The risks of high debt levels at China's local and regional governments (LRGs) have long been well-known. But what's contentious is just how big those underlying debts really are. In particular, debate surrounds the amount of "implicit" or hidden debt, which is loosely defined. But where everyone seems to agree is that the official figures are probably an understatement of the LRGs' indebtedness.

S&P Global Ratings has outlined the key economic and fiscal debt indicators by province (see "Chinese LRG Risk Indicators By Province," published on RatingsDirect on Oct. 16, 2018). In that report, we highlight the relative strengths and weaknesses of provinces across the country, particularly their official debt burden and our estimated underlying debt burden.

Investors frequently ask us about the credit quality of Chinese LRGs, especially around the possibility of another debt-for-swap program and the role of new LRG bonds in reducing implicit debt. We address these issues in this report.

Frequently Asked Questions

How big is the LRG sector's underlying debt burden?

We estimate the LRGs' underlying debt at Chinese renminbi (RMB) 37 trillion (US$5.5 trillion) in 2017, or 45% of GDP (see chart 1). This includes RMB31 trillion in the LRGs' cumulative new borrowings under their own names and the debt from their local government financial vehicles (LGFVs). Our estimate is much larger than the RMB16.5 trillion stated officially and recorded on LRG balance sheets.

We use the term "underlying debt burden" to distinguish it from the official LRG debt, which we believe underestimates the sector's actual debt burden.

Our estimate of underlying debt highlights that LGFV debt grew rapidly over the past decade, at almost 20% a year, which is much faster than nominal GDP. Plotting this LGFV debt growth against infrastructure investment reveals that LRG underlying debt growth goes hand in hand with infrastructure investment.

The debt leverage ratio may appear daunting but it's not, in itself, the problem. Rather, LGFV borrowers or their local government owners operate under varying credit conditions and LRG revenue sources cannot match the debt buildup. This presents significant challenges to fiscal sustainability.

We expect LRGs' underlying debt leverage to stabilize at around 44% of GDP over the next three years, given the central government's policy rigor and the knock-on effects from tighter financial regulation.

Chart 1

What constitutes an LGFV?

There's no official definition. But in a nutshell, LGFVs are mainly engaged in infrastructure and land development. They are mandated to do so by a local government, which provides ongoing financial and administrative support to enable LGFVs to meet their debt obligations and maintain their financial standing.

The definition is important because LGFVs make up the bulk of LRGs' underlying debt. But whether a company is a state-owned entity (SOE) or an LGFV can't be judged by its name; some may contain the word "construction" or "investment" but could be an SOE or LGFV.

No LGFV has ever defaulted in the onshore or offshore bond market. The default by Xinjiang Sixth Agriculture State-Owned Assets Management in August this year was widely reported, but we don't view this entity as an LGFV. Instead, we view it as a normal corporate SOE that is engaged in agriculture business.

We've identified 1,350 LGFVs nationwide, based on their public investment role and close linkage with their governments. Of course, this may not be an exhaustive list. We can only aggregate those LGFVs that have issued bonds in the onshore market, which allows us to see their financial statements. If we assess an SOE to be an LGFV, we will include its entire balance sheet liabilities as part of the underlying debt burden of its respective LRG (see chart 2).

Chart 2

Could new LRG bonds replace the role of LGFVs for LRGs?

Not yet, but they are reducing the LGFVs' financing function for LRGs, particularly related to projects that aren't financially viable.

The launch of LRG bonds in 2015 is one of the most prominent remedial measures that Chinese policymakers have taken to reduce LRGs' reliance on off-budget financing through LGFVs. These bonds are issued in provincial governments' names on behalf of the entire LRG sector. In 2017, the LRG sector issued RMB1.6 trillion in new bonds, bringing cumulative issuance to RMB3.4 trillion for 2015-2017. We expect new LRG bond issuance to increase to around RMB4 trillion in 2020 (see chart 3).

However, in the next few years, we don't believe the size of new LRG bonds will be big enough to replace LGFVs' financing and investment role for LRGs.

New LRG bonds may help fully finance small-scale capital programs (typically around RMB1 trillion) with no cash-flow generation. However, most infrastructure programs are large and expensive, with long-term duration construction phases. It's therefore unrealistic for LRGs to rely on bond proceeds or even tax revenues to fully fund such projects. Instead, LRGs provide bond proceeds to LGFVs as equity injections, so that LGFVs can leverage from them to channel financial resources. In such cases, LGFVs will maintain their financing function. But if LRGs can self-fund more of the equity component, it would help LGFVs reduce leverage and slow their debt buildup.

Chart 3

What's special about project-revenue-backed LRG bonds?

Project-revenue-backed (PRB) LRG bonds are likely to remain the key driver of the increase in new LRG bond issuance. This is mainly because increases in PRB bonds have no implications for China's headline fiscal deficit. That means they will be a key channel for China to expand its fiscal space if needed.

When the central government launched LRG bonds, they were classified into two types: (1) general-revenue-backed (GRB) LRG bonds; and (2) PRB LRG bonds. Market commentators sometimes refer to PRB bonds as "special purpose bonds."

The key differences between GRB and PRB bonds lie in where they are recorded in the government's budgetary accounts and associated repayment funding sources. GRB bonds will be repaid through tax revenue, while PRB bonds will be repaid mainly through land sales. PRB bonds were introduced in 2015, but policymakers have particularly encouraged LRGs to use this funding channel since 2017.

China reports its headline fiscal deficit based on the debt financing recorded in the country's general budget. Any debt-financing activities stated in the government's funds budget is excluded from that deficit. This is why LRGs' overall new bond quota increased 3.6x over 2015-2018, even though PRB bonds' share of issuance jumped to 61% from 20%; compared with flattish growth for GRB bonds.

China's policymakers are now under pressure to stabilize infrastructure investments. They are also being urged to adopt more pro-growth fiscal expansion and expenditure measures, including aggressive tax cuts. We expect the central government to rapidly increase the quota for new bond issuance among LRGs over the next few years, and can they only do so by increasing PRB bonds to achieve that goal.

What capacity do LRGs have to service their debt obligations?

It varies. Debt-servicing capacity is closely related to China's sovereign credit story (see "People's Republic Of China Rating Affirmed At 'A+/A-1'; Outlook Stable," published on Sept. 21, 2018). Currently, the only legal obligor of Chinese LRGs official debt are the country's 31 provincial governments plus five cities that enjoy province-level status in state economic planning. A default on a provincial bond by an LRG would have an immeasurable impact on China's reputational risk, and may have a chain effect on financial stability and confidence.

The LRG bond market is in its infancy, and policymakers are trying hard to foster this market by making LRG bonds more attractive as an asset class. This could enable governments to tap into the market more often for new borrowing and refinancing purposes. In August 2018, the central government rolled out policies aimed at shoring up the market confidence for LRG bonds. To attract more investors, these measures include reducing the risk weighting of banks holding LRG bonds and ordering that LRG bond yields should have a spread of at least 40 basis points to central government bonds.

These steps suggest that the likelihood of a default by a provincial government is unlikely, but the central government's support will be key. As such, the creditworthiness of provincial governments--the legal obligors--has relatively small variance and is close to sovereign credit quality.

Is China reversing the risk controls over local government debt issues?

We don't see any U-turn in policy, and believe the central government remains committed to reducing the risks of the LRG debt overhang. It's true, though, that China's policymakers didn't act or sound tough toward LGFVs from June and July 2018, when the State Council's Executive Committee stated support for LRGs' "necessary" infrastructure investments and LGFVs' "reasonable" borrowing needs. In addition, China's overall macroeconomic setting has shifted toward a more accommodative stance, as reflected in multiple reserve requirement ratio cuts this year to ease liquidity conditions.

In our view, policymakers have become more cautious about the adverse impact on infrastructure investment and economic stability if they act too harshly toward LGFVs, and won't want to risk them going under. The central government is therefore striking a balance between maintaining economic and financial stability and taming LRG debt risks.

We believe the central government's perceived softness toward LGFV financing is mainly intended for existing infrastructure projects under construction. Approvals for new projects and their financing framework will be regulated more strictly due to a series of regulatory measures rolled out last year.

How will the central government deal with LRGs' implicit debt?

Chinese policymakers seem to be admitting that LRGs' underlying debt burden is much larger than what current official debt stock presents. This is evidenced by widely media reports that the central government is quietly auditing LRGs on their underling debt stock, particularly the amount occurred between 2015 and 2017. So, now the question is what the central government would do to absorb these implicit debt risk?

In our opinion, the central government will unlikely shift large amount of implicit debt from LGFVs' balance sheets onto LRGs' balance sheets and embark on another round of debt-swap program as what they did in 2014. Instead, policymakers will opt for a combination measures including land sales or increasing new LRG bond quota to gradually absorb the risks associated with implicit debt.

This view is based on two reasons. First, the central government won't want to admit that its reported LRG debt figures for 2014 were incorrect. Second, the current official debt stock, which amounted to RMB16.5 trillion as end of 2017, is already a heavy debt burden on LRGs that won't be able to take on additional explicit debt obligations.

In fact, in 2018, the majority of provinces have started to issue refinancing LRG bonds to roller over their three-year LRG bonds issued in 2015 that come due this year. With the debt maturity peaking in 2021 and 2022, LRGs will be facing increasing debt servicing pressure from now on. In this case, LRGs have no room to add more implicit debt onto their official debt scope.

So what does all this mean for LGFVs? We believe they may gain some breathing space in terms of financing conditions that may help their liquidity conditions. We have already factored LRGs' apparent or hidden debt into our estimates of their debt burdens, so we don't expect additional debt surprises to affect our assessments of their credit profiles. But the fast increase in new LRG bond issuance should lead to wider budgetary deficit for certain provinces, and that may weigh on their credit quality.

Related Research

• Chinese LRG Risk Indicators By Province, Oct. 16, 2018
• People's Republic Of China Rating Affirmed At 'A+/A-1'; Outlook Stable, Sept. 21, 2018