LONG VIEW: LGFV Fragility
LONG VIEW: LGFV Fragility
Using macro data we build a Fragility Index for 33 cities in China to be used as a backdrop proxy for Local Government Financing Vehicle (LGFV) fragility. Data file attached.
The Setup:
As China’s property crash sends tremors across the economy, investors looking for the next big quake are zeroing in on local government financing vehicles [LGFVs]. It’s a reasonable choice -- LGFVs are loaded with debt, disclosure is poor and many ply a living in real estate. It’s also off the mark.
That doesn’t mean there won’t be upsets as the housing market deflates. No LGFV has defaulted on a bond payment yet in this cycle. We don’t rule one out ahead.
— Bloomberg Intelligence: “CHINA INSIGHT: Housing Rout Spells LGFV Defaults, Not Crisis (2)”
Using macro data, we build a Fragility Index for 33 cities in China to be used as a backdrop proxy for Local Government Financing Vehicle (LGFV) fragility (i.e. the likelihood of default). The idea is that — due to the often close relationship between local government and LGFVs — if the municipality itself is negatively exposed to real estate and its deteriorating conditions, the associated LGFVs will likely be so too. Idiosyncratic risks between LGFVs remain and should be analysed separately. This is not investment advice, do your own research.
The data file is attached for those who want to tinker with the calculations.
Contents:
Fragility Index Formula & Rationale
Fragility Index Chart
Index Components in Charts
Data File (.csv)
1. Fragility Index Formula & Rationale
The composite index is formulated using the following components:
Real estate overhang
The NBS publishes a statistic called “Floorspace under construction”. This stock of residential housing has begun construction yet remains unfinished. For many cities, the statistic exhibits an upward trend. Housing starts outstrip housing completions, and this accumulates over the years. It’s unclear whether NBS ever writes off any unfinished housing stock. When compared to the city’s population, we can get an idea of the municipality’s overextension to runaway construction projects on a per capita basis.
Land sales dependence
This is the percentage of municipal revenues attributed to land transfer fees. More details on the years-long trend of increasing reliance on this particular source of revenues by local governments are in our previous post, “THE BRIEF: The Local vs. Central Divide”. The idea is that a slowdown in the property sector, by extension, negatively impacts local government finances, which then spend less. Depending on their magnitude, budget cuts may precipitate ripple effects across the city.
Crisis of collateral
For years, many have seen real estate prices as incapable of falling. Much collateral rests on this idea of ever-increasing prices. See our previous post, “LONG VIEW: The End of a Chapter”, for more details. But once prices falter, a crisis of collateral may ensue. In our formulation, we use the decline from the rolling historical peak for both new and secondhand homes as an input into the index.
Please remember that the formula shown above is our attempt at quantifying the problem. There may be other or better ways of doing it. We are attaching a .csv file below for those who would like to tweak the formula. There are a total of 10 variables (columns) available for calculations. E.g. you may want to add fiscal deficit as a component of the index or adjust the component weights.
2. Fragility Index Chart
The index is constructed so that as long as property prices keep hitting all-time highs, even high exposure to real estate and dependence on land transfer fees don’t influence the index much. For example, Chengdu, with all-time high housing price readings for both new and secondhand homes, nullifies the city’s exposure to the other two index components. The game of musical chairs is still on. However, the vulnerabilities in the other two index components will kick in when prices drop. The Fragility Index is meant to be tracked and updated monthly.
3. Index Components in Charts
Here’s what goes into the index visualised. Perhaps that will help in building intuition about the computation.
I. Real Estate Overhang
The chart shows NBS’s “Floorspace under construction”.
Floorspace under construction divided by population is a point-in-time calculation for the end of November 2022 and uses the latest available population numbers (depending on the city, from 2018 to 2020).
II. Public Finances: Dependence on land sales for revenue
Here is a nice chart from Bloomberg showing local government fiscal deficits and dependence on land sales revenues. This data is by province. We extrapolate provincial averages onto the cities in our study.
Nice comment by Bloomberg on regional differences between LGFVs:
LGFVs in developed regions, such as Beijing, Shanghai and coastal areas provinces, tend to be more diversified and independent from local governments.
In many western and northern provinces, such as Guizhou, Guangxi and Gansu provinces -- where governments have wide budget deficits and rely heavily on land sales for revenue -- LGFVs tend to rely on land-related businesses and the local purse. The lower the level of government that an LGFV is affiliated with, the stronger the connection between the two.
— Bloomberg Intelligence: “CHINA INSIGHT: Housing Rout Spells LGFV Defaults, Not Crisis (2)”
III. House Prices: Crisis of collateral
This price chart is essentially a visualisation of drawdowns. Readings of 1 signify all-time highs. A reading of 0.90 indicates a drawdown of 10% from the peak. Some smaller provincial cities are hurting quite badly right now, with readings way below people’s expectations (not shown here).
4. Data File (.csv)
The file includes collated data points and calculations for 33 Chinese cities from publicly available sources in one neat table, ready for further analysis and tweaks.
Keep reading with a 7-day free trial
Subscribe to China Charts to keep reading this post and get 7 days of free access to the full post archives.