscal and credit easing.

Current Account Surplus Near 15 Year Lows

(In percent of GDP, four quarter rolling sum)

Sources: CEIC Data Company Ltd.; and IMF staff estimates.

4. Stronger domestic demand reflects a mixture of substantial policy easing and some progress in supply-side reforms.

  • Stimulus. First, between September 2014 and December 2015, the benchmark lending rate was reduced by 165 basis points. Second, the authorities eased real-estate macro-prudential policies in 2015 and early 2016 (e.g. lower down-payment requirements and higher discounts on mortgage rates) which helped lower inventories. Third, general government net borrowing widened by 2¾ percent of GDP between 2014 and 2016, driving a similar increase in the “augmented” deficit which reached an estimated 12¼ percent.

    Residential Real Estate Inventory Ratio Falls

    (In years)

    Sources: Local Housing Administrative Bureau (Fangguanju), Wigram Capital Advisors, IMF staff estimates.

  • Supply-side reforms. The authorities initiated reforms to reduce overcapacity in the industrial sector, achieving capacity reduction targets for the coal and steel sectors, restructuring some weak state-owned enterprises (SOEs), and more tightly enforcing environmental regulations. While not complete, the measures, along with the above-mentioned stimulus, have helped reverse the deflationary trend and trigger a recovery in producer prices and industrial profits. Moreover, the supply-side reform effort extends beyond overcapacity: for example, a 2014 reform to facilitate business registration has helped increase the number of new businesses from 6,000 before the reform to 15,000 per day in 2016.

    PPI Drives Recovery in Industrial Revenues

    (In percent, year-on-year growth, 3mma)

    Sources: CEIC Data Company Ltd.; and IMF staff estimates.

5. Amid strong growth, the authorities have pivoted toward tightening measures, reflecting a greater focus on containing financial sector risks:

  • In the second half of 2016, the authorities started tightening macro-prudential measures for the real estate sector, reversing much of the previous easing (See Selected Issues Papers (SIP)).
  • In early 2017, the PBC increased its 7-day repo rate twice by 10bps and clarified that the policy stance was now neutral.

    Tier I Cities’ Real Estate Price Growth Moderates

    (In percent, year-on-year growth, 3mma)

    Sources: Local Housing Administrative Bureau for city-level prices, NBS, and IMF staff estimates.

  • Also in early 2017, the PBC extended the coverage of its “Macro-Prudential Assessment (MPA)” to off-balance sheet activity for the first time by including Wealth Management Products (WMPs).
  • The CBRC published several new documents aimed at stricter enforcement of existing regulations and reducing regulatory arbitrage across financial products.

Thus far, the key impact of these measures has been a tightening in financial conditions and a sharp fall in intra-financial sector credit. Interbank interest rates have risen sharply—since mid-2016, the three-month Shibor has risen 124 basis points, the one-year government bond yield has risen 107 basis points and AAA corporate bond yields have risen 124 basis points. Meanwhile, bank claims on non-bank financial institutions (NBFIs) and off-balance sheet WMPs have largely stopped growing on a month-to-month basis after booming in recent years. Total credit to the non-financial private sector has also started to moderate at the margin, but has thus far been relatively less affected.

Interest Rates Have Risen

(In percent)

Sources: Bloomberg LP; CEIC Data Company Ltd.; and IMF staff estimates.

Intra-financial Claims No Longer Growing

(In RMB trillions)

Sources: Haver Analytics.

6. In addition to tightening financial conditions, the authorities have taken several steps to stabilize exchange rate expectations.

  • Foreign exchange intervention. In late 2015 and early 2016, the RMB depreciated by 8 percent in effective terms (using the basket of currencies published by the China Foreign Exchange Trade System or CFETS), reversing most of the appreciation that had resulted from the previous tight link to the U.S. dollar. Then, since mid-2016, the RMB traded within a narrow range against the CFETS basket. This stability in the RMB effective rate required considerable sales of foreign exchange when the U.S. dollar was strengthening in the second half of 2016, but once that reversed and capital outflows moderated (see below), FX reserves stabilized.

    Previous Appreciation Almost Reversed

    Sources: Bloomberg LP; and IMF staff estimates.

  • Tighter enforcement of CFMs. Enforcement of existing measures was tightened substantially from late 2016.1 Together with a stronger near-term growth outlook and a weaker U.S. dollar, these steps helped reduce net capital outflows substantially.
  • Change in mechanism for setting daily fix. In May 2017, the CFETS revised the guidance to banks for calculating their quotes for the daily opening exchange rate to reduce what they viewed as “irrational” depreciation expectations that raised the risk of overshooting (Box 1).

    Capital Outflows Moderate Sharply

    (In US$ billion)

    Sources: Institute of International Finance (IIF); and IMF staff estimates.

Box 1.The RMB/USD Central Parity Fixing Mechanism

China officially maintains a managed floating exchange rate arrangement. After keeping the RMB closely linked to the U.S. dollar during the global financial crisis, the PBC in June 2010 returned to the managed floating exchange rate regime. In this regime, the RMB’s daily trading prices against the dollar were allowed to move up to ±0.5 percent from a central parity rate, the so-called “fix”, released daily by China’s Foreign Exchange Trading System (CFETS—a PBC subsidiary). To inform the fix, market-making banks were supposed to provide quotes for the RMB/USD rates based on their sense of market supply and demand in the morning, an opaque system that had allowed substantial discretion and, in practice, resulted in the central parity barely moving from day to day, even though the market rate had often closed at some distance from the previous day’s morning fix. The daily trading band was subsequently widened to ±1 percent in April 2012 and to ±2 percent in March 2014, to allow additional flexibility of the exchange rate.

In August 2015, the PBC announced a new “central parity exchange rate formation mechanism.” Under the new mechanism, banks were asked to submit quotes that took account of the closing spot rate of the previous day as well as market supply and demand. The PBC’s announcement explained that the change was designed to increase the role of market forces, in the context of the policy objective of gradually increasing exchange rate flexibility.

In December 2015, CFETS started publishing an RMB Effective Exchange Rate Index (“CFETS basket”). Subsequently, in February 2016, they formulated new guidance to banks for their daily quotes for the RMB/U.S. dollar central parity. Henceforth, the quotes would be based on the “previous closing rate plus overnight changes in a currency basket”, with the “changes in a currency basket” referring to the adjustment in the RMB/USD rate needed to offset the impact of changes in cross-rates among basket currencies.

In May 2017, CFETS adjusted its guidance to banks further. Banks were requested to include a “countercyclical adjustment factor” in their quotes with the objective of reducing “irrational” depreciation expectations and “pro-cyclical” herding behavior. The adjustment factor was not defined, with each bank calculating it using its own parameters to reflect their assessment of economic fundamentals, in addition to the previous closing rate and the RMB’s move relative to a currency basket. In the first week after the new factor was introduced, the RMB appreciated by almost 100 basis points relative to the CFETS basket and roughly 90 basis points against the U.S. dollar.

Outlook: Stronger Near-Term Prospects but Still Rising Medium-Term Tensions

7. Ahead of the fall Party Congress, the economy is well-positioned to meet the 2017 growth target amid continuing stability in the balance of payments.

  • Growth drivers. The government set the 2017 growth target at around 6½ percent but added “or higher if possible in practice” to convey an upside bias. Staff now project 6.7 percent (up from 6.2 percent in the last Article IV) reflecting the momentum from last year’s policy support and strengthening external demand. Nonetheless, over the course of the year, momentum will likely decline reflecting recent regulatory measures which have tightened financial conditions and contributed to a declining credit impulse. Indeed, high frequency indicators from Q2 suggest that activity likely peaked in Q1. These factors will likely keep core inflation broadly stable around 2 percent.

    Credit Impulse Starting to Weigh on Real Estate

    (In percent, year-on-year growth)

    1/ Credit impulse is the change in the flow of adjusted TSF over the last four quarters relative to GDP.

    Sources: Haver analytics and IMF staff estimates.

  • External. The current account surplus is projected to fall by 0.3 percentage points to 1.4 percent of GDP, due primarily to robust domestic demand and a projected 5-percent deterioration in terms of trade. However, that fall will be more than offset by a moderation in capital outflows relative to 2016 levels amid tighter enforcement of CFMs and more anchored exchange rate expectations.

8. In the baseline, staff projects that China will meet the authorities’ 2020 output target, but at the cost of further large increases in public and private debt.

  • Staff has revised up average GDP growth over 2018-20 to 6.4 percent, from 6.0 percent in the last Article IV. Given the extensive use of stimulus in recent years, there is a greater expectation that the authorities can and will maintain a sufficiently expansionary macroeconomic policy mix to meet their target of doubling 2010 real GDP by 2020.

    Upward Revision to Path of Real GDP Growth

    (In percent)

    Sources: CEIC Data Company Ltd.; and IMF staff estimates.

  • To achieve this higher path, staff now assume that the authorities will broadly maintain current levels of public investment over the medium term and not substantially consolidate the “augmented” deficit (which includes staff’s estimate of off-budget investment). Therefore, “augmented” debt no longer stabilizes over the medium term, as projected in the 2016 Article IV, instead reaching 92 percent of GDP in 2022 on a rising path.

    Slower Consolidation in Augmented Deficit

    (In percent of GDP)

    1/ Definition of augmented deficit has been expanded to include government guided funds. This has also resulted in a revision to the historical path back to 2015.

    Sources: CEIC Data Company Ltd.; and IMF staff estimates and projections.

  • Private sector credit is projected to continue increasing over the medium term, largely unchanged from the last Article IV. Thus, total non-financial sector debt reached about 235 percent of GDP in 2016 and is projected to rise further to over 290 percent of GDP by 2022.

    Non-Financial Debt Rises Further

    (In percent of GDP)

    Sources: Haver analytics and IMF staff estimates.

9. Downside risks around the baseline have increased. A key consequence of the new baseline is that it envisions China using up valuable fiscal space to support a growth path with slower rebalancing and a higher probability of a sharp adjustment. Thus, if a sharp adjustment were to materialize, China would have lower buffers with which to respond. Such a potential adjustment could be triggered by several risks, including:

  • Funding. A funding shock could come from at least two (related) pressure points. The first is the mostly short-term, “interbank” wholesale market (which includes banks’ claims on each other and on NBFIs). The second is a loss of confidence in short-term asset management products issued by NBFIs, or a run on the WMPs which fund them.

    Significant Reliance on Wholesale Borrowing 1/

    (In percent)

    1/ Wholesale borrowing includes liabilities to other banks, to non-bank financial institutions, to the PBC, and bond issuances.

    Sources: Haver analytics and IMF staff estimates.

  • Retreat from Cross-Border Integration. Should higher trade barriers be imposed by trading partners, the impact would depend on their coverage and magnitude, how exchange rates respond, and whether China retaliates. For example, an illustrative simulation in the IMF’s Global Integrated Monetary and Fiscal Model suggests that if the U.S. puts a 10-percent tariff on Chinese exports and China allowed its real exchange rate to adjust, real GDP in China would fall by about 1 percentage point in the first year. If China retaliated with similar tariffs on U.S. imports, its GDP would contract further. However, given the complexity of global trade relationships and uncertainty regarding how exchange rates would adjust, the effect could be larger and more disruptive.

    Tariff Retaliation Lowers GDP Further

    (In percent difference from baseline; impact on Chinese GDP of 10 percent tariff on Chinese imports to US)

    Source: IMF staff estimates.

  • Capital Outflows. Pressure on the exchange rate could resume because of a faster-than-expected normalization of U.S. interest rates, much weaker growth in China, or some other shock to confidence. In an extreme scenario, the pressure could lead to renewed large reserve loss and eventually a potential disruptive exchange rate depreciation. However, this risk is likely small in the short run due to the stronger enforcement of CFMs, the prominence of state-owned banks in the foreign exchange market, and ample foreign exchange reserves.

    Record Capital Outflows

    (In US$ billion, four quarter rolling sum)

    Sources: CEIC Data Company Ltd.; and IMF staff estimates.

10. Growth could also be lower than projected, but more sustainable, if the authorities de-emphasize quantitative growth targets. Staff’s baseline assumes that the authorities will do what it takes to attain the 2020 GDP target, adjusting macro and financial sector policies as necessary. However, the authorities’ public pronouncements, as well as their reactions to staff projections (see below), suggest they are putting increasing weight on the quality, rather than the quantity, of growth. While reducing nearer-term growth, such an outcome would be welcome as it would raise the longer-term sustainability of growth.

11. In a fast-reform “proactive” scenario, China could stabilize GDP growth above medium-term baseline projections if it implemented two key reforms.

  • The growth slowdown since the global financial crisis has largely been driven by slower productivity growth. To reverse this trend, greater progress on resolving weak firms and converging to the cross-country efficiency frontier is needed. Staff estimates that such efforts could increase the contribution of productivity to growth by about 1 percentage point over the long term. Such gains could allow China to reduce the currently excessive contribution of investment to GDP growth by 1 percentage point, keeping headline GDP growth broadly unchanged.

    Better Resource Allocation from Resolving Weak Firms Can Raise Growth Potential

    (In percent, per year)

    Sources: Hsieh and Klenow (2009); IMF Fiscal Monitor (2017); NBS; and IMF staff estimates.

  • A change in the composition of fiscal policy to support rebalancing toward consumption and away from investment would allow faster consumption growth financed by a drawdown in household savings rather than higher debt.

12. In the short term, such reforms to resolve weak firms and reduce investment may have a negative impact on output and employment. While the size of this impact is uncertain, staff estimates that short-term growth in a proactive reform scenario could fall to about 5½ percent absent a policy response. However, China has some fiscal space to delay consolidation and could smooth the growth impact during a transitional period. Such stimulus should be on-budget and targeted at improving the social safety net. While these reforms would improve China’s trajectory relative to the baseline, total debt (public and private) would still increase by 35 percentage points of GDP over the medium term and stabilize at a high 270 percent of GDP (compared to 293 percent in the baseline). Overall, it should be emphasized that the growth path in a proactive reform scenario is uncertain, depending on many factors, including the depth, pace and sequence of reforms, cyclical fluctuations, and external conditions. The key policy imperative is to replace precise numerical growth targets with a commitment to reforms that achieve the fastest sustainable growth path.

Increasing Downside Risks in Growth Outlook

(In percent, year-on-year growth)

Sources: CEIC Data Company Ltd.; and IMF staff estimates and projections.

13. The revised baseline scenario—with higher growth, slower rebalancing, and a tighter capital account than in the 2016 Article IV—has three primary near-term spillovers and one major medium-term global risk. First, the baseline implies ongoing support for global commodity exporters which benefit from strong public investment, real estate, and heavy industry (e.g. Brazil, Australia). Second, there are fewer global mergers and acquisitions and Chinese-financed real estate purchases because of residents’ more limited access to foreign exchange (e.g. Canada, U.K., Australia). And third, China will continue to gain upstream market share in the Asian supply chain (e.g. Japan, Korea, Malaysia,) but at a slower pace than if there were a better investment climate for the domestic private sector and FDI. But this scenario of high growth and high debt accumulation also raises the probability of a disruptive adjustment to Chinese demand which would result in a sharp contraction in imports and a contractionary impulse to the global economy. In this regard, it is in the global economy’s interest that China transitions now to a new model of economic growth, even if the transition may be bumpy.

Authorities’ Views

14. While agreeing on the growth outlook, the authorities disagreed about the associated risks. The authorities agreed that 2017 growth was likely to exceed marginally the 6.5 percent full-year target. This implied some deceleration during the course of the year and would result in inflationary pressure remaining contained and a broadly unchanged current account. For the medium term, though the authorities shared the view that their 2020 target of doubling 2010 real GDP would likely be reached, they viewed the debt build-up thus far as manageable and likely to slow further as their reforms take effect. They also explained that their “projected growth targets” were anticipatory and not binding. They underscored that reaching the desired quality of growth was a greater priority than the quantity of growth. The authorities viewed domestic concerns, such as high financial sector leverage, as manageable considering ongoing reforms and Chinese-specific strengths, such as high domestic savings. They saw the external environment as facing many uncertainties, such as an unexpected fall in global demand or a retreat from globalization.

Policies: Switching Growth Engines to Sustain Strong, Inclusive, and Green Growth

Decisive implementation of reforms could greatly improve the baseline outlook. This requires (1) switching faster from investment to consumption so growth is less reliant on debt, (2) increasing the role of market forces to improve resource allocation, and (3) further building the “soft infrastructure” of modern policy frameworks to enhance policymakers’ ability to manage the modern Chinese economy. China has buffers to boost growth if needed, but they should be used to support accelerated reforms.

A. Fueling Sustainable Growth by Improving Resource Allocation

Excess savings

15. Low consumption and high national savings translate into lower welfare for Chinese citizens and excessive investment and debt. Estimated at 46 percent of GDP, China’s national savings are 26 percentage points higher than the global average, largely due to the household sector (SIP). Structural characteristics of the Chinese economy are part of the explanation, but so are policy factors, especially a comparatively weak social safety net. Excessive savings are problematic for welfare and macroeconomic stability:

  • Allocating a low share of income to consumption reduces current welfare (e.g. via a lower standard of living).

    China is Major Outlier in Savings and Consumption

    (In percent of GDP, 155 countries)

    Sources: World Economic Outlook (WEO); and IMF staff estimates.

  • Given home bias, capital account restrictions, and high growth targets, such high savings translate into similarly large amounts of domestic investment, which are unlikely to be absorbed efficiently—the risk is that returns from such high investment do not commensurately support future growth and debt service obligations.
  • Amid such high savings, if investment were curtailed, the current account surplus would widen, worsening global imbalances, reducing China’s contribution to global demand, and undermining the multilateral trade system which has served the world, including China, well in recent decades.

16. While demographic change will gradually reduce savings over time, targeted structural reforms, particularly fiscal, can help further boost consumption and ameliorate income inequality. Given China’s rapid aging, the ratio of elderly to working age population is expected to rise from 15 percent in 2015 to 50 percent in 2050. Model simulations suggest that this aging will lower the 2030 household savings rate by about 6 percentage points of GDP. As this would be a very gradual process, still leave domestic savings at excessive levels, and do little to reduce inequality, further reforms are needed with the following priorities:

  • Make the tax system more progressive: Reduce the personal income tax (PIT) basic exemption (which results in about 80 percent of urban workers not paying the PIT) and remove imputed minimum earnings for social contributions which results in prohibitively high effective tax rates for the working poor.

    Highly Regressive Taxes on Labor Income

    (In percent of labor income, urban household, 2012)

    Note: Labor income taxes include the Individual Income Tax on labor earnings and employee social security contributions for pensions, medical and unemployment insurance.

    Source: China Household Finance Survey; and IMF staff estimates.

  • Further increase social transfers to poor households. Lower-income households in China have a savings rate of plus 20–30 percent compared to minus 20 percent in many peers. A key driver of the difference is China’s lower public transfers. A move toward international norms on social assistance would help lower excessive precautionary savings and reduce income inequality, which (while moderating recently due to government efforts) is among the highest in the world (See SIP).

    High Chinese Savings at All Incomes

    (In percent)

    Sources: Chinese Household Income Project (CHIP); and Luxembourg Income Study (LIS).

  • Continue to increase public spending on health, pensions, and education. Higher health and pension spending will increase government consumption directly and private consumption indirectly by reducing households’ need for precautionary savings. Intensified “hukou” (residency) reform and improved access to social services, particularly less-developed areas, is needed to ensure an adequate safety net for all Chinese citizens.

    Rising Social Spending But Well Below OECD

    (In percent of GDP)

    Sources: CEIC, IMF FAD Expenditure Assessment Tool and IMF staff estimates.

  • Increase SOE dividend payments to the budget. The share of aggregate SOE profits currently paid to the budget is estimated to be well below the government’s 30-percent target by 2020. Addressing this can help establish appropriate budget constraints for weaker SOEs, reduce wasteful investment, and finance needed social spending.

    Low Transfer of SOE Dividend to Fiscal Budget

    (In percent of aggregate SOEs’ profit)

    Sources: Ministry of Finance; Unirule Institute of Economics; and IMF staff estimates.

    1/ SOEs were required to contribute their profits to the fiscal budget since 2007. Data for 2007 did not cover all SOEs. The third plenum reform required the transfer 30 percent of SOE profits to the fiscal budget by 2020.

Authorities’ Views

17. The authorities shared staff’s view that national savings remained high. They noted that demographic factors would help reduce savings going forward but agreed that policy initiatives such as strengthening the social safety would help further boost consumption and promote a more sustainable composition of growth. They noted that social spending had already risen materially in recent years and further discretionary increases would need to be balanced against underlying spending pressures, such as those due to aging.

Reform of State-Owned Enterprises and Promotion of Competition

18. SOEs have been structurally less efficient than the private sector, reducing economy-wide productivity. SOE profits have fallen and are significantly lower than those of the private sector (SIP). While partly due to higher exposure to overcapacity industries, SOE productivity is also a quarter lower on average than non-SOEs when controlling for the sector, partly driven by still-significant social responsibilities and weak corporate governance. The relatively lower profitability is especially striking given that SOEs receive substantial implicit support (e.g. credit, land), which are estimated at about 3 percent of GDP, even excluding other benefits such as operating in protected markets (Lam and Schipke 2017). Another way to illustrate the relatively lower efficiency is that in the industrial sector, SOEs account for more than half of corporate debt and 40 percent of industrial assets but less than 20 percent of industrial value added.

Weak SOE Return on Equity

(Net return on total owners’ equity; in percent)

Sources: Statistical Yearbook (2015); Unirule Institute of Economics (2015); and IMF staff estimates.

1/ Based on nominal profits of industrial SOEs net of fiscal subsidies, implicit support through the use of land and natural resources, and lower implicit financing cost.

19. Though SOE reform is considered a top priority of the government and progress has been made, reforms should be accelerated. The authorities have classified SOEs into two categories (commercial and social functions) to clarify the mandate for their managers, phased out some social responsibilities, and initiated governance and ownership reforms. This is important progress, but many concerns remain:

  • After many years of downsizing, SOEs started growing after the global financial crisis, accounting for three-quarters, or 60 percentage points, of the rise in corporate debt/GDP since then, and now have assets of over 200 percent of GDP. Under the current reform plans, SOEs would extend their size further, potentially crowding out private sector development.
  • “Mixed-ownership” reforms, in which private capital is allowed to invest in government-run enterprises, and other measures to strengthen SOE governance are still at initial stages. Actual increases in private sector participation in SOEs remain limited, political influence has been institutionalized, and it is still unclear whether the mixed-ownership model is sufficient to improve efficiency. In addition, while several troubled SOEs have announced restructuring plans, the focus appears more on mergers and consolidation, rather than operational restructuring.
  • Efforts to resolve unviable debt are underway but incomplete. While 20 percent of identified “zombie” (unviable) central SOEs were reportedly resolved, SOEs continue to account for 50 percent of zombie debt outstanding, suggesting that significant further progress is necessary.
  • Despite several high-level announcements (e.g. that the number of industries that are restricted for foreign investment would be reduced from 93 to 63), there seems as yet little effective progress in exposing SOEs to greater competition, due to the limited breath of the reforms and their slow implementation.

20. While privatization may not strictly be necessary to improve resource allocation, “competitive neutrality” is.

  • SOEs. The authorities should expedite implementation of their existing reform initiatives to foster competitive neutrality. This includes moving SOEs’ social functions to the budget to allow firms to focus on commercial objectives, raising the share of SOEs classified as “commercial-competitive”; opening additional protected sectors to greater competition from private and foreign investment; and faster restructuring of SOEs’ underperforming debt. The SOE reform agenda should also be broadened to include hardening budget constraints on SOEs by phasing out implicit subsidies on factor inputs and forcing non-viable firms to default and exit if market forces warrant, with fiscal support for the affected workers. As competitive neutrality under state ownership may prove elusive in practice, these reforms would usefully be complemented by transferring more state-owned assets to private ownership.
  • Investment climate and trade. Greater opportunities and a level playing field for the private sector, including foreign firms, would increase growth by promoting competition and attracting technology from abroad. There are two key priorities. The first is reducing barriers to entry. The OECD’s Product Market Regulations Index found services (financial, IT, transportation, logistics) highly closed relative to the OECD and other emerging markets—SOEs are particularly dominant in services. Second, a greater focus on giving the private sector (including foreign firms) equal access to resources (land, natural resources, credit and government subsidies) and ensuring equal treatment in regulations, taxation, government procurement and administrative approvals. Both can be fostered by following through on the government’s announced plans to expand private and foreign market access and local governments’ fair-competition review. China also has scope to gain from trade by reducing trade barriers, including tariffs which are still considerably higher than those of its main trading partners. Progress in these areas could help reduce trade tensions, foster positive outcomes (greater opening up and more trade), rather than negative ones (higher barriers and less trade) and encourage greater openness amongst trading partners by demonstrating China’s commitment to an open rules-based trading system.

    Product Market More Closed than Most OECD and EM Peers

    (OECD’s product market regulation (PMR) index)

    Note: The PMR index assesses degree to which policies promote competition in the product market. Policies assessed include state control of business enterprises; legal and administrative barriers to entrepreneurship; and barriers to international trade and investment.

    Sources: OECD Product Market Regulation Statistics; and IMF staff estimates.

Authorities’ Views

21. The authorities argued that SOE productivity continued to improve as a result of the supply-side and mixed-ownership reforms. They also noted that SOEs’ commercial performance was comparable with that of private firms, particularly if one accounted for their significant social responsibilities. Citing significant progress in phasing out overcapacity, deleveraging, and SOE governance, they noted that SOE profits had risen 10 percent year-over-year while leverage ratios had declined. They also disagreed with staff’s assessment that SOEs received material state support, in particular, with staff’s estimate of 3 percent of GDP: SOEs were independent market entities and thus were treated the same as other market entities. To the extent that banks chose to extend preferential credit access to SOEs, this reflected the creditworthiness of these borrowers. Looking forward, the authorities emphasized that they had completed their classification of central SOEs into commercial or social functions and had accelerated mixed-ownership reforms which allowed SOEs to benefit from both public and private perspectives. In addition, they argued that institutionalizing the Communist Party’s leadership within SOEs would help increase efficiency.

22. The authorities underscored their commitment to opening up to the private and foreign sectors. The authorities emphasized the importance of recent announcements on reducing barriers to entry for private investment (e.g. in oil refining, electricity, natural gas, telecommunications, civil aviation, and certain financial services such as credit ratings), reducing the size of the negative list restricting foreign investment, and improving the business environment for foreign firms. They stressed their commitment to promoting global trade and investment liberalization, opposing all forms of protectionism, and seeking “win-win” cooperative solutions to trade tensions (noting recent progress in the dialogue with the U.S.).

Reduction in Overcapacity
ActualTarget
2013201420152016201620172018-20
Crude steel
Capacity (in millions of tonnage) 1/1,0821,1601,2001,1441,1551,1041,097
Employment (millions) 2/4.94.74.23.83.93.52.9
Coal
Capacity (in millions of tonnage) 1/5,7005,4105,4505,3705,160
Employment (millions) 2/5.34.94.44.04.23.93.9
Sources: CEIC Data Company Ltd.; WIND; and IMF staff estimates.

The official targets announced were net changes relative to 2015 levels. These targets are staff estimates of the capacity target levels that those announcements implied.

The authorities intend to reduce 0.5 million employment in 2017 (or a total of 1.8 million workers over the medium term) for coal and steel sectors. Here assumes the reduction target for 2016 is one-fifth of the total target based on 2015 employment levels. Beyond 2016 are forecast numbers based on recent reports not targets.

Sources: CEIC Data Company Ltd.; WIND; and IMF staff estimates.

The official targets announced were net changes relative to 2015 levels. These targets are staff estimates of the capacity target levels that those announcements implied.

The authorities intend to reduce 0.5 million employment in 2017 (or a total of 1.8 million workers over the medium term) for coal and steel sectors. Here assumes the reduction target for 2016 is one-fifth of the total target based on 2015 employment levels. Beyond 2016 are forecast numbers based on recent reports not targets.

Overcapacity

23. While related to the role of SOEs, the overcapacity challenge straddles the public and private sectors and has domestic and international dimensions. Defined by low capacity utilization rates and a large share of firms incurring losses, overcapacity in China encompasses at least ten sectors including coal, steel, cement, plated glass, aluminum, chemicals, paper, solar power, ship building, and coal-fueled power. China’s protracted excess capacity has contributed to downward pressure on global prices, rising market share for Chinese firms, and tensions with key trading partners. But overcapacity is also damaging for China by weighing on medium-term growth, the environment, and financial stability.

Overcapacity: Deteriorating Utilization and Large Global Share

(In percent of total legal designated capacity (left-scale) and global output (right-scale))

Sources: European Chamber of Commerce, Goldman Sachs, and IMF staff estimates.

1/ Coal capacity exceed 100 percent because coal mines produced over legal designed capacity level in earlier years. Coal capacity utilization in 2016 fell because coal production fell sharper than the cut in coal capacity relative to 2015.

24. Initial progress has been made in reducing overcapacity. The authorities outlined capacity and employment reduction targets for the coal and steel sectors for 2016-2020. They clarified that the targets were in net terms so any new capacity, aimed at improving efficiency, would require deeper cuts to existing facilities. According to official data, the capacity targets were over-achieved in 2016 and are on track to be met in 2017, in part due to tighter enforcement of environmental and regulatory standards. Total employment in the two sectors is also 25-30 percent (around 2 million) below 2013 levels. Key concerns are that, per some accounts, capacity reduction included the closure of already-idle plants, and there is still limited restructuring of overcapacity firms’ outstanding debt.

25. Going forward, the reform effort should be broadened and deepened with greater reliance on market forces. Efforts should focus on more ambitious net targets within coal and steel and a broadening of targets to other sectors. Enforcing targets should avoid excessive reliance on administrative measures such as cuts in work days, mergers, and window guidance on prices. Reform efforts also need to address the origin of overcapacity which is a combination of high GDP growth targets met through state-directed investment spending, soft budget constraints which allow loss-making firms to stay in business, and underpricing of long-term environmental damage. Indeed, until market forces more fully drive resource allocation, new overcapacity sectors are likely to emerge, possibly including in the high-technology sectors that the government is now making a priority in its industrial policy. In this regard, China’s participation in the OECD Global Forum on Steel Excess Capacity is welcome and staff encourages further efforts to strengthen multilateral cooperation in reducing overcapacity.

Authorities’ Views

26. The authorities felt that staff did not sufficiently appreciate the progress in reducing overcapacity in steel and coal. Due in large part to the government’s supply-side reforms, there had been substantial improvements in key operational indicators in these industries (capacity utilization, prices, firm profitability), which had in turn contributed to a reduction in global overcapacity. The authorities explained that this progress stemmed largely from optimizing corporate structure, upgrading product quality, strict enforcement of environmental regulations, a crackdown on illegal construction, and forceful elimination of inefficient production techniques. This progress had also provided useful experience for extending efforts to reduce over-capacity in other sectors.

B. Tackling Financial Stability Risks

27. China now has one of the largest banking sectors in the world. At 310 percent of GDP, China’s banking sector is above the advanced economy average and nearly three times the emerging market average. The sharp growth in recent years reflects both a rise in credit to the real economy and intra-financial sector claims. The increase in size, complexity and interconnectedness of these exposures have resulted in sharply rising risks.

Banking Sector Among Largest in the World

(In percent of GDP, 2015)

Note: Non-standardized report forms (non-SRFs) for Argentina, China, India, Saudi Arabia, and United Kingdom. All other countries report in SRF.

Sources: Bank for International Settlements (BIS); and IMF staff estimates.

Nonfinancial private sector debt

28. The recent growth in non-financial sector debt raises concerns for medium-term macroeconomic stability. Lending to the private sector rose 16 percent in 2016, twice nominal GDP growth, pushing the credit gap to about 25 percent of GDP. Since 2008, private sector debt relative to GDP has risen by 80 percentage points to about 175 percent—such large increases have internationally been associated with sharp growth slowdowns and often financial crises. Staff estimates that, had credit growth been kept to a sustainable rate, ceteris paribus, real GDP growth would have been around 5½ percent between 2012 and 2016, rather than 7¼ percent (SIP). As recommended by staff in previous Article IV consultations, the growth-subtracting effect of credit restraint can be mitigated by pro-rebalancing, on-budget fiscal stimulus, and in the medium term, by productivity-enhancing structural reforms—which would ensure growth is both strong and sustainable.

China’s Large Credit Gap

(In percent of GDP)

Note: Based on credit to private non-financial sector.

Sources: Bank for International Settlements (BIS); and IMF staff estimates.

29. In the past year, the authorities have started to take important initial steps to facilitate private sector deleveraging. For example, guidelines were issued to broaden the number of tools that firms could use to restructure their debt. In addition, as explained in ¶5, a range of prudential and administrative measures were introduced to contain financial sector risks.

30. Given strong growth momentum, now is the time to intensify these deleveraging efforts.

  • Stock. The focus should be on greater recognition of the underlying stock of bad assets. This requires a reduction in implicit subsidies, especially for SOEs, and more decisive action by supervisors. The immediate focus could be the debt of zombie firms, overcapacity companies, and underperforming SOEs.

    Nonviable Zombie Firms are Rising Again 1/

    (In percent of total industrial firms, weighted by number of firms and total liabilities)

    Sources: NBS Industrial Firm Survey and staff estimates.

    1/ Data for 2010 are missing and based on average of 2009 and 2011. Estimates are average between two definitions of zombies (State Council and Fukurama and Nakamura (2001).

  • Flow. Ultimately, private sector deleveraging will require credit to grow more slowly than GDP. To achieve this, the overarching priority remains focusing more on the quality and sustainability of growth and less on quantitative targets. To reduce the drag on growth from slowing credit expansion, new credit needs to become more efficient. On the credit demand side, this includes imposing hard budget constraints on SOEs and macro-prudential measures to contain mortgages and broader risks to the real estate sector. On the credit supply side, micro-and macro-prudential regulations should be tightened further, by requiring additional buffers for the financial system and continuing recent efforts to eliminate regulatory arbitrage.

The Size and Complexity of Intra-Financial Sector Credit

31. The large stock of intra-financial sector credit continues to raise important risks for financial stability. Intra-financial sector credit is a way for institutions to boost leverage and profits while avoiding regulatory hurdles such as capital and provisioning requirements on bank loans. The key channels are (1) large banks lending to small banks and NBFIs in the largely short-term and collateralized wholesale market, (2) banks or NBFIs investing in other banks’ WMPs and negotiable certificates of deposit (NCDs), and (3) banks and NBFIs investing in shadow products issued by other NBFIs. These exposures raise several concerns:

Large Banks Fund Small Banks, NBFIs in Wholesale Market

(In RMB trillion)

Sources: The People’s Bank of China.

  • Capital. When banks purchase investment products, the risk-weighting may be lower than that of regular loans even when the underlying assets are loan-like. In addition, losses on wealth management products may ultimately be absorbed by the issuing bank given the perceived guarantee.
  • Liquidity. Even though intra-financial sector credit is often short-term and collateralized, these products may prove less easy to redeem in practice, NBFIs do not have access to PBC funding, and the price of the underlying assets could gap lower in the event of large-scale liquidity stress, amplifying the shock.
  • Complexity. The numerous stages of leverage make “seeing-through” to the underlying asset more difficult for banks, regulators and investors.

32. Recent regulatory/supervisory tightening is thus critically important and should continue, even if it means some financial tension and slower growth. This tightening is welcome but, given the size of accumulated imbalances, it is bound to create difficulties, and defaults of overexposed firms and to dampen activity in related sectors. To prevent such tension from derailing the supervisory tightening and leading to another “stop-go” cycle, regulators should: (1) coordinate effectively to ensure that financial conditions do not tighten excessively, (2) ensure that all solvent banks have equal access to the PBC’s standing lending facilities provided they have the required collateral, and (3) allow insolvent financial institutions to exit.

33. The ongoing Financial Sector Assessment Program (FSAP) will provide staff’s comprehensive assessment of the financial sector framework, risks, and recommendations. While this Staff Report has been coordinated with the FSAP progress to date, the FSAP work is ongoing, with Executive Board consideration tentatively envisaged around November, and thus the above analysis and recommendations should be considered preliminary.

Authorities’ Views

34. The authorities disagreed that the stock of bad assets was underestimated. They argued that weak firms were already facing significantly harder budget constraints due to pressure on financial institutions to reduce financing to overcapacity and real estate sectors. More broadly, while the authorities underscored that deleveraging remained a top priority, they considered a more gradual pace of deleveraging than that recommended by staff as desirable and sustainable. They pointed out that lending to firms was already slowing and would stabilize as a share of GDP over the medium term.

35. The authorities recognized that the growth in the size and complexity of the financial sector raised risks but argued the problem was manageable. They pointed to the recent intensification of supervision and resulting slowdown in financial sector leverage as evidence of their commitment. The authorities recognized that the increase in market interest rates could create some tension, but underscored that they had recently stepped up efforts in policy coordination and the PBC would maintain sufficient liquidity to avoid disorderly adjustment and systemic risks. They emphasized that all banks had access to the PBC’s standing facility against required collateral, but disagreed with staff’s recommendation not to use penalty rates for banks with low MPA ratings. The authorities looked forward to the finalized FSAP recommendations.

C. A More Sustainable Macro-Policy Mix Should Play a Key Role in the Transition

Fiscal Stance

36. China has some fiscal space, but the extent depends on whether liabilities from off-budget investments are considered. China’s general government net borrowing and debt were 3¾ and 37 percent of GDP, respectively, in 2016. Given that the primary balance is projected to be broadly stable and China has a favorable growth-interest rate differential, official government debt is projected to rise gradually and stabilize over the long run. However, including debt of Local Government Financing Vehicles (LFGVs) and entities such as government-guided funds, “augmented” debt is estimated at 62 percent of GDP in 2016 and projected to rise to 92 percent in 2022. The argument for using the “augmented” concept is that these obligations financed spending that appeared to be mostly non-market based with uncertain returns and by entities that are largely government-controlled; it thus likely provides a more accurate estimate of the fiscal impulse and potential debt burden on public finances. Indeed, in 2014, two-thirds of LGFV debt (22 percent of GDP) was recognized as government obligations. On the other hand, given recent reforms, especially the 2014 Budget law and subsequent regulations that legally removed government responsibility for such debt, there is now somewhat larger uncertainty regarding the degree to which “augmented” debt will eventually become a burden on public finances (SIP). Further analysis of firm-level data is needed to assess the extent to which LGFVs operate on a fully commercial basis with sound earnings and debt outlook, or otherwise.

Rising Augmented Debt

(In percent of GDP)

Source: IMF staff estimates.

1/ Data through 2015, 2016 estimated, 2017 projection. Large jump in 2014 reflects official recognition of 22 percent of GDP in LGFV debt.

2/ Government guided funds (GGF) and special construction funds (SCF). Social capital portion only.

37. Fiscal policy going forward should aim to support rebalancing and ease the transition to a new growth model. China has some fiscal space given mitigating factors such as the relatively captive market provided by large national savings and the substantial stock of government assets. As a result, while consolidation is important, improving the composition of fiscal policy is more urgent. The preferred approach would involve:

  • Fiscal reform for faster rebalancing. The composition of fiscal policy should favor rebalancing by increasing spending on health, education, and social security and reducing infrastructure investment. Revenues should also be re-calibrated away from land sales and social security contributions in favor of higher personal, property, and environmental taxes as well as higher SOE dividends. Substantially raising taxes on fossil fuel and pollution (e.g., a carbon/coal tax) would not only raise revenue, but also curtail emissions, improve energy efficiency, and prevent almost 4 million premature deaths by 2030 (WP/16/148).
  • Gradual “augmented”-deficit reduction. The “augmented” deficit should gradually fall to its debt stabilizing level. A tightening of some ½ percent of GDP per year could balance the need to achieve sustainability while limiting the drag on growth. On-budget deficits can be maintained and even expanded if combined with faster off-budget consolidation.
  • Easing costs of transition in the short to medium term. If an accelerated reform agenda were to weigh on growth and risk an excessively sharp adjustment, China would still have some fiscal space which could be used to limit the transition cost, ideally via an enhanced social safety net for affected workers. Thus, if necessary, consolidation could be slowed or even temporarily reversed.

Authorities’ Views

38. The authorities continued to disagree with the “augmented” debt and deficit concepts used by staff. They argued that the 2014 budget law and subsequent regulations had clarified that Local Government Financing Vehicles (LGFVs) were standard firms and thus new borrowing was not part of the government sector. All obligations of local governments had now been explicitly recognized within official public debt statistics and local governments would not assume any legal repayment responsibility going forward for financing vehicles, government guided funds, or special construction funds. They argued that many LGFV projects were commercial and backed by real assets. The authorities also argued that staff should only use the general budget deficit and official debt numbers of roughly 3 and 37 percent of GDP (as of end-2016), respectively.

Monetary Stance

39. A gradual removal of monetary policy accommodation would be justified if core inflation continued to tick up. The current stance of monetary policy is still accommodative. At 4.35 percent, the benchmark lending rate is at its historical low and, at 2.45 percent, the PBC’s 7-day repo rate (a key policy instrument for the new interest rate corridor) is barely positive in real terms. In addition, GDP and credit growth are strong, the labor market is robust, and core inflation has risen from 1½ to 2 percent in the last year. Indeed, current rates are well below Taylor Rule-suggested levels. At the same time, recent regulatory and supervisory measures have significantly tightened financial conditions and, coupled with other factors, may contain core inflation pressures. Thus, while the PBC should expect to increase gradually the 7-day repo rate, the pace should be data dependent with a focus on near-term developments in core inflation and activity. Moderately higher interest rates could also help reduce excessive leverage and limit pressure on the exchange rate, but these considerations are secondary as interest rates should not be the primary tool for tackling financial stability concerns or stabilizing the exchange rate (SIP).

Current Rates Below Taylor Rule Suggested Levels

(In percent)

Note: The Taylor rule assumes an inflation target of 3% and a neutral benchmark lending rate of 6%.

Sources: CEIC Data Company Ltd.; and IMF staff estimates.

Authorities’ Views

40. The authorities disagreed with staff that the monetary stance was accommodative. After the recent adjustments in banking system liquidity and interest rates, they saw the current stance as neutral (they agreed that the stance had been accommodative in recent years) and preferred to have no bias regarding future interest rate moves. Though headline inflation had fallen sharply in recent months due to food prices, the authorities saw underlying inflation dynamics as broadly consistent with their desired range of 2-3 percent. They also viewed the current stance as appropriate for their exchange rate and financial sector stability objectives.

Exchange Rate Management

41. The authorities have taken several steps since 2015 that make China better prepared to increase exchange rate flexibility, even in the near term.

  • Established reference to a basket. Though the CFETS basket has not fully replaced the dollar as a reference, the authorities have made progress in reducing the dollar link and increasing market focus on the nominal effective exchange rate.
  • Reduced Intervention. Tighter enforcement of CFMs, a better growth outlook, and a weaker U.S. dollar have helped reduce FX intervention and the near-term risk of large outflows.
  • Exchange rate broadly in line. Staff assesses the real effective exchange rate to be broadly consistent with fundamentals, unchanged from the 2015 assessment.
  • Containing depreciation expectations. The authorities have helped stabilize expectations by explicitly countering, both in words and actions, market views that they were seeking depreciation as a way to boost growth and competitiveness.

    Still Strong Export Market Share

    (In percent)

    Sources: Haver Analytics; and IMF staff estimates.

42. Nevertheless, administrative control over FX flows and the exchange rate has increased over the past year, and progress toward a more market-determined, flexible, exchange rate should resume. Besides the tighter enforcement of CFMs, the recently introduced “counter-cyclical” factor to guide banks in their fixing quotes reduces the role of market forces and FX intervention in day-to-day exchange rate management in favor of more administrative control. While this may still leave room for significant flexibility over time, the change could have been better explained and, as many market participants/observers perceived, it appears a step back toward a closer link to the U.S. dollar. It is also unlikely to reduce durably the need for foreign exchange intervention if underlying pressures are not addressed. Thus, there remains considerable scope to further improve exchange rate policy as well as the communication of it.

43. Against this background, the framework suggested in last year’s Article IV remains appropriate. The authorities should maintain an implicit and widening band around an equilibrium effective rate within which the spot rate could fluctuate with market forces. If the equilibrium rate is assessed to have changed, the center of the band could be adjusted. Such a framework would be implemented by foreign exchange intervention and public communication.

44. China’s foreign currency reserves, at US$ 3 trillion, are more than adequate to allow a continued gradual move to a floating exchange rate. There are two key considerations when applying the IMF’s composite metric for assessing reserve adequacy to China.

  • First, given that China has a neither fully open nor fully closed capital account and an exchange rate which is closer to, but not fully, fixed, it is not obvious which weights to apply. The US$ 1 trillion threshold implied by the composite metric for a floating exchange rate with capital controls is too low and the US$ 2.9 trillion threshold for a fixed but open regime is too high. On balance, the appropriate precautionary level lies in between.

    Reserves Well Above Relevant Adequacy Metrics

    (In US$ billion)

    Source: IMF staff estimates.

  • Second, the composite metric is designed to guide the appropriate level of reserves to acquire in a normal period so they can serve as a buffer if a capital account crisis materializes. It is not meant to be a minimum to maintain at all stages of capital outflow pressure.

45. Further capital account opening, while desirable over the medium term, should be carefully sequenced. The recent tightening in CFMs was broadly consistent with the IMF’s institutional view on capital flows: the necessary supporting reforms (effective monetary policy framework, sound financial system, and exchange rate flexibility) had not kept pace with the de facto liberalization of capital flows (SIP). Nonetheless, there are several concerns to consider. First, tighter enforcement of CFMs is weighing on the business climate, in part because of the less-than-transparent way in which it has been implemented and its uneven enforcement over time and across provinces. Second, there is a risk that the use of CFMs allows for a delay in reforms and increases the risk of domestic asset bubbles. And third, the effectiveness of CFMs is likely to erode as investors find loopholes and portfolio flows increase. With these considerations in mind, there are two key priorities going forward:

  • Accelerated progress on the necessary supporting reforms is needed to support ongoing liberalization. In the near term, only carefully targeted liberalization—e.g. more FDI in services or reducing the reserve requirement for onshore hedging—should be considered.
  • Existing CFMs should be consistently and transparently enforced. They should also not restrict current international payments and transfers, in line with China’s IMF obligations.

46. The “Belt and Road” Initiative (BRI) could foster multinational cooperation in trade, investment and finance, bringing much needed infrastructure and connectivity to the region. Fully reaping the benefits of this initiative will require strong governance of projects to make sure that they are financially viable and that the recipient countries have sufficient institutional and macroeconomic capacity to manage them.

Authorities’ Views

47. The authorities believed that capital flow pressures had become more balanced since early 2017. However, they continued to see some irrational and self-reinforcing dynamics in market forces. For example, they saw the RMB’s stability against the dollar in the first quarter of 2017—a period of depreciation in the global dollar index, little official foreign exchange intervention, and improving fundamentals in China - as evidence of persistent irrational market behavior; in their view, the RMB should have strengthened during that period. To break these expectations, they considered policy steps to generate two-way expectations as appropriate, particularly before a period of dollar strength resumes. Such steps could at times be through intervention (selling FX to move the exchange rate against market expectations), and they were confident that the current level of FX reserves was adequate. Regarding the daily fixing, they explained that the previous mechanism had at times led to self-fulfilling market expectations with market pressures one day influencing directly the following day’s morning fix (which carries a policy flavor). Given the irrational and self-reinforcing pressures in the foreign exchange market, the authorities introduced a “counter-cyclical factor” to reflect better macroeconomic fundamentals and underlying market forces.

48. The authorities noted that the tighter enforcement of existing regulations had helped reduce excessive capital outflows. They underscored that legitimate transactions should not be delayed and intended to increase their administrative capacity to ensure smooth processing of these transactions in the future. They stressed that they remained committed to not restricting current international payments and transfers, in line with their obligations to the IMF. In addition, they remained committed to further gradual capital account liberalization and did not think that progress in key reform areas was insufficient for further opening of the capital account. Rather, they argued that such opening could support the reforms. They saw the BRI as an open and inclusive regional economic cooperation framework that benefited all countries involved.

D. Modernizing Policy Frameworks2

Fiscal Framework

49. Ongoing fiscal framework reforms should be deepened to support consumption, reduce inequality, and ensure medium-term debt sustainability. Important steps have been taken with formulating the new Guidelines for Local Government Debt and the blueprint for reforming inter-governmental fiscal relations (SIP). As these reforms move forward, a number of issues should be considered:

  • Targeted Amount of Revenues and Spending. The authorities should consider whether the level and structure of tax and spending is adequate for China’s development goals. China’s tax revenue is relatively low at 20 percent of GDP (versus OECD average of 34 percent) due primarily to low PIT and property receipts. In addition, the tax structure is regressive and China lags major emerging markets in public spending on education, health and other social assistance. This is reflected in one of the highest levels of income inequality in the world, which, while easing slightly recently, is projected to increase without policy action.
  • Allocation of responsibilities. Tax and spending responsibilities should be assigned to the level of government that can most efficiently execute them. For example, social insurance functions (e.g. pensions and employment insurance) should be centralized to take advantage of economies of scale, remove barriers to mobility and ensure benefits are equalized across regions. Currently, local governments in China have the highest share of national spending responsibility in the world yet very limited revenue autonomy. Introducing a recurrent property tax with rates set by local governments within a centrally-approved band is one option to help address this imbalance.

    Highly Decentralized Spending Responsibilities

    (In percent)

    Sources: CEIC Data Company Ltd.; Government Finance Statistics; and IMF staff estimates.

  • Intra-governmental transfers. Even after such reallocation of responsibilities, some “vertical imbalance” will likely remain. This imbalance should be reduced via central government transfers (i.e. non-revenue sharing) which are rules-based to improve predictability and reduce the pro-cyclicality of local government funding. Fiscal disparities across regions could be reduced further by increasing the size of equalization grants. Targeted transfers should be rationalized and simplified with stronger emphasis on outputs/quality of services rather than inputs.
  • Financing. A key advantage of this framework is that provided the debt authorization is adequate, it reduces pressure for off-budget borrowing or reliance on land sales. Together with developing comprehensive medium-term budgets and stronger coordination with the National Development and Reform Commission (NDRC) on investment projects, this framework would also support the Ministry of Finance’s recent intensive focus on eliminating potential government liabilities from LGFV borrowing, currently included in staff’s “augmented” debt/deficit concepts—staff will continue to review the size, coverage and appropriateness of this concept as these efforts gain traction.

Authorities’ Views

50. The authorities broadly shared staff’s view that direct taxes (including on personal income and property) should increase but argued that the process should be gradual and consistent with the broader tax reform strategy. They also agreed that more centralization of social insurance spending could be appropriate to take advantage of economies of scale. Key areas of difference with staff focused on whether the new budget law and recent steps to remove further government guarantees, increased reliance on PPPs and existing local government borrowing quotas were adequate to prevent further off-budget fiscal spending. Unlike staff, the authorities did not see a need for further increasing pollution tax rates or expanding the base, notably by adopting a carbon tax. The authorities also favored tax sharing as a means of financing spending of local governments, as opposed to given them more revenue autonomy through a property tax and personal income tax surcharge.

Monetary Policy Framework

51. China’s transition to a more market-based economy requires continued progress in modernizing the monetary policy framework:

  • Objective. The PBC’s mandate includes multiple goals, including stabilizing the level of the exchange rate and domestic prices while supporting growth and financial sector soundness. However, monetary instruments are unable to deliver on multiple objectives (if inconsistent) and are limited in their ability to affect real variables directly (e.g. output) in the long term. Thus, monetary policy effectiveness would be considerably improved if price stability was formally identified as its primary objective. In addition, the PBC should have clear accountability around the target (such as a medium-term inflation target set by the government) and the necessary operational independence to achieve it.
  • Key policy instruments. The conduct of monetary policy increasingly resembles a standard interest rate-based framework, based on the 7-day interbank reverse repo rate. This would be strengthened by (1) formally acknowledging this framework, (2) dropping monetary aggregate targets and the publication of benchmark lending rates, (3) gradually reducing the distortionary high reserve requirements (offset as needed by open-market operations), (4) basing pricing and access to the PBC’s lending facilities on clearly defined collateral rules and not supervisory criteria, and (5) aligning lending instruments more closely to the PBC’s monetary policy objectives.

    PBC Raises Short-Term Interest Rates

    Sources: Bloomberg LP; CEIC Data Company Ltd.; and IMF staff estimates.

  • Communication. PBC communications have improved, but considerable scope remains to (1) clarify the objectives of monetary policy and how the policy instruments relate to those objectives, (2) communicate transparently with financial institutions, and (3) publish policy communications simultaneously in English, given the growing global importance of Chinese financial markets.

Authorities’ Views

52. The PBC articulated that price stability was the primary, but not the only, objective of monetary policy. The authorities explained that, per the PBC Law, the aim of monetary policy was to maintain the stability of the value of the currency and thereby promote economic growth. As China remained an economy in transition, this meant that price stability had the highest weight when making monetary policy decisions, but that other objectives had to be considered as well including employment, balance of payments, and financial sector stability. Regarding the implementation of monetary policy, the authorities agreed that they had made progress in moving to a market-based system where prices and interest rates played an increasingly important role relative to quantities. However, they felt it premature to drop monetary aggregate targets, did not consider current levels of reserve requirements as distortionary and feared that lowering them would inadvertently signal a looser monetary policy. They also thought it premature to refer explicitly to the 7-day repo as the policy rate and saw a continued need for benchmark rates to guide market pricing.

Data Frameworks

53. While some progress has been made, major data gaps remain, undermining policy making and credibility, IMF surveillance, and G20 commitments.

  • Coverage. Key data are still missing, including a full breakdown of real GDP by expenditure (e.g. quarterly levels of consumption, investment, exports, and imports) and by supply (e.g. a decomposition of “other services” and “industry”).
  • Integrity. Significant effort has been made to improve data integrity, including the newly introduced regulation on implementing the Statistics Law. Continued follow-through is critical.
  • Quasi-fiscal. Closer monitoring of all types of public investment (e.g. LGFVs, PPPs, government-guided funds) is necessary both to avoid potential sudden increases in public debt, as in 2014, and to assess the impact of public policy on aggregate demand.
  • Communication. More detailed communication alongside data releases would improve transparency and credibility.

Authorities’ Views

54. The authorities agreed with the need to broaden their publication of macroeconomic data. However, with respect to the specific gaps noted by staff, they considered that further technical work was needed and did not see imminent publication as realistic. The authorities argued that significantly increased monitoring of new debt incurred by LGFVs was not necessary given that recent reforms had largely eliminated the risk that such debt could migrate to the government balance sheet. They also emphasized several steps taken to improve data integrity, including more frequent provincial inspection visits and greater penalties for falsification.

Staff Appraisal

55. China continues to transition to a more sustainable growth path and reforms have advanced across a wide domain. Important supervisory and regulatory action has been taken to contain financial sector risks, corporate debt is growing more slowly, and local government borrowing frameworks are being improved.

56. The near-term growth outlook has firmed but at the cost of higher medium-term risks. Policy support, recovering external demand, and reform progress have helped keep growth strong. Amid strong momentum and an expectation that the authorities will do what it takes to achieve their medium-term growth target, staff have increased their medium-term baseline growth projections. However, risks around this baseline have also increased. The main cost of this stronger growth outlook is further large increases in public and private debt. Such large increases have internationally been associated with sharp growth slowdowns and often financial crises. Staff thus recommends replacing precise numerical growth targets with a commitment to reforms that deliver the fastest sustainable growth path.

57. China has the potential to sustain safely strong growth over the medium term. But as has been widely recognized, including in the government’s reform plans, this requires deep reforms to transition from the current growth model that relies on credit-fed investment and debt. It is critical to accelerate such reforms now while growth is strong and buffers sufficient to ease the transition.

58. China needs to boost consumption further to ensure sustainable and inclusive growth. Continued increases in public spending on health, pensions, education, and transfers to poor households would reduce excessive precautionary savings and, combined with making the tax system more progressive and greener, boost growth while further reducing China’s high income inequality and pollution.

59. To increase the role of market forces, the existing reform agenda for SOEs should be accelerated and broadened, and trade and investment further liberalized. SOEs should face harder budget constraints by having their implicit state support removed and by being forced to default and exit if market forces warrant. Building on recent announcements, barriers to entry should be removed, especially in the highly closed service sector. Efforts to reduce overcapacity should have more ambitious targets with greater reliance on market forces and more attention to underperforming debt. Gains from trade should also be increased by reducing trade barriers and achieving “win-win” cooperative solutions to trade tensions Such actions could also encourage greater openness amongst trading partners by demonstrating China’s commitment to an open, rules-based trading system.

60. A more sustainable macro-policy mix should include focusing more on the quality and sustainability of growth and less on quantitative targets. The fiscal stance should be gradually tightened and monetary policy accommodation reduced. To reduce nonfinancial sector debt, the focus should be on greater recognition of losses, especially of underperforming SOEs and zombie enterprises. Reducing the flow of new debt and increasing its efficiency require cutting off-budget public investment and imposing hard budget constraints on SOEs.

61. The critically important recent focus on tackling financial sector risks should continue, even if it entails some financial tensions and slower growth. The ongoing FSAP will provide staff’s comprehensive assessment of the financial sector framework, risks and recommendations.

62. The monetary policy framework should continue to be strengthened. This should include phasing out monetary targets, resuming progress towards a flexible exchange rate, and improving communications. While China’s external position remains moderately stronger compared to the level consistent with medium-term fundamentals, the renminbi is assessed as broadly in line with fundamentals. CFMs should be applied transparently and consistently. Further capital account liberalization should be carefully sequenced with the necessary supporting reforms, including an effective monetary policy framework, sound financial system, and exchange rate flexibility.

63. The government’s guidelines on reforming central-local fiscal relations are welcome. Some expenditure responsibilities, such as social insurance, should be centralized, while local governments should be given more revenue-raising authority, as well as sufficient debt quotas to reduce their incentive to rely on off-budget borrowing and land sales.

64. China also needs to address remaining data gaps to further improve policy making and meet G20 commitments.

65. It is proposed that the next Article IV consultation with China take place on the standard 12-month cycle.

Figure 1.Activity: High-Frequency Indicators Remain Firm

Sources: CEIC Data Company Ltd.; Haver Analytics; and IMF staff estimates.

Figure 2.Rebalancing: Continued Gradual Progress

Sources: CEIC Data Company Ltd.; IEA; Haver Analytics; and IMF staff estimates.

Note: Green indicates substantial progress, yellow indicates some progress, and red means lack of progress. For more details on the color coding, see IMF working paper 16/183.

1/ IMF staff estimates.

2/ Based on flow of funds data, available up to 2014.

Figure 3.Fiscal: Continued Loosening

Sources: CBONDS; CEIC Data Company Ltd.; IMF FAD Expenditure Assessment Tool; IFS; HKMA; Haver Analytics; WIND; and IMF staff estimates.

Figure 4.Monetary: Rising Price Pressure

Sources: Bloomberg; CEIC Data Company Ltd.; Haver Analytics; and IMF staff estimates.

Figure 5.External: Outflows Begin to Ease

Sources: Bloomberg; CEIC Data Company Ltd.; HKMA; SAFE; and IMF staff estimates.

Figure 6.Banking: Ongoing Rapid Expansion

Sources: CEIC Data Company Ltd.; HKMA; Haver Analytics; WIND; IMF Global Financial Stability Report (GFSR); and IMF staff estimates.

Figure 7.Credit: Rising Household Borrowing

Sources: BIS; CEIC Data Company Ltd.; WIND database; Authorities’ websites; and IMF staff estimates.

Figure 8.Financial Markets: Bond Yields Rise

Sources: Bloomberg; CEIC Data Company Ltd.; TMA; WIND database; Authorities’ websites; and IMF staff estimates.

Figure 9.Corporate Sector: Vulnerabilities Remain High

Sources: China Court; CEIC Data Company Ltd.; Credireform; Euler Hermes; European Chamber of Commerce; Goldman Sachs; NBS Industrial Firm Survey; Sinotrust; UK Insolvency Service; US Trust Offices; WIND database; Authorities’ websites; and IMF staff estimates.

Table 1.China: Selected Economic Indicators
20122013201420152016201720182019202020212022
Projections
(Annual percentage change, unless otherwise indicated)
NATIONAL ACCOUNTS
Real GDP (base=2015)7.97.87.36.96.76.76.46.46.36.05.8
Total domestic demand7.98.17.27.27.47.06.96.86.66.25.9
Consumption8.77.27.28.38.48.68.07.47.16.66.2
Investment7.19.17.16.16.35.25.56.06.05.85.6
Fixed9.09.36.86.76.75.25.66.16.26.05.8
Inventories (contribution)−0.60.10.2−0.2−0.10.10.00.00.00.00.0
Net exports (contribution)0.30.10.4−0.1−0.5−0.1−0.2−0.2−0.2−0.2−0.1
Total capital formation (percent of GDP)47.247.346.844.744.243.742.942.442.041.741.4
Gross national saving (percent of GDP) 1/49.748.849.047.545.945.144.343.642.942.341.8
LABOR MARKET
Unemployment rate (annual average) 2/5.05.15.15.05.04.94.94.94.94.9
Wages (migrant workers)33.112.910.09.57.17.06.86.76.66.56.5
PRICES
Consumer prices (average)2.62.62.01.42.02.02.42.52.62.62.6
GDP Deflator3.22.41.01.10.02.22.12.02.12.12.0
FINANCIAL
7-day repo rate (percent)4.65.45.12.52.63.0
10 year government bond rate (percent)3.64.63.72.93.13.5
Real effective exchange rate (average)5.66.33.210.2−5.6
Nominal effective exchange rate (average)5.05.33.19.5−6.5
MACRO-FINANCIAL
Total social financing 3/19.117.514.312.412.913.111.012.211.410.89.9
In percent of GDP169.0180.0189.8197.6209.0216.9221.6229.1235.3240.9245.5
Total domestic nonfinancial sector debt18.417.514.316.117.015.713.412.612.211.410.5
In percent of GDP178.7190.3200.7215.8236.4251.0262.1272.0281.3289.6296.7
Domestic credit to the private sector19.816.613.114.716.716.012.111.510.810.29.6
In percent of GDP134.6142.3148.5157.6172.3183.3189.2194.3198.4202.1205.3
House price 4/8.77.71.49.111.310.48.68.37.97.26.8
Household disposable income (percent of GDP)59.460.060.761.061.461.762.062.062.162.262.5
Household savings (percent of disposable income)40.838.538.037.635.935.334.133.032.031.331.0
Household debt (percent of GDP)29.633.035.438.244.246.348.751.354.057.160.5
Non-financial corporate domestic debt (percent of GDP)105.0109.3113.0119.4128.1134.9138.5141.1142.7143.5143.3
GENERAL GOVERNMENT (Percent of GDP)
Net lending/borrowing 5/−0.3−0.8−0.9−2.8−3.7−3.7−3.7−3.9−4.0−4.1−4.2
Revenue27.827.728.128.528.227.427.327.226.926.826.7
Expenditure28.128.529.031.331.931.131.131.130.930.930.8
Debt 6/15.516.038.636.436.637.538.439.340.241.242.2
Structural balance−0.1−0.5−0.5−2.5−3.6−3.7−3.7−3.9−4.0−4.1−4.2
BALANCE OF PAYMENTS (Percent of GDP)
Current account balance2.51.52.22.71.71.41.31.20.90.70.4
Trade balance3.63.74.15.14.44.13.93.73.43.23.0
Services balance−0.9−1.3−2.0−1.9−2.2−2.3−2.3−2.3−2.4−2.4−2.5
Net international investment position21.820.715.214.916.016.716.716.616.215.614.9
Gross official reserves (bn US$)3,3883,8803,8993,4063,0982,9342,9022,8812,8482,7922,714
MEMORANDUM ITEMS
Nominal GDP (bn RMB) 7/54,09959,69664,71869,91174,63181,34488,36495,919104,067112,608121,482
Augmented debt (percent of GDP) 8/44.148.152.358.262.268.173.278.183.187.691.5
Augmented net lending/borrowing (percent of GDP) 8/−5.1−7.6−7.2−8.4−10.4−10.6−10.8−11.1−11.2−11.0−10.7
Augmented fiscal balance (percent of GDP) 9/−7.8−10.3−9.8−10.2−12.4−12.6−12.6−12.6−12.6−12.3−11.9
Sources: CEIC Data Co., Ltd.; IMF, Information Notice System; and IMF staff estimates and projections.

IMF staff estimates for 2015 and 2016.

Surveyed unemployment rate.

Not adjusted for local government debt swap.

Average selling prices estimated by IMF staff based on housing price data (Commodity Building Residential Price) of 70 large and mid-sized cities published by National Bureau of Statistics (NBS).

Adjustments are made to the authorities’ fiscal budgetary balances to reflect consolidated general government balance, including government-managed funds, state-administered SOE funds, adjustment to the stabilization fund, and social security fund.

Official government debt (narrow definition). Estimates of debt levels before 2015 include central government debt and explicit local government debt (identified by MoF and NPC in Sep 2015). The large increase in general government debt in 2014 reflects the authorities’ recognition of the off-budget local government debt borrowed previously. The estimation of debt levels after 2015 assumes zero off-budget borrowing from 2015 to 2021.

Expenditure side nominal GDP.

Augmented fiscal data expand the perimeter of government to include local government financing vehicles and other off-budget activity.

“Augmented fiscal balance” = “augmented net lending/borrowing” - “net land sales proceeds” (in percent of GDP) i.e. with land sales treated as financing.

Sources: CEIC Data Co., Ltd.; IMF, Information Notice System; and IMF staff estimates and projections.

IMF staff estimates for 2015 and 2016.

Surveyed unemployment rate.

Not adjusted for local government debt swap.

Average selling prices estimated by IMF staff based on housing price data (Commodity Building Residential Price) of 70 large and mid-sized cities published by National Bureau of Statistics (NBS).

Adjustments are made to the authorities’ fiscal budgetary balances to reflect consolidated general government balance, including government-managed funds, state-administered SOE funds, adjustment to the stabilization fund, and social security fund.

Official government debt (narrow definition). Estimates of debt levels before 2015 include central government debt and explicit local government debt (identified by MoF and NPC in Sep 2015). The large increase in general government debt in 2014 reflects the authorities’ recognition of the off-budget local government debt borrowed previously. The estimation of debt levels after 2015 assumes zero off-budget borrowing from 2015 to 2021.

Expenditure side nominal GDP.

Augmented fiscal data expand the perimeter of government to include local government financing vehicles and other off-budget activity.

“Augmented fiscal balance” = “augmented net lending/borrowing” - “net land sales proceeds” (in percent of GDP) i.e. with land sales treated as financing.

Table 2.China: Balance of Payments(In percent of GDP, unless otherwise noted)
20122013201420152016201720182019202020212022
Projections
Current account balance2.51.52.22.71.71.41.31.20.90.70.4
Trade balance3.63.74.15.14.44.13.93.73.43.23.0
Exports23.022.321.319.117.718.317.616.916.215.615.0
Imports19.418.617.214.013.314.213.713.312.812.412.0
Services balance−0.9−1.3−2.0−1.9−2.2−2.3−2.3−2.3−2.4−2.4−2.5
Income balance−0.2−0.80.1−0.4−0.4−0.3−0.2−0.1−0.10.00.0
Current transfers0.0−0.10.0−0.1−0.1−0.1−0.1−0.1−0.1−0.1−0.1
Capital and financial account balance−0.43.6−0.5−3.8−3.7−2.8−1.6−1.3−1.1−1.0−0.9
Capital account0.00.00.00.00.00.00.00.00.00.00.0
Financial account−0.43.6−0.5−3.9−3.7−2.8−1.6−1.3−1.2−1.0−0.9
Net foreign direct investment2.12.31.40.6−0.4−0.8−1.1−1.1−0.9−0.7−0.6
Portfolio investment0.60.50.8−0.6−0.6−0.4−0.4−0.4−0.4−0.3−0.3
Other investment−3.00.7−2.6−3.9−2.7−1.7−0.20.10.10.10.1
Errors and omissions 1/−1.0−0.7−0.6−1.9−2.00.00.00.00.00.00.0
Overall balance1.14.51.1−3.1−3.9−1.4−0.3−0.1−0.2−0.3−0.4
Reserve assets−1.1−4.5−1.13.13.91.40.30.10.20.30.4
International investment position:
Asset60.862.161.154.857.658.457.256.255.053.852.6
Direct investment6.26.98.49.811.713.214.315.215.716.016.2
Secuirties investment2.82.72.52.33.34.14.85.56.16.77.3
Other investment12.312.313.212.415.016.415.815.314.914.614.3
Reserve assets39.540.337.030.327.624.722.320.318.316.614.9
Liability39.041.445.939.941.541.840.840.039.238.638.1
Direct investment24.124.224.724.025.525.624.623.522.421.520.6
Secuirties investment3.94.07.67.37.27.57.57.67.67.88.0
Other investment11.013.213.78.68.88.88.78.99.19.39.5
Net international investment position21.820.715.214.916.016.516.416.215.815.214.6
Memorandum items:
Export growth (value terms)9.28.94.4−4.5−7.28.44.84.54.14.33.9
Import growth (value terms)5.27.71.1−13.4−4.512.15.05.14.95.24.7
FDI (inward)2.83.02.52.21.51.31.10.90.80.80.8
External debt (Billion of US$)1,162.81,532.81,788.11,383.01,420.71,511.71,637.81,808.82,015.92,233.92,463.0
As a percent of GDP13.615.917.012.312.612.812.813.013.313.613.8
Short-term external debt (Billion of US$, remaining maturity)853.51,201.51,311.3920.6870.9898.2972.61,073.51,189.51,311.61,439.8
Gross reserves (Billion of US$) 2/3,387.93,880.43,899.33,406.13,097.82,934.12,901.72,880.82,848.02,792.32,714.3
As a percent of ST debt by remaining maturity397.0323.0297.4370.0355.7326.7298.3268.4239.4212.9188.5
Terms of trade (percentage change)2.81.21.77.67.7−2.80.90.70.50.30.1
Real effective exchange rate (2010 = 100)108.5115.4119.0131.1123.8
Nominal GDP (Billion of US$)8,570.39,635.010,534.511,226.211,232.111,808.612,841.413944.415179.916458.117805.4
Sources: CEIC Data Co., Ltd.; IMF, Information Notice System; and IMF staff estimates and projections.

Includes counterpart transaction to valuation changes.

Includes foreign currency reserves and other reserve assets such as SDRs and gold.

Sources: CEIC Data Co., Ltd.; IMF, Information Notice System; and IMF staff estimates and projections.

Includes counterpart transaction to valuation changes.

Includes foreign currency reserves and other reserve assets such as SDRs and gold.

Table 3.China: External Vulnerability Indicators
20122013201420152016
Monetary and financial indicators
General government debt (in percent of GDP, narrow definition) 1/15.516.038.636.436.6
Broad money (M2: annual percentage change)14.413.611.013.311.3
Foreign currency deposits to broad money (percent)2.62.42.92.93.2
Local currency loans to the economy (annual percentage change)15.014.113.614.313.5
Foreign currency loans to bank domestic credit (in percent)5.35.14.74.03.4
Stock exchange index (end-of-period, December 19, 1990 = 100) 2/2,3762,2143,3893,7043,249.6
Stock exchange capitalization (percent of GDP)53.551.670.394.391.9
Number of listed companies (A-share)2,4722,4682,5922,8083,034
Balance of payments indicators
Exports (annual percentage change, U.S. dollars)9.28.94.4−4.5−7.2
Imports (annual percentage change, U.S. dollars)5.27.71.1−13.4−4.5
Current account balance (percent of GDP)2.51.52.22.71.7
Capital and financial account balance (percent of GDP)−0.43.6−0.5−3.8−3.7
Of which: gross foreign direct investment inflows2.83.02.52.21.5
Reserve indicators
Gross reserves (billions of U.S. dollars) 3/3,3883,8803,8993,4063,098
Gross reserves to imports of goods & services (months)22.324.022.118.218.6
Gross reserves to broad money (M2) (percent)22.021.719.515.213.3
Gross reserves to short-term external debt by remaining maturity (percent)397.0323.0297.4383.8355.7
External debt and balance sheet indicators
Total external debt (percent of GDP)13.615.917.012.312.6
Total external debt (billions of U.S. dollars)1,162.81,532.81,788.11,383.01,420.7
Short-term external debt by original maturity (billions of U.S. dollars)853.51,201.51,311.3887.4870.9
Net foreign assets of banking sector (billions of U.S. dollars)300.4177.3189.4443.8539.8
Total debt to exports of goods & services (percent)53.565.172.658.664.6
Total debt service to exports of goods & services (percent)25.128.952.937.839.9
Of which: Interest payments to exports of goods & services (percent)0.20.20.20.20.2
Foreign-currency long-term sovereign bond ratings (eop)
Moody’sAa3Aa3Aa3Aa3Aa3
Standard and Poor’sAA-AA-AA-AA-AA-
Memorandum items:
International investment position (billions of U.S. dollars)1,866.51,996.01,602.81,672.81,800.5
Nominal GDP (billions of U.S. dollars)8,5709,63510,53511,22611,232
Exports of goods & services (billions of U.S. dollars)2,1752,3562,4632,3602,198
Real effective exchange rate (annual percentage change)5.66.33.210.2−5.6
Sources: CEIC Data Co. Ltd; Bloomberg; IMF, Information Notice System; and IMF staff estimates.

Estimates of debt levels before 2015 include central government debt and explicit local government debt (identified by MoF and NPC in Sep 2015). The large increase in general government debt in 2014 reflects the authorities’ recognition of the off-budget local government debt borrowed previously. The estimation of debt level in 2015 assumes zero off-budget borrowing during 2015.

Shanghai Stock Exchange, A-share.

Includes foreign currency reserves and other reserve assets such as SDRs and gold.

Sources: CEIC Data Co. Ltd; Bloomberg; IMF, Information Notice System; and IMF staff estimates.

Estimates of debt levels before 2015 include central government debt and explicit local government debt (identified by MoF and NPC in Sep 2015). The large increase in general government debt in 2014 reflects the authorities’ recognition of the off-budget local government debt borrowed previously. The estimation of debt level in 2015 assumes zero off-budget borrowing during 2015.

Shanghai Stock Exchange, A-share.

Includes foreign currency reserves and other reserve assets such as SDRs and gold.

Table 4.China: Monetary and Credit Developments
201120122013201420152016
MONETARY SURVEY
(Annual percentage change)
Net foreign assets11.32.98.62.6−2.7−5.9
Monetary authority (contribution)9.11.911.72.4−8.6−9.1
Depository institutions (contribution)2.31.0−3.10.35.93.1
Domestic credit17.117.115.116.223.720.1
Claims on government, net (contribution)1.31.2−0.20.64.04.8
Claims on nonfinanical sectors (contribution)13.413.612.711.413.88.6
Claims on other financial sectors (contribution)2.42.32.64.15.96.6
Broad money (M2)17.314.413.611.013.311.3
M1 (contribution)3.22.22.91.04.36.1
Quasi-money (contribution)14.112.210.710.09.05.2
Reserve money21.212.37.48.5−6.011.8
TOTAL SOCIAL FINANCING
(Annual percentage change)
TSF 1/18.119.117.514.315.016.1
Bank loans (contribution)12.111.510.09.28.88.5
Shadow banking (contribution)3.14.05.12.30.60.9
Net corporate bond financing (contribution)2.13.02.02.22.42.3
Non-financial enterprise equity (contribution)0.70.30.20.40.60.9
LG debt swap0.00.00.00.02.63.5
Others (contribution)0.20.20.20.20.00.0
(In percent of GDP)
TSF 1/157.9169.0180.0189.5202.9218.4
Bank loans117.3121.8125.7130.9137.5143.5
Shadow banking23.026.431.733.131.831.2
Net corporate bond financing10.713.815.618.021.023.9
Non-financial enterprise equity6.05.85.65.96.57.7
LG debt swap0.00.00.00.04.510.8
Others0.91.11.41.61.51.4
MEMORANDUM ITEMS
(In percent)
Nonperforming loans ratio1.01.01.01.31.71.7
Provision coverage ratio (provisions/NPLs)278.1295.5282.7232.1181.2176.4
Liquidity ratio (liquid assets/liquid liabilities)43.245.844.046.448.047.6
Loan to deposit ratio64.965.366.165.167.267.6
Return on assets1.31.31.31.21.11.0
Return on equity20.419.919.217.615.013.4
Capital adequacy ratio12.213.213.513.3
Tier 1 capital adequacy ratio10.010.811.311.3
Core tier 1 capital adequacy ratio10.010.610.910.8
Net open FX position (in percent of capital)4.63.93.73.53.73.5
Sources: Haver Analytics; and IMF staff estimates.

After adjusting local government debt swap.

Sources: Haver Analytics; and IMF staff estimates.

After adjusting local government debt swap.

Table 5.China: General Government
In RMB billions2013201420152016201720182019202020212022
Projections
Est.
Balance: General Budget (authorities’ definition)
(1)Revenue (incl. adjustments) (1a)+(1b)13036141371603216682171061869120378220052386025755
(1a)Headline revenue12921140371522715955168631845520118217182354525409
(1b)Adjustments 1/115100806727243237260286315346
(2)Expenditure (incl. adjustments) (2a)+(2b)14276154971765818862194862134723236251052726029438
(2a)Headline expenditure14021151791758818784194362129223175250382718729358
(2b)Adjustments 2/2553197078505561677381
(3)Fiscal balance (Authorities) (1)-(2)−1240−1360−1626−2180−2380−2656−2857−3100−3400−3683
In percent of GDP 3/−2.1−2.1−2.3−2.9−2.9−3.0−3.0−3.0−3.0−3.0
Balance: General government (staff estimates)
(4)General government revenue (1a)+(4b)+(4c)16538181581994921043222762415026110280253018232395
(1a)General budget headline revenue12921140371522715955168631845520118217182354525409
(4b)Social security revenue3452391944664827517954385709599562956609
(4c)SOE fund revenues 4/165202256260234257283311342377
(5)General government expenditure (2a)+(5b)+(5c)+(5d)18651204632319925316269082895531293336273619938818
(2a)General budget headline expenditure14021151791758818784194362129223175250382718729358
(5b)Social security expenses2862336739364392484551365444577161176484
(5c)SOE fund expenditures 4/151200208217196216237261287316
(5d)Managed funds’ expenditure financed by land sales, bond issuance or carryover 5/1617171714671923243123112437255726082661
(5e)of which: net expenditure financed by land sales1617171713021498159115111487145714081361
(6)Fiscal balance (staff estimate) (4)-(5)−2113−2305−3250−4274−4633−4805−5183−5602−6016−6423
(In percent of GDP)−3.5−3.6−4.6−5.7−5.7−5.4−5.4−5.4−5.3−5.3
General government net borrowing/lending (4)-(5)+(5e)−497−587−1948−2775−3042−3294−3696−4145−4609−5062
(In percent of GDP)−0.8−0.9−2.8−3.7−3.7−3.7−3.9−4.0−4.1−4.2
Balance: Augmented (staff estimates)
(6)Overall fiscal balance−2113−2305−3250−4274−4633−4805−5183−5602−6016−6423
(6a)Additional infrastructure spending financed by LGFV debt4061407234213417377242384572500650204981
(6b)Additional spending of special construction (SCF) and gov’t guided funds (GGF)0−294691566182520492367245927643004
(7)Augmented balance (6) - (6a) - (6b)−6175−6348−7140−9256−10230−11092−12122−13066−13800−14407
(In percent of GDP)−10.3−9.8−10.2−12.4−12.6−12.6−12.6−12.6−12.3−11.9
(8)Augmented net lending/borrowing (7)+(5e) 6/−4558−4630−5838−7758−8639−9581−10635−11609−12393−13047
(In percent of GDP)−7.6−7.2−8.4−10.4−10.6−10.8−11.1−11.2−11.0−10.7
Debt: General government and Augmented (staff estimates)
(9)Central government debt 7/867595661066012007137701528517147193972204225064
(10)Explicit local government debt 8/862154001480015316167051860520505224052430526205
In percent of GDP33.537.238.841.944.747.950.853.555.657.4
(10a)non-swap LG bonds86211621626255239415841774196411154113441
(10b)other recognized LG debt0142381317412764127641276412764127641276412764
(11)General government debt (narrow definition) (9)+(10)9536249662546027323304753389037652418024634751269
(In percent of GDP)16.038.636.436.637.538.439.340.241.242.2
(12)Additional debt likely to be recognized as general government debt 9/125498703293572981711091813924172592065024066
(13)General government debt (staff estimate) (11)+(12)22086258352875333052386474480851575590616699775334
(In percent of GDP)37.039.941.144.347.550.753.856.859.562.0
(13a)Additional LGFV debt not included in staff estimated GG debt66057828903810254114731284414345160101770319409
(13b)Additional debt tied to SCF and GGFs 10/017418063118524870718956114531396816406
(14)Augmented debt (broad definition) (13)+(13a)+(13b)286913383739597464245536864723748768652498669111149
(In percent of GDP)48.152.356.662.268.173.278.183.187.691.5
Sources: CEIC, Data Co. Ltd.; China Ministry of Finance; NAO; and IMF staff estimates and projections.

Includes central and local governments’ transfers to general budget from various funds, carry-over.

Includes adjustments for local government balance carried forward, redemption of local government bonds prior to 2014 and government bond issued under government managed funds.

GDP in this table refers to expenditure side nominal GDP.

Including only revenues/expenditures for the year, and excluding transfers to general budget and carry over.

Includes carry over counted as revenue, adjustments to local government spending, proceeding from issuing special purpose bonds, and net expenditure financed by land sales estimated by subtracting the acquisition cost, compensation to farmers, and land development from the gross land sale proceeds.

The overall net lending/borrowing includes net land sale proceeds as a decrease in nonfinancial assets recorded above the line.

Ministry of Finance debt only, excludes bonds issued for bank recapitalization and asset management companies.

Includes local government bonds and explicit debt.

10% of government contingent debt in 2014. Contingent debt in 2014 is estimated using LGFV total debt minus explicit LG debt of 15.4 Tr. Thereafter, 2/3 of new contingent debt are assumed likely to be recognized as general government debt.

Total social capital constribution to SCF and GGFs.

Sources: CEIC, Data Co. Ltd.; China Ministry of Finance; NAO; and IMF staff estimates and projections.

Includes central and local governments’ transfers to general budget from various funds, carry-over.

Includes adjustments for local government balance carried forward, redemption of local government bonds prior to 2014 and government bond issued under government managed funds.

GDP in this table refers to expenditure side nominal GDP.

Including only revenues/expenditures for the year, and excluding transfers to general budget and carry over.

Includes carry over counted as revenue, adjustments to local government spending, proceeding from issuing special purpose bonds, and net expenditure financed by land sales estimated by subtracting the acquisition cost, compensation to farmers, and land development from the gross land sale proceeds.

The overall net lending/borrowing includes net land sale proceeds as a decrease in nonfinancial assets recorded above the line.

Ministry of Finance debt only, excludes bonds issued for bank recapitalization and asset management companies.

Includes local government bonds and explicit debt.

10% of government contingent debt in 2014. Contingent debt in 2014 is estimated using LGFV total debt minus explicit LG debt of 15.4 Tr. Thereafter, 2/3 of new contingent debt are assumed likely to be recognized as general government debt.

Total social capital constribution to SCF and GGFs.

Table 6.China: Non-Financial Sector Debt
(In RMB trillion)(In percent of GDP)Coverage
201420152016201420152016
Total134155180207222242
Central government9.61112151516
Local governmentMOF GG
Regular financing1.223223Debt
Former LGFV debt 1/141313221917Staff GG
Local government financing vehicles (LGFV)Debt
“Likely” to be recognizedAugmented Debt
As per the 2014 audit111111
New borrowing in 2015-16 (staff estimate) 2/3/025037
“Unlikely” to be recognized
As per the 2014 audit888121110
New borrowing in 2015-16 (staff estimate) 2/3/013023
Government funds 4/023034
Households232733353844
Corporates (excluding LGFV)Private
Domestic738597113121129Sector Debt
External444666
Memo items:
Corporates (including LGFVs)86101117133145157
of which LGFVs91216131722
Households232733353844
General government (MOF definition)252527393637
Government funds 4/023034
Nominal GDP657075
Sources: CEIC Data Co., Ltd.; Ministry of Finance; and IMF staff estimates.

LGFV debt recognized as LG debt as of 2014 (by the 2014 audit).

New LGFV borrowing estimate for 2015-16 is based on infrastructure fixed asset investment data.

Relative share of “likely to be recognized” new LGFV borrowing is based on the historical recognition ratio.

Government guided funds (GGF) and special construction funds (SCF). Social capital portion only.

Sources: CEIC Data Co., Ltd.; Ministry of Finance; and IMF staff estimates.

LGFV debt recognized as LG debt as of 2014 (by the 2014 audit).

New LGFV borrowing estimate for 2015-16 is based on infrastructure fixed asset investment data.

Relative share of “likely to be recognized” new LGFV borrowing is based on the historical recognition ratio.

Government guided funds (GGF) and special construction funds (SCF). Social capital portion only.

Appendix I. External Sector Report
China
Foreign asset and liability position and trajectoryBackground. The net international investment position (NIIP), while positive, has deteriorated to 16 percent of GDP at end-2016 after peaking at 33 percent of GDP in 2007. This deterioration is driven by much-reduced current account (CA) surpluses, valuation changes, and sustained high GDP growth. Gross foreign assets (58 percent of GDP by end-September 2016) are dominated by foreign reserves, while gross liabilities (42 percent of GDP) mainly represent inward FDI, largely because direct investment has been the most open part of the capital account. Reserve assets fell to US$3.0 trillion by end-2016 (about 27 percent of 2016 GDP), from US$3.3 trillion at end-2015 (about 29 percent of 2015 GDP), due to net sales of reserve assets and valuation changes.

Assessment. The NIIP-to-GDP ratio is expected to remain around current levels over the medium term, consistent with projected CA surpluses. The NIIP is not a major source of risk at this point, given the large foreign reserves and FDI-dominated liabilities. However, capital outflow pressure may persist and reserves fall further as the private sector seeks to accumulate foreign assets faster than non-residents seek to accumulate domestic assets.
Overall Assessment

The external position in 2016 was moderately stronger compared with the level consistent with medium-term fundamentals and desirable policies.

The renminbi, despite moving closer to the level consistent with overall assessment, remained broadly in line with fundamentals and desirable policies. Moreover, even though the RMB is assessed to be in line with fundamentals, the external position remains moderately stronger, reflecting remaining distortions and policy gaps that affect the saving-investment balance, such as inadequate social spending, as well as the result of exchange rates reacting more quickly to short-term market sentiment and not translating rapidly into elimination of the current account gap. Going forward, there are potential risks related to protectionist policies by key trading partners.

Potential Policy Responses

External imbalances have declined considerably since the global financial crisis. Achieving a lasting balance in the external position will require continued progress in closing the remaining domestic policy gaps and addressing distortions. Success will move the economy to a more sustainable growth path, with higher consumption and lower overall saving. This can be achieved through successful implementation of the authorities’ reform agenda as well as consistent macroeconomic policies. Priorities include improving the social safety net; SOE reform and opening markets to more competition; creating a more market-based and robust financial system; taking steps to attract more inward FDI, including by ensuring that foreign investors receive the same treatment as domestic investors; and achieving a flexible, market-based exchange rate with a better communication strategy. Continuing the move toward a more market-based and transparent monetary policy framework is a key element in ensuring an orderly transition to an effective float, which may also require use of foreign exchange reserves to smooth excessive volatility.
Current accountBackground. The CA surplus declined to 1.7 percent of GDP in 2016 (1.8 percent of GDP cyclically adjusted), which was about 1.0 percent of GDP lower than in 2015. The decrease of the CA surplus was mainly due to shrinking trade balance (driven by high import volume growth), notwithstanding REER depreciation. From a longer perspective, the CA surplus has fallen substantially relative to its peak of about 10 percent of GDP in 2007, reflecting strong investment growth, REER appreciation, weak demand in major advanced economies, and, more recently, a trend widening of the services deficit.

Assessment. The EBA-estimated CA gap narrowed to about 1.5 percent of GDP in 2016 from 2.4 percent in 2015, primarily due to the declining actual CA surplus. The remaining total gap is mostly accounted for by the residual, reflecting factors other than policy gaps identified in the EBA model, including distortions that encourage excessive savings. The contribution of identified policy gaps is on net mutually offsetting, with loose fiscal policy and excessive credit growth contributing to narrowing the CA gap, largely offset by inadequate health spending and capital controls (which widen the CA gap). There is large uncertainty about China’s cyclical position and possible underestimation of the CA (outbound tourism figures may be somewhat overstated). Also, given the ongoing rebalancing away from investment and high and sticky savings in the short term, the downward trend of the CA could be reversed. Overall, staff assesses the CA remains ½ to 2½ percent of GDP stronger than implied by medium-term fundamentals and desirable policies.
Real exchange rateBackground. In 2016, the average REER depreciated by about 5.1 percent relative to 2015, driven by the depreciation in the NEER (6.5 percent), and reflecting in part the strengthening of the US dollar. As of May 2017, the REER is 2.8 percent weaker relative to the average 2016 level.

Assessment. The EBA REER index regression estimates China’s REER to be 2.7 percent weaker than levels warranted by fundamentals and desirable policies in 2016, compared to 3.9 percent stronger in 2015. 1/ The move to a marginally negative gap reflects the depreciation of the REER in 2016. However, this assessment is subject to large uncertainties related to the outlook and shifts in portfolio allocation preferences. 2/ Overall, staff assesses the REER to be broadly consistent with fundamentals and desirable policies, with the gap being in the range of −10 to +10 percent.
Capital and financial accounts: flows and policy measuresBackground. After a long period of net capital inflows, the financial account recorded a net outflow of US$647.4 billion in 2015 and US$639.7 billion in 2016. Net direct investment inflows shrank, as FDI inflows slowed and Overseas Direct Investment surged in 2016. The substantially negative errors and omissions (2.0 percent of GDP) are included as capital outflows as they are likely to be unrecorded capital rather than CA transactions. Notwithstanding some new inflow liberalization measures in 2016, China’s capital account remains relatively closed in a de jure sense and the authorities have materially increased the enforcement of existing measures to help reduce outflow pressure. Following measures taken in 2016Q4, outflows have moderated in recent months.

Assessment. Over the medium term, the sequence of capital control loosening, consistent with exchange rate flexibility should carefully take into account domestic financial stability. The further opening of the capital account is likely to lead to sizable gross flows in both directions. The adjustment path is hard to predict, and equilibrating such balance sheet adjustments and shifts in market sentiment argues for prioritizing the move to an effective float (while using foreign reserves to a limited degree to smooth excessive volatility) and strengthening domestic financial stability, over substantial further liberalization of the capital account. Efforts should be stepped up to encourage inward FDI, which would generate positive growth spillovers from the import of foreign technology and improving corporate governance standards.
FX intervention and reserves levelBackground. After a long period of reserve accumulation, FX reserves declined in 2015 and 2016 by US$513 billion and US$320 billion, respectively, of which intervention accounted for about US$342 billion and US$448 billion.

Assessment. Reserves stood at 105 percent of the IMF’s composite metric unadjusted for capital controls at end-2016 (down from 118 in 2015); relative to the metric adjusted for capital controls, reserves stood at 171 percent (down from 190 in 2015). The decline of the ratio is driven not only by lower reserves but also by higher broad money (M2) growth which is driving up the metric. Given that the progress made in capital account liberalization over time was partly reversed by the recent capital account tightening measures, the capital account is considered partially open. Consequently, reserves would be considered adequate in the range indicated by the adjusted and unadjusted metrics. Overall, staff assesses the current level of reserves to be adequate.
Technical Background Notes1/ The EBA REER Level model estimates a total REER gap of 6.3 percent, with identified policy gaps of 3.4 percent. However, the model fit of the EBA REER Level model is very poor for China.

2/ Changing expectations about monetary and exchange rate policy, re-evaluation of the government’s reform agenda, or a desire by residents to diversity into foreign assets can trigger large changes in capital flows and exchange rate pressures, even in the absence of significant changes in fundamentals as captured by EBA.
Appendix II. Debt Sustainability Analysis

Because of uncertainty about the perimeter of general government, the debt sustainability analysis assesses government debt under narrow and broad definitions. Government debt under the narrow definition remains low under the baseline but on a slight upward path. “Augmented” debt (broad definition), however, is high and increases strongly. These results reflect a deterioration of debt dynamics compared to last year’s DSA due to an upward revision of the deficit path throughout the projection period. The risk of debt stress depends fundamentally on the willingness to reduce public investment.

China’s public debt sustainability analysis (DSA) is based on the following assumptions:

  • Public debt coverage. Two definitions of debt are used. The main difference is the coverage of local government debt.
    • The narrow coverage scenario includes central government debt and “on-budget” local government debt identified by the authorities. For 2004–13, general government debt includes central government debt and local government bonds (issued by the central government). From 2014, general government debt includes central government debt and explicit local government debt (which consists of local government bonds and other recognized off-budget liabilities incurred by end-2014). The change of definition in 2014 is mainly a result of the change of official data coverage when 2/3 of LGFV debt was explicitly recognized as government liability.
    • “Augmented” debt is used in the broad coverage scenario. It adds other types of local government borrowing, including off-budget liabilities (explicit or contingent) borrowed by Local Government Financing Vehicles (LGFVs) via bank loans, bonds, trust loans and other funding sources, estimated by staff. It also covers debt of government-guided funds and special construction funds, whose activities are considered quasi-fiscal and are new additions to this Article IV. The augmented deficit is the flow counterpart of augmented debt. Augmented data are a complement to general government data. Data limitations mean some nongovernment activity is likely included, and some LGFV and funds may end up having substantial revenues. But there is also the possibility that some further government activity is not included even in the augmented definitions, such as PPPs (many of which do not include private partners and are solely government funded—the stock of approved PPP projects in end-2016 is 3.2 trillion or 4.3 percent of GDP). PPPs are not currently included in the “Augmented” estimate due to data gaps but may be in the future (SIP).
  • Macroeconomic assumptions:The projection reflects a gradual slowdown of real GDP growth to 5¾ percent y/y by 2022 and GDP deflator of about 2¼ percent. The fiscal assumptions differ in the scenarios with general government debt or augmented debt.
    • Fiscal balance in the narrow coverage scenario. This scenario assumes all spending is done within the confines of the budget, and thus that the new budget law is strictly implemented and off-budget public investment is sharply reduced. Under this scenario, primary fiscal deficits are assumed to decline from 5 percent of GDP in 2016 to 4 percent of GDP in 2022, driven chiefly by the assumption of a gradual decline in managed funds’ spending financed by land sales while the remaining on-budget primary balance remains flat.
    • Fiscal balance in the broad coverage scenario. Off-budget local government spending is assumed to remain elevated and decline only marginally (instead of being immediately discontinued, as in the narrow coverage scenario). The augmented primary deficit, which includes the on-budget fiscal deficit and off-budget spending financed by LGFV debt and government funds, is projected to decline from around 9½ percent of GDP in 2016 to around 7½ percent of GDP by 2022. The interest bill increases significantly from around 3 to 4¼ percent of GDP reflecting the rapid buildup of augmented debt. Augmented expenditure to GDP ratio is projected to decline owing to: (i) lower expenditure financed by net land sales; and (ii) the authorities’ reforms to limit local government borrowing.
    • Local government financing. While many local governments relied on net revenue from land sales and LGFV borrowing to finance their investment in the past, the DSA assumes that future financing needs will be increasingly met by bond issuance, in line with the authorities’ plan to replace all local government debt with bonds within three years.
    • Interest rates and amortization. The interest rates for central government and local government bonds are assumed to be about 3–4 percent (in line with historical average). The interest rates of off-budget borrowing (only in the augmented scenario) are assumed to be about 6–7 percent (based on the yield differential between sovereign bond and LGFV bank loans and other short-term instruments). The amortization profile is not central to the analysis because staff assume all maturing debt will be rolled over, although we note a strong front loading of refinancing due to the ongoing swap of legacy LGFV loans for LG bonds.

In the narrow coverage scenario, general government debt is on a slightly increasing path.

General government debt under narrow coverage at 37 percent of GDP in end-2016 is increasing gradually and would only stabilize at 82 percent of GDP in the very long-run if the 2016 primary balance is maintained (see text chart in the box below). This despite a favorable growth-interest rate differential. Off-budget local government spending is assumed to stop immediately after the implementation of the new budget law in 2015 (which would likely imply a significantly weaker growth path than under the baseline, but this is not factored in).

In the broad coverage scenario, debt continues to rise rapidly and consolidation would be needed to prevent it from stabilizing at a very high level.

  • Augmented debt rises rapidly to about 92 percent of GDP in 2022 from around 62 percent of GDP in 2016. This is because the broad coverage scenario assumes that local government off-budget spending continues after 2015 (although expected to decline marginally in the medium term). The debt dynamics are such that debt would only stabilize at a very high level (above 300 percent of GDP, see text chart in the box below) if the 2016 primary balance is maintained.

China faces relatively low risks to debt sustainability, but is vulnerable to contingent liability shocks.

  • In the narrow coverage scenario, general government debt remains relatively low and stabilizes at a still reasonable level in all standard stress tests except for the scenario with contingent liability shocks. A contingent liability shock in 2018 will result in a sharp increase from about 37 percent of GDP in 2016 to about 67 percent of GDP in 2019.3 While the debt level is still manageable, the authorities would potentially have to deal with an increase in gross financing needs that could be sensitive to market financing conditions. Without any extra fiscal consolidation, the debt-to-GDP ratio would stay around 70 percent of GDP over the medium term under the contingent liability shock.

China’s debt profile will largely depend on the implementation of the new budget law and, more fundamentally, on the willingness to reduce public investment.

  • Based on the projected debt dynamics under the narrow coverage, China’s debt profile is still manageable, especially given that it is mostly domestically financed.
  • However, the debt profile crucially depends on the implementation of the new budget law (and thus the size of public investment), as it determines whether debt dynamics will be closer to the narrow coverage scenario or broad coverage scenario. If the new budget law is strictly implemented and future LGFV borrowing will be completely on a commercial basis, debt-to-GDP will rise only gradually and stabilize around 82 percent of GDP in the very long-run without a need for consolidation. However, if local governments continue to incur off-budget liabilities, the debt profile will move closer to that in the broad coverage scenario and the debt-to-GDP ratio will continue to rise in the medium term and only stabilize at very high levels, underpinning the need for consolidation under that scenario.

Box 1.How much fiscal space does China have?

While China’s government debt in 2016 is comparable to that of EMEs…

Depending on the definition of government, China’s debt in 2016 is between 37 percent – just above the 25th percentile of EMEs – and 62 percent, closer to the 75th percentile of EMEs.

The lower-end debt number is narrowly-defined general government debt that includes only explicit debt recognized by the MOF, while the upper-end number also includes off-budget debt (“Augmented” debt).

…there is some fiscal space under the narrowest definition of government…

Despite the favorable growth–interest rate differential, China’s primary balance is still too large to stabilize debt. Under narrowly-defined government, the primary balance in 2016 would stabilize debt at around 82 percent of GDP in the long-run, considerably above the current 75th percentile of EMEs (see chart).

But there is reason to believe China could tolerate significantly higher levels of government debt than almost all EMs—very high savings rate, strong external position, capital controls, limited alternative domestic assets, strong state control and confidence etc.

However, even if debt were to stabilize at 75th percentile of AE debt level (around 100 percent of GDP), that would still only allow for 0.9 percentage points of GDP wider primary deficit from its 2016 level (see chart).

Importantly, temporary fiscal support should be reserved for supporting the reform process or smoothing large shocks, and not geared towards achieving growth targets.

…but under a broader definition of government fiscal space is more limited.

Augmented debt is projected to be larger than the EM 75th percentile in 2017 (see chart). The mild consolidation assumed in staff’s baseline projections of around 2 p.p. of GDP is still insufficient to stabilize debt and thus consolidation would have to be maintained well beyond 2022.

Thus, under the “augmented” definition, there is limited room for permanent fiscal expansions and the window for temporary expansions is closing. These concerns are all the more important given the spending pressures from looming demographic shifts.

Debt Dynamics: Only Some Fiscal Space

(percent of GDP)

Note: Calculations take into account financing from land sales and assume steady-state real growth rate of 5.75 percent and real interest rate of 0.8 and 3 percent for GG and augmented government, respectively.

Explaining the Debt Dynamics Chart

  • The chart shows combinations of debt (x-axis) and primary deficit (y axis).
  • The blue and red bubbles are projections of general government and augmented debt and deficits respectively.
  • The blue and red solid lines represent the debt-stabilizing primary deficit and debt for each level of government.
  • Point A shows that if the 2016 GG deficit is maintained indefinitely, debt stabilizes at 82 percent.
  • Point B shows that if the GG deficit were to widen by 0.9 percent of GDP, debt would stabilize at the 75th percentile of advanced economies.
  • Point C shows that the projected primary balance in 2022 is still 4 percentage points of GDP away from stabilizing debt. Point D shows that if the 2016 augmented deficit is maintained indefinitely, debt stabilizes at 320 percent.

Figure 1.China: Public Sector Debt Sustainability Analysis (General Government Debt Under Narrow Coverage)

(In percent of GDP, unless otherwise indicated)

Source: IMF staff.

1/ Public sector is defined as general government as per authorities definition.

2/ Based on available data.

3/ Long-term bond spread over U.S. bonds.

4/ Defined as interest payments divided by debt stock (excluding guarantees) at the end of previous year.

5/ Derived as [(r - π(1+g) - g + ae(1+r)]/(1+g+π+gπ)) times previous period debt ratio, with r = interest rate; π = growth rate of GDP deflator; g = real GDP growth rate; a = share of foreign-currency denominated debt; and e = nominal exchange rate depreciation (measured by increase in local currency value of U.S. dollar).

6/ The real interest rate contribution is derived from the numerator in footnote 5 as r - π (1+g) and the real growth contribution as -g.

7/ The exchange rate contribution is derived from the numerator in footnote 5 as ae(1+r).

8/ Includes asset changes and interest revenues (if any). For projections, includes exchange rate changes during the projection period.

9/ Assumes that key variables (real GDP growth, real interest rate, and other identified debt-creating flows) remain at the level of the last projection year.

Figure 2.China: Public DSA—Composition of General Government Debt (Narrow Coverage) and Alternative Scenarios

1/ The contingent liability shock scenario also assumes that 10 percent of banking assets would turn into government liabilities.

Figure 3.China: Public DSA (Narrow Coverage) – Realism of Baseline Assumptions

Source : IMF Staff.

1/ Plotted distribution includes all countries, percentile rank refers to all countries.

2/ Projections made in the spring WEO vintage of the preceding year.

3/ China has had a cumulative increase in private sector credit of 46 percent of GDP, 2013-2016. For China, t corresponds to 2017; for the distribution, t corresponds to the first year of the crisis.

4/ Data cover annual obervations from 1990 to 2011 for advanced and emerging economies with debt greater than 60 percent of GDP. Percent of sample on vertical axis.

Figure 4.China: Public DSA (Narrow Coverage) – Stress Tests

Source: IMF staff.

Figure 5.China: Public DSA (Narrow Coverage) – Risk Assessment

Source: IMF staff.

1/ The cell is highlighted in green if debt burden benchmark of 70% is not exceeded under the specific shock or baseline, yellow if exceeded under specific shock but not baseline, red if benchmark is exceeded under baseline, white if stress test is not relevant.

2/ The cell is highlighted in green if gross financing needs benchmark of 15% is not exceeded under the specific shock or baseline, yellow if exceeded under specific shock but not baseline, red if benchmark is exceeded under baseline, white if stress test is not relevant.

3/ The cell is highlighted in green if country value is less than the lower risk-assessment benchmark, red if country value exceeds the upper risk-assessment benchmark, yellow if country value is between the lower and upper risk-assessment benchmarks. If data are unavailable or indicator is not relevant, cell is white. Lower and upper risk-assessment benchmarks are:

200 and 600 basis points for bond spreads; 5 and 15 percent of GDP for external financing requirement; 0.5 and 1 percent for change in the share of short-term debt; 15 and 45 percent for the public debt held by non-residents; and 20 and 60 percent for the share of foreign-currency denominated debt.

4/ Long-term bond spread over U.S. bonds, an average over the last 3 months, 16-Dec-16 through 16-Mar-17.

5/ External financing requirement is defined as the sum of current account deficit, amortization of medium and long-term total external debt, and short-term total external debt at the end of previous period.

Figure 6.China: Public Sector Debt Sustainability Analysis (Broad Coverage)

(In percent of GDP unless otherwise indicated)

Source: IMF staff.

1/ Public sector is defined as the Augmented public sector.

2/ Based on available data.

3/ Long-term bond spread over U.S. bonds.

4/ Defined as interest payments divided by debt stock (excluding guarantees) at the end of previous year.

5/ Derived as [(r - π(1+g) - g + ae(1+r)]/(1+g+π+gπ)) times previous period debt ratio, with r = interest rate; π = growth rate of GDP deflator; g = real GDP growth rate; a = share of foreign-currency denominated debt; and e = nominal exchange rate depreciation (measured by increase in local currency value of U.S. dollar).

6/ The real interest rate contribution is derived from the numerator in footnote 5 as r - π (1+g) and the real growth contribution as -g.

7/ The exchange rate contribution is derived from the numerator in footnote 5 as ae(1+r).

8/ Includes asset changes and interest revenues (if any). For projections, includes exchange rate changes during the projection period.

9/ Assumes that key variables (real GDP growth, real interest rate, and other identified debt-creating flows) remain at the level of the last projection year.

Figure 7.China: Public DSA — Composition of Augmented Debt (Broad Coverage) and Alternative Scenarios

Figure 8.China: Public DSA (Broad Coverage) – Stress Tests

Source: IMF staff.

Figure 9.China: Public DSA (Broad Coverage) – Risk Assessment

Source: IMF staff.

1/ The cell is highlighted in green if debt burden benchmark of 70% is not exceeded under the specific shock or baseline, yellow if exceeded under specific shock but not baseline, red if benchmark is exceeded under baseline, white if stress test is not relevant.

2/ The cell is highlighted in green if gross financing needs benchmark of 15% is not exceeded under the specific shock or baseline, yellow if exceeded under specific shock but not baseline, red if benchmark is exceeded under baseline, white if stress test is not relevant.

3/ The cell is highlighted in green if country value is less than the lower risk-assessment benchmark, red if country value exceeds the upper risk-assessment benchmark, yellow if country value is between the lower and upper risk-assessment benchmarks. If data are unavailable or indicator is not relevant, cell is white. Lower and upper risk-assessment benchmarks are:

200 and 600 basis points for bond spreads; 5 and 15 percent of GDP for external financing requirement; 0.5 and 1 percent for change in the share of short-term debt; 15 and 45 percent for the public debt held by non-residents; and 20 and 60 percent for the share of foreign-currency denominated debt.

4/ Long-term bond spread over U.S. bonds, an average over the last 3 months, 16-Dec-16 through 16-Mar-17.

5/ External financing requirement is defined as the sum of current account deficit, amortization of medium and long-term total external debt, and short-term total external debt at the end of previous period.

Appendix III. Risk Assessment Matrix 1/
LikelihoodImpactPolicy Response
Upside
High
Medium1.Faster-than-anticipated implementation of reforms. Faster implementation of reforms, especially on more opening up of the service industries to private and foreign sectors, corporate debt restructuring, SOE governance and eliminating implicit guarantees, can help address vulnerabilities, increase productivity, and rebalance the economy toward more sustainable growth.
MediumMedium
2.Stronger-than-anticipated recent stimulus. The recent stimulus measures may either have a larger and longer impact on the economy, for example, robust growth in real estate investment. It would increase near-term growth prospects (though likely reduce medium-term growth prospects and increase vulnerabilities).
Downside
Medium
High1. Retreat from cross-border integration. A fraying consensus about the benefits of globalization could lead to protectionism and economic isolationism, leading to reduced global and regional policy collaboration with negative consequences for trade, capital and labor flows, sentiment, and growth. The retreat from cross-border integration is likely to be a drag on China’s export growth.• Support multilateralism and ensure WTO trade rules enforced
High2. Significant further strengthening of the US dollar and/or higher rates. As investors reassess policy fundamentals, as term premia decompress, or if there is a more rapid Fed normalization, leveraged firms, lower-rated sovereigns and those with un-hedged dollar exposures could come under stress. It could result in capital outflow pressure in China, leading to sustained large reserve loss which eventually leads to a disorderly exchange rate depreciation.• Guard against financial risks, enhance crisis preparedness, sustainable macro policies (especially credit). Use on-budget, pro-consumption fiscal stimulus if growth threatens to fall excessively.
High/Medium3. Structurally weak growth in key advanced and emerging economies: Low productivity growth, a failure to fully address crisis legacies and undertake structural reforms, and persistently low inflation undermine medium-term growth in advanced economies. Resource misallocation and policy missteps, including insufficient reforms, exacerbate declining productivity growth in emerging markets. The weak growth in key advanced economies and emerging economies is likely to be a drag on China’s export growth.
High
Medium1. Short-term risks: Defaults in the wake of a sudden tightening in liquidity or a sharp fall in investor confidence could result in a funding shock in the wholesale market in which small banks and NBFIs borrow to expand their balance sheets or a run on the short-term asset management products issued by NBFIs (or a run on the WMPs which fund them). These could be amplified by renewed capital outflows and exchange rate pressure.• Move, flexibly, to effectively floating exchange rate regime, but intervene to mitigate disorderly market conditions.
Medium2. Medium-term risks: slow progress on reform and continued reliance on policy stimulus and unsustainable credit growth, would add to vulnerabilities, worsen resource misallocation, and lead to permanently lower growth.• Enhance reforms to switch growth engines from investment/state to consumption/private sector.
1/ The Risk Assessment Matrix (RAM) shows events that could materially alter the baseline path (the scenario most likely to materialize in the view of IMF staff). The relative likelihood is the staff’s subjective assessment of the risks surrounding the baseline (“low” is meant to indicate a probability below 10 percent, “medium” a probability between 10 and 30 percent, and “high” a probability between 30 and 50 percent). The RAM reflects staff views on the source of risks and overall level of concern as of the time of discussions with the authorities. Non-mutually exclusive risks may interact and materialize jointly. “Short term” and “medium term” are meant to indicate that the risk could materialize within 1 year and 3 years, respectively.
1/ The Risk Assessment Matrix (RAM) shows events that could materially alter the baseline path (the scenario most likely to materialize in the view of IMF staff). The relative likelihood is the staff’s subjective assessment of the risks surrounding the baseline (“low” is meant to indicate a probability below 10 percent, “medium” a probability between 10 and 30 percent, and “high” a probability between 30 and 50 percent). The RAM reflects staff views on the source of risks and overall level of concern as of the time of discussions with the authorities. Non-mutually exclusive risks may interact and materialize jointly. “Short term” and “medium term” are meant to indicate that the risk could materialize within 1 year and 3 years, respectively.
Appendix IV. Progress on Previous Reform Recommendations
Reform Scorecard on the Recommendations in the 2016 Article IV Staff Report
A. Tackling corporate debt
• Proactive loss recognition and sharing of losses
• Harden budget constraints
• Overcapacity cut
• Mitigating social costs of layoffs
B. Accepting the slowdown
• Lower and sustainable growth targets
• Reining in local government borrowing
• Structural fiscal reforms
C. Guarding against financial risks
• Stronger supervision on shadow finance
• Address real estate sector risks
D. Progressing to effective floating exchange rate
• Achieving an effective float over the medium term.
• Adopt a more market-based monetary framework.
• Careful sequencing of reforms on capital account
E. Enhance transparency and communication
F. SOE reforms
Sources: national authorities and staff assessment.
Legend
Substantial progress
Some progress
Limited progress
Recent Reform Measures and Staff Recommendations in 2016 Article IV Report
Staff Recommendations in 2016 Article IV ReportAnnounced Reform Measures since July 2016Date
A. Tackling The Corporate Debt
• High-level decision with coordinated action of all involved public agencies to tackle debt vulnerabilities.• A joint-ministerial committee led by NDRC was established to oversee and coordinate the reforms on deleveraging.Oct 2016
• Harden budget constraints by removing implicit guarantees and better pricing of risks.• A small number of SOEs were allowed to enter bankruptcy procedures.Jul 2016
• Triage debt and restructure weak firms via workouts or liquidation. Greater use of market-based restructuring mechanism, rather than relying on forced mergers.• Focus has been on deleveraging at firm-level, including NDRC’s medium-term target to reduce liabilities-to-asset ratio for steel sector. Indicative target for credit growth slowed to 12 percent y/y (by 1ppt) in 2017. About 20 percent of identified central SOE zombies were resolved.Dec 2016 and Mar 2017
• Recognize losses and burden sharing• Private workouts have increased with banks setting up 12,800 creditor committees to resolve debt (7 percent of bank loans) and initiating 30 debt-equity swaps. Greater use of insolvency framework (over 5,600 cases in 2016, doubling in two years).Jan 2017
• Address overcapacity issues as part of tackling corporate debt• Over-achieved capacity cut target for 2016 and broaden the capacity cut to coal-fueled power and building materials.Nov 2016; Mar 2017
• Provide social safety net for displaced workers and raise SOE dividend payout to the budget to targeted 30 percent.• The restructuring fund of RMB 100 billion has been used to assist 0.7 million affected workers. SASAC also commits to transfer SOE shares to social security funds to address legacy problems.Mar 2017
• Facilitate market entry, particularly in state-dominated services sector.• The government committed to strengthen property rights and open up more sectors for foreign and private investment.Jan-Feb 2017
B. Accepting The Slowdown
• Set macropolicies consistent within reforms, including downplaying the growth target by setting it at a wider range and sustainable level of around 6 percent in 2017, and slowing credit growth substantially.• GDP growth target was set around 6.5 percent, or higher if possible in practice; intend to adopt a gradual approach on deleveraging to prevent adverse feedback loops.Mar 2017
Fiscal reforms
• Adopt a multi-year budget framework to reduce augmented deficits moderately and raising on-budget pro-reform spending.• Introduced a broad timetable for local government budget disclosure by 2017-18.Nov 2016
• Implement the revised budget law and other directives to rein in and improve transparency of local government borrowings; resolve intergovernmental structural imbalances.• Announced guidelines to align intergovernmental relations by 2020; enhanced monitoring and imposing penalties for violations on local government debt borrowings. A new publicprivate partnership (PPP) center was established to strengthen the PPP management.Aug, Oct-Nov 2016; Feb 2017
• Fiscal reforms to modernize tax system, expanding social security with greater portability of benefits, liberalizing residency restrictions.• VAT reforms to reduce number of multiple tax rates; household residency ID card to be introduced nationwide by 2017 could improve portability of social benefits; corporates’ contribution rates on worker insurance were reduced.Jan, Mar 2017
• Raise taxes on fossil fuel and pollution to address environmental externalities.• MoF extended subsidies on renewable energy vehicles and provided tax incentives for technology and innovation sectors. Overcapacity cuts will enforce environmental standards and be expanded to cover polluting coal-fueled power and resource-intensive building materials.Dec 2016; Mar 2017
C. Guarding against Financial Risks
• Proactive recognition of potential loss and strengthen capital ratios.• Average core Tier 1 capital ratios for listed banks has been stable at 11.1 percent in 2016 (relative to 11.2 percent in 2015). Nine banks announced capital raising plan.Mar 2017
• Stronger supervisory focus on liquidity and rapid expansion of shadow and interbank finances.• The MPA framework was introduced to strengthen supervision of financial system, which covers the WMP activity; broad guidelines introduced on internet finance.Jul, Dec 2016; Jan 2017
• CBRC issued several documents on strengthening the enforcement of banking regulations, enhancing risk management and closing avenues for arbitrage activities.Mar - Apr 2017
• Contain asset price bubbles in real estate markets• Multiple measures (e.g., purchase restrictions, higher down payment requirement and mortgage rates) were introduced by local governments to contain asset bubbles.Sep 2016 - Mar 2017
• Upgrade supervisory cooperation and coordination framework and crisis preparedness• Cross regulatory agencies issued regulations on internet finance, including peer-to-peer lending. Regulators plan to roll out a more comprehensive regulatory framework to tighten supervision of asset management products.Aug 2016; Jan 2017
D. Progressing Toward an Effective-Floating Exchange Rate Regime
• Achieving an effective float within the next couple of years.• PBC has made greater reference to the currency basket, including adding more currencies in the calculation and revising the methodology on central parity fixing.Jan - Feb 2017
• Adopt a more market-based monetary framework.• PBC has increasingly communicated on the short-term interbank rates to better guide market liquidity through an interest rate corridor.Dec 2016 and Mar 2017
• Careful sequencing of reforms on capital account liberalization.• Tighter enforcement of capital flow management (de-facto tightening) by requiring additional documentation for outward direct investmentAug 2016 - Jan 2017
• Foreign institutional investors were allowed to hedge foreign exchange risks onshoreFeb 2017
• Lifted restrictions on the asset allocation of QFII and RQFII; Shenzhen-Hong Kong stock connect was launched; plan to launch Shanghai-Hong Kong Bond Connect in 2017.Sep 2016; Mar 2017
• Tighter enforcement of rules on offshore RMB lending by domestic non-financial institutions, by requiring banks to strictly examine 1) whether the busines operating scale of the overseas borrower is suitable for the loan size, and 2) the authenticity and reasonableness of the use of the outbound loan.
E. Enhance Transparency and Communication
• Addressing data gaps for macroeconomic surveillance.• NBS has publicized provincial data falsification and imposed penalty on related personnel. It also published a new index on service sector activity.Jan-Feb 2017
• Strengthen communications on policy objectives and align expectations.• Regular press conferences were held to disseminate policy direction; public agencies (including PBC).Mar 2017
F. Other structural reforms
SOE reforms
• Accelerate SOE reforms including through removing their nonmarket responsibilities.• An SOE restructuring fund of RMB 350 billion was established to facilitate restructuring.Sep 2016
• SASAC aimed to accelerate the mixed-ownership reform in 2017, including Shanghai’s pilot program of employee stock holding. It also encouraged the consolidation of coal, steel, and electricity SOEs.Mar 2017
• The government announced to separate SOEs’ social functions (provision of utilities and property management) by 2018.Jul 2016; Mar 2017
• Complementary reforms to improve labor market flexibility and facilitate SOE reforms• Pricing reform underway in energy, public utility sectors, and some agricultural products.Feb-Mar 2017
• Aim to resettle 100 million rural residents to cities by 2020.Oct 2016
• Measures to strengthen property rights of rural land were introduced, including guidelines on the pledging of rural land as collateral.Oct - Dec 2016; Jan 2017
Sources: 2017 Government Work Plan; national authorities, and media news.
Sources: 2017 Government Work Plan; national authorities, and media news.
1Enforcement was tightened on existing CFMs on capital outflows, including on overseas direct investment and offshore RMB lending; other CFMs were also tightened, including on overseas RMB withdrawals by payment cards (see Informational Annex).
2See the recent IMF book “Modernizing China-Investing in Soft Infrastructure.”
3Mechanically, the standard contingent liability shock in the IMF’s DSA toolkit assumes that 10 percent of non-government banking system assets would turn into government liabilities. Non-government banking system assets were about 220 percent of GDP in 2016. It also assumes that the real GDP growth in 2018 and 2019 would be 2–2.5 percent lower (a one standard deviation shock).

Other Resources Citing This Publication