Tag: Economy

China housing inventory – more than 2 years

If you look at US new home inventory, it shall take about 9 month to clear, as of Apr 2024 (see previous post).

What’s the number in China?

[The following numbers apply to 100 cities in China; source: https://mp.weixin.qq.com/s/BKnnF9QJTkYnAb1u0ip3Sw]

More than 2 years – 26.5 month to clear as of Apr 2024.

For lower tier cities (green line), the month to clear can be more than 30 months, or 2.5 years.

What’s would be the normalized rate of demand?

[Note the difference between 100 cities the the entire country]

Assuming future China has 7 million couples per year and new home demand is ~1/3 of that, it would be ~2.3mn new home demand.

Assuming new home is ~85 sqm, (2 x 42 sqm per person), it’s roughly 200mn sqm per year.

If every couple buys a new home, it would be ~600mn sqm.

There are other demand such as replacements. But just look at the US, which records less than 700k new home sales per year currently. China has 4x the population, so eventually new demand might come to 2.8mn, which translates to 280mn sqm if average is 100 mn per home.

This assumes new homes can sustain long enough.

There might be other renovation demand that can be recorded as new home sales, but real new demand could be lower than people think.

As of Apr 2024, reported inventory of new homes for 100 cities stood at ~500mn sqm.

In 2023, new home supply for 100 cities is ~291.1 mn sqm while sqm sold is ~316.5 mn.

In gov’s report, 2023 new residential sold was 948 mn sqm in 2023. So these 100 cities accounts for ~1/3 of total demand.

 

First yoy drop in US new home sales after 12 month of growth

US Apr new home sales (seasonally adjusted) dropped 7.7% yoy, reported last month.

The drop was long overdue, as interest rate (thus mortgage rate) has been high for quite some time.

The yoy strength from Apr 2023 – Mar 2024 was due to

A) the base effect

As most of 2022 was pretty weak after the war in Ukraine broke out. Fed also started the rate hiking cycle in Mar 2022.

B) the amazing AI-driven tech rally

Nasdaq rose over 60% from end of Sep 2022 to end of Mar 2024.

It reached all-time-high so even if you invested at the peak of 2021, you will still make money.

I bet many people enjoyed the wealth effect from stocks.

The base for the next 6 month (May 2023 – Oct 2023) looks pretty high, so yoy comparison should look weak, which is good if Fed is looking for “weak data”.

Meanwhile, inventory (new home for sale) has reached ATH since Feb 2024.

Previous high was 466k, seen in Oct 2022.

Time to clear the inventory is climbing in recent months. Now (Apr 2024) it’s 9.1 month vs. an average of 9.8 month between Jun 2022 to Oct 2022.

The median price for new homes sold is quite stable since Fed increased interest rate.

 

 

 

Living space in China vs. developed countries

According to the China Population Census Yearbook 2020, China has 462 million households, with average area of 111 sqm per household and 41.76 sqm per person. Average room per household is 3.2 and average room per person is 1.2.

Average space per capital of 41.76 sqm is not a small number. 

To compare, I asked ChatGPT for other countries’ numbers – 40-45 is quite the average. Only US, Canada and Australia are meaningfully above that number.

Just to double check in case ChatGPT is wrong, I looked up for German’s average living space, which was 46 sqm per person in 2018.

This number hasn’t changed much in the last decade or so. It was ~42 sqm per person in 2006 already, after decades of improvements (from 19 sqm per person in 1960).

Some further cross -check

Take a look at average home size in Europe, 100 sqm is already quite good.

Source: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8073340/pdf/ijerph-18-04278.pdf

Average room per person is 1.6 among EU countries.

 

Foreclosure during GFC

2008

A total of 861,664 families lost their homes to foreclosure last year, according to RealtyTrac.

A total of 3,157,806 foreclosure filings—default notices, auction sale notices and bank repossessions—were reported on 2,330,483 U.S. properties during the year, an 81 percent increase in total properties from 2007 and a 225 percent increase in total properties from 2006. The report also shows that 1.84 percent of all U.S. housing units (one in 54) received at least one foreclosure filing during the year, up from 1.03 percent in 2007.

2009

RealtyTrac, the online marketer of foreclosed homes, reported that one in 45 households — or 2,824,674 properties nationwide — were in default last year. That’s 21% more than in 2008, and more than double 2007’s total.

2010

RealtyTrac, a leading online marketplace for foreclosure properties, released its Year-End 2010 U.S. Foreclosure Market Report, which shows a total of 3,825,637 foreclosure filings—default notices, scheduled auctions and bank repossessions—were reported on a record 2,871,891 U.S. properties in 2010, an increase of nearly 2% from 2009 and an increase of 23% from 2008.

And a 14-year graph

https://www.attomdata.com/news/most-recent/2018-year-end-foreclosure-market-report/

March CPI 3.5%

March CPI is 3.5% for the past year and increased 0.4% MoM in March.

Higher than expected, but again, like the post last month, Truflation indicates <2% inflation.

Here is another discussion of “lagging” in shelter CPI, which was 0.4% in March

This lag occurs for a few reasons. First, the market indices capture rents of units currently on the market, not rents for units occupied by continuing renters, like the CPI does. Rents change when leases expire, which typically happens annually. In addition, landlords may be less likely to raise rents to market prices for continuing tenants, and so it might take even longer for rents on all units to catch up with rents charged to new tenants.

A new index was introduced to more closely track market prices of renting.

 

PingAn Trust?

A coin has two sides.

A balance sheet has two side.

If there is a hit on the asset side, there must be a hit somewhere else.

When properties are not selling, the property linked trust products would take a hit.

PingAn Trust is not responsible for those products’ underlying assets, which are troubled developers; it’s like an investment bank that packages products and sells those to “retail investors”.

Maybe many of those “retail investors” are not sophisticated enough.

In China, anyone with 500k rmb annual income can become a “qualified investor”, vs. $200k in the US, only ~1/3.

In China, if you have 5mn rmb financial assets (not net assets), or 3mn rmb net financial assets, you could become a “qualified investor”, vs. $1mn net worth (not counting primary residence) in the US.

Someone can borrow against the house and buys financial assets? Sure. He/she may get a >5mn loan from a tier-one city home (a 100+sqm home can be easily over 10mn rmb during good times).

It’s such a perfect designed chain of problems when property prices are coming down.

  • Underlying assets of trusts are worsening.
  • Trust investors can’t pay mortgages or home equity loans (or banks lower the est. value so can only take out less loan).
  • Fire sale from developers (trust products) & home owners (trust investors) shall further put pressure on the property market.

Is it possible that US inflation data is “tweaked”?

US reported Feb CPI (inflation) data: +0.4% mom and +3.2% yoy

Core CPI: +0.4% mom and +3.8% yoy

Slightly ahead, which makes it more difficult for Fed to cut rates.

However, is the data a bit inflated? I say possible.

Here is another inflation gauge by Truflation: only +1.64% yoy in March; below 2% in Feb

It’s true that US economy is stronger than most other places around the world.

However, it’s hard not to think that maybe there are some political reasons to “influence” CPI, which can influence Fed decision. After all, Fed is “data dependent”. See this post (Fed is bullied by the market – why?) as well.

Some elements from official CPI data can be more easily to “influence” e.g. used cars.

I will just stop there.

Why property price to income ratio this high?

China doesn’t have a nationwide recurrent property tax that applies to all residential.

  • in some cities with property tax experiment, usually certain area is exempted from taxation, e.g. Chongqing is 180 sqm

Hukou policy means residential property comes with other value. Not owning one will cause inconveniences: e.g. renting is not similar to ownership in education etc.

Renter’s right vs landlord is not like what usually seen in mature markets.

Most people don’t pay tax on rental income.

One-child policy plus decades of economic growth created massive buying power for this one-child generation. When you have 4 parents supporting the new couple to buy 1 home, that purchasing power is hugely inflated.

Local gov’s budget is partially financed by selling lands (use rights) for residential property development. High price -> higher budget.