China’s coffee consumption growth slows

Luckin Coffee, the largest coffee chain in China, posted 1.2% company-owned same-store sales growth for 4q25, which is weak.

During earnings call, Luckin says China’s coffee market is still in a rapid growing phase – is that so?

I think there is still room, but current “coffee intensity” is already like in a mature stage.

Starbucks reward members (35mn) consume 3-5 cups per month.

Luckin members consumes close to 342mn cups per month in 2025 (4.1bn cups annually) and close to 100mn monthly transacting consumers, which translate to over 3 cups per customer per month approximately.

Why there is still room (but may be hard to penetrate)?

1) Luckin’s 450mn customer base means there could be 1.5x to 2x room in China, excluding children and elderly.

2) If Starbucks can get 16k stores in US, Luckin may get up to 4x of that which is over 60k stores in China, or 2x from current 30k stores.

3) Starbucks Reward member is an underestimation of active consumers of Starbucks, which could be like 70-90mn. 4x of that gives you 210-360mn potential which is 2-3x from current 100mn for Luckin.

Why hard to penetrate?

1) lower-hanging fruit / easy regions already have footprint.

2) China has more “layers” of consumers; thus hard to have one-size-fit-all offering. There are more competitions in China. Can’t handle price sensitive and premium customers together.

When gov requires buyback vs when gov buys you back

When gov requires buyback

When gov invest money as LP, the GP usually requires investees to buy back after say 5 years, with annual rate of up to 8%.

The equity investment is essentially a 5-year convertible loan that carries interest rate.

If things go well, gov should never lose money on these investments.

When gov buys you back

In rare cases, gov may promise to buy you back – e.g. Shenzhen gov promises to buyback 保障性配售住房, it applies an annual discount of 1% or more.

You will earn a negative interest rate annually.

What I don’t understand about robotaxi…

For the same destination, Baidu’s Apollo robotaxi in Shenzhen will charge RMB 125 (before coupon) vs RMB 40 on Didi express (affordable tier, before coupon) and RMB 50 for regular taxi.

Didi charges 125 before coupon
Didi Express charges 40 before coupon
Regular taxi charges 50

What’s also interesting is that Baidu’s robotaxi estimates that it will take 79 minutes!

Meanwhile Didi estimates it’s about 31 minutes, which is in-line with other map apps’ estimates.

Baidu robotaxi charges more than 2x the taxi price and takes more than 2x the time…

Well done.


Attaching the breakdown of Baidu robotaxi fare (before coupon)

HKD is USD or not?

HKD is pegged to USD. The fixed range is 1 USD = 7.75 – 7.85 HKD.

However, a weird thing is happening recently.

HIBOR (Hong Kong Interbank Offered Rate) is dropping dramatically.

For a USD-pegged currency, its borrowing rate normally increases or decreases with USD, and is similar to USD, otherwise arbitrage may happen.

Fed’s fund rate currently is 4.25% to 4.50%.

HIBOR 1-month dropped from 4% to below 1%. from May to Jun.

May 2nd, 1 Month HIBOR is 3.98363% vs. Jun 19th, it’s 0.53506%

A 350 bps drop in 1.5 month! Crazy world.


Three things may happen:

1/ Fed to decrease interest rate.

That will probably happen within the next 12month, but not in the July meeting.

Possible in Sep and/or Dec meeting.

2/ HK to limit money outflow.

It will become more like mainland.

But why?

HK doesn’t have the incentive to do this. And Beijing also wants a special international finance center.

Thus, unlikely.

However, there might be more mainland’s RMB converting into HKD – think about SOEs.

Since they are under the direction of Chinese gov, they won’t participate in every corner of the global capital market.

These HKD will behave like under the capital control.

3/ HKD-USD peg breaks

Too big a thing.

In short-term, seems unlikely.

In early May, HKMA needs to sell HKD and buy USD.

In early May, the exchange rate hit the strong-side CU of HK$7.75 to US$1 four times on three trading days, during both Hong Kong and non-Hong Kong trading hours. The HKMA sold HK$129.4 billion in exchange for US$16.7 billion in accordance with the LERS mechanism.

After Fed’s rate decision in June, HKMA expects carry trades may drive USD stronger against HKD and makes HIBOR higher.

If carry trades are to persist, the Hong Kong dollar exchange rate may weaken further, and may even trigger the weak-side Convertibility Undertaking. In such a case, the HKMA would then sell US dollars in exchange for Hong Kong dollars in accordance with the LERS, leading to a corresponding decline in the Aggregate Balance, hence driving Hong Kong dollar interbank rates to gradually increase.

Fourth year of decline in household’s new debt

Charting this…

From Jan to May, China’s household added 834.7 billion mid-to-long-term rmb debt.

This is only ~30% of 2021 level (near 3tn), and 40% of 2016 level (2tn rmb). This is the 4th year of decline of newly added debt; it’s only 3% below 2024 level though.

For the full year of 2016 and 2021, China’s household added 5.7 trillion and 6.1 trillion mid-to-long-term rmb debt.

Household’s mid-to-long-term debt is mostly for home buying.

 

Bizarre Numbers (4)

New home sales is ~60% of China’s home transaction, according to sqft from official stat. (Source: 全年新建商品房销售面积97385万平方米。二手房交易网签面积71812万平方米)

In the US, in 2024, ~700k new home sales vs. 4mn existing home sales. New home sales is ~15% of US home transactions.

60% vs 15%

Bizarre Numbers (2)

Detective Conan IP generated less than 1500mn rmb revenue in a year, with film being the largest contribution (~50%).

PopMart’s Labubu generated over 3000mn rmb revenue last year, while most people don’t know Labubu’s stories, its origin, its value, etc.

Bizarre Numbers

UPS and FedEx reported revenue per piece/package is ~$13-$14 (1q25).

SF Express, the most premium express in China, reported ASP of 15.5 rmb in 2024 and ASP of 14.6 rmb in 1q25.

Other express delivery companies in China reported ASP of 2.3 rmb (YTO) and 2.05 rmb (STO and Yunda) in 2024.


Meanwhile, ZTO+YTO+STO+Yuanda+SF -> 330k package per day!

FedEx +UPS is ~40k package per day.