Conclusion: The threat local government financing vehicle debt poses to China’s banking system is very real. While there is certainly still time and policy wiggle room to deal with the current headwinds, we do think this issue will hang like a dark cloud over the Chinese economy for as long as it remains officially unaddressed.
To start, we think it’s worth stressing that wrapping one’s hands around the exposures and liabilities of the Chinese financial system is akin to explaining the true cause of gravity – insomuch that doing so remains much closer to “impossible” than physicists would prefer. That said, however, we think this growing issue is at least worth attempting to quantify and color.
By now, the headwinds facing Chinese property developers and local government financing vehicles (LGFVs) are not new news, but for those of you who do not follow our work on the Chinese economy closely, it’s worth re-hashing the numbers:
How Much LGFV Debt Is Out There?
Provincial and city governments in China, which are barred by law from issuing municipal bonds to fund investment in infrastructure projects, had set up 6,576 arms-length corporate entities (local government financing vehicles) by the end of 2010 – a reported 25% increase from 2008.
Arriving at a total amount of LGFV debt obligations is no walk in the park; in the year-to-date, we’ve seen estimates as high as ~14.3 trillion yuan (People’s Bank of China) and as low as 7.7 trillion yuan (China Banking Regulatory Commission). A June report out of the National Audit Office appears to have fixated the market on their 10.7 trillion yuan ($1.68 trillion) estimate as of year-end 2010, so we’ll use that admittedly stale figure in our analysis. For reference, the CNY10.7 trillion total is equivalent to 26.9% of China’s 2010 nominal GDP, which compares to the U.S.’s 24.6% muni-bond debt/nominal GDP ratio (not entirely apples-to-apples, however).
As an aside, we think China’s own inability to quantify these obligations speaks volumes to the convoluted nature of this matter.
How Are LGFV Liabilities are Structured?
Roughly 79% (CNY8.5 trillion) of the 10.7 trillion yuan of LGFV debt was in the form of bank loans, 70% of which will mature over the next five years. Nearly one-fourth (24%) percent of the total outstanding debt comes due in 2011 alone.
In terms of who owes what, the PBOC’s 2010 Regional Financial Performance Report suggests that local governments have the responsibility to repay 6.7 trillion yuan (62.6%) of the CNY10.7 trillion total debt outstanding and provided an explicit backstop for another 2.33 trillion yuan (21.7%), leaving the remaining CNY1.67 trillion (15.6%) hanging in the balance. The NAO’s mid-summer report – which consensus has adopted as the most relevant – suggests that some 60% (CNY6.42 tril.) of the total is the “direct” responsibility of local governments for repayment and another CNY2.55 trillion (23.8%) have an explicit backstop in the form of pledged land sales revenue, leaving CNY1.73 trillion (16.2%) hanging in the balance.
How Risky Are LGFV Credits?
As it relates to credit quality, the CBRC’s report on LGFV debt classified half of the loans as “low-risk”, or possessing the ability to be repaid with cash flows, fiscal revenue, or local government land sales (~40% of total local gov’t revenue, per China Real Estate Information Corp). Another 27% of the total was classified as “normal risk”, or possessing the ability to be serviced and repaid with future cash flows. The remaining 23% was classified as “high-risk”, or not backed by collateral or a visible cash flow stream.
Extrapolating these ratios to the NAO’s CNY8.5 trillion figure for total LGFV bank loans, we see that ~CNY4.25 trillion of such loans are of high credit quality, ~CNY2.3 trillion are of average credit quality, and the remaining ~CNY1.96 trillion are of low credit quality.
These findings rhyme with a recent China Asset Management Co. study whose results were published by the nation’s official bond clearinghouse. The study found that 28% of LGFVs have negative cash flow from operations and 22% of them had debt-to-asset ratios greater than 70%. Obviously we must absorb these numbers with a grain of salt, as they are indeed from a potentially conflicted source. Moreover, 28% of local government financing vehicles printing negative cash flows doesn’t necessarily mean 28% of the outstanding loan value is likely to go into default, as LGFVs vary in both size and loan guarantees. Still, the numbers are indeed eye-opening and, at face value, they do lend some credence to the widely-held belief that China faces a potential banking crisis should we see a material pickup in defaults across this space.
Attempting to Quantify the Downside
At 8.5 trillion yuan, loans to local government financing vehicles accounted for 17.7% of the total amount of loans held on Chinese bank balance sheets at the end of 2010. For the purpose of using consistent numbers in our analysis, we’ll ignore China’s oft-bandied about off-balance sheet banking assets. Simply put, there isn’t enough consistent data out there to analyze, but, for whatever it’s worth, Fitch Ratings estimates the disclosed off-balance sheet items to be ~25% of total assets (not all being in the form of loans, of course).
Assuming that LGFV bank loans maintained a similar ratio to total bank loans throughout 2011 we can back our way into a CNY9.3 trillion estimate for local government financing vehicle debt as of the latest reporting period, which shows a total of CNY52.4 trillion in outstanding bank loans throughout the Chinese financial system.
Currently, the Chinese banking system in aggregate has a rather healthy 1% non-performing loan ratio, which is the lowest on record (data going back to 1Q04). Over the last seven years, the ratio has fallen dramatically from a peak of 40% for state-owned banks during China’s last banking crisis (per the Cato Institute), which cost the Chinese government nearly 5 trillion yuan (27% of 2005 nominal GDP) in the form of bank recapitalizations, etc. in the eight years though 2005 (per Barclays). Like the current [potential] crisis, which was brought on by the Chinese government’s CNY4 trillion stimulus package introduced in November ’08, China previous banking crisis was brought upon by the government forcing banks to extend credit to largely unprofitable state-owned enterprises – whose cash flows were under the control of local governments then just as LGFV cash flows are currently.
So where could sour LGFV debt take China’s NPL ratio go from here? Well, we’d be lying if we told you we had an official estimate. We’ve seen estimates as high as 12% (Moody’s), which obviously seems quite aggressive. But without data to perform detailed credit analysis on each of the loans, as well as a crystal ball that could provide us a trustworthy estimate of credit conditions and the ability of these borrowers to obtain refinancing over the next five years, all we could do is speculate – which we’d argue isn’t entirely helpful.
What we can do that is helpful is keep you informed of any major developments pertaining to policy remedies and reported credit quality improvement/deterioration, all the while monitoring the health of the Chinese banking system along the way.
What we do know is that, as a result of well-documented government efforts to cool real estate speculation and consumer-price inflation, Chinese property prices have been under assault since 1Q10, which creates a headwind for local government revenue collection (40% of total) – the same revenues that are being pledged to backstop ~84% of LGFV debt. As we’ve outlined in our previous work, Deflating the Inflation will ultimately allow the PBOC to relax monetary policy, which should be accompanied by easing credit conditions, which, in theory, should then provide a boost on the margin to Chinese property prices. Still, a chart of the Shanghai Stock Exchange Property Index (34 developers and financiers) tells us that the headwinds facing the Chinese property market (and, as an extension, Chinese financials) are not going away any time soon.
As it relates to policy developments, in early August, it was leaked by an unnamed source that the Chinese Finance Ministry was drafting a preliminary plan to allow local governments to issue [long-term] bonds directly to investors. If implemented, this would be quite a constructive development as it would allow for the smoothing of any LGFV asset/liability mismatches arising from negative cash flows over the short-to-medium term. While we do not find that piling debt-upon-debt to be a “solution” that is likely to end well, we’d be remiss to ignore the positive impact this would ultimately have on alleviating any potential liquidity headwinds facing underwater LGFV borrowers.
All told, the threat local government financing vehicle debt poses to China’s banking system is very real.
While there is certainly still time and policy wiggle room to deal with the current headwinds (central gov’t debt/GDP = 33.8% per IMF; deficit/GDP = 2% per Bloomberg), we do think this issue will hang like a dark cloud over the Chinese economy for as long as it remains officially unaddressed.
Moreover, any large-scale “fix” of the issue could potentially be well over a year away if it is likely that it doesn’t occur until after the 18thPolitburo takes the leadership reins in March ‘13. That might potentially work in the Communist party’s favor in two ways: 1) it would allow current leadership to not have to admit it made a consequential mistake with the aggressive stimulus package of 2008-09 and 2) it could provide a “save the day” moment for China’s new leaders.
Net-net-net, a potential Chinese banking crisis is noteworthy risk to keep on your radar, and the likelihood that it hangs sustainably over the both China and the global economy for an extended period of time certainly won’t help investor confidence.