I work in the power sector. I have lived in China for 15 years. I visit the province where the dam is several times a year to visit in-laws. And yet…until just a few months ago, I had never been to the dam.

Thankfully, I have now corrected that, as part of my recent trip to Yichang (catch up on that visit here).

The 3GD is a very cool site visit. Aside from being a massive hydroelectric generation and flood control facility, it’s also a national 5A tourist attraction with multiple onsite museums. With free entry and just 35 CNY for the (mandatory) shuttle bus, it’s also cheap, and total worth a visit if you happen to find yourself in central China. Here’s how I got here, what I saw, and what I thought

This essay was first posted to Twitter as a thread on 27 June 2026. It has been edited lightly for long-essay format.

The Three Gorges Dam is in Zigui County (秭归县), about 50km west of Yichang City proper. As a power generation facility, it boasts 32 main turbines at 700 MW, plus two smaller turbines for house load, for a total of 22,500 MW (22.5 GW) of installed capacity. It is the largest in the world.

Fully completed in 2012, it now produces roughly 100 TWh a year, equal to the annual power generation of a mid-sized country (but only ~1% of China's annual power use).

But power generation isn’t its only function. Actually, that’s not even its primary function, as I would learn during my visit.

As a water management facility, the Three Gorges Reservoir has a surface area of ~1000 square km and a holding capacity of ~39 cubic km of water. The reservoir behind the 181m dam stretches 600 km up the Yangtze River, beyond the Hubei provincial border and even extending back west to regions of Chongqing. We can only see one small part of it at Zigui County, i.e., the lake-like portion that formed immediately behind the dam.

In recognition of its dual purpose of flood control and power generation, its full proper name is “The Yangtze River Three Gorges Water Conservancy Project” 长江三峡水利枢纽工程, but in practice it’s almost always shortened to “The Three Gorges Project” (三峡工程) in Chinese.

Zigui County is administered by Yichang, but it’s still a good distance upriver. You can get there from Yichang via bus or self-driving, but if you have a Chinese driver’s license, I really recommend driving. There's a provincial highway (the S48) running between the two cities that'll get you there in less than an hour, but the slower National Road G348 has better views and also a few scenic viewpoints where you can park and take pictures. G348 is a fairly new road (opened in just 2023) that connects Wuhan to Dali, and the Yichang section already has a reputation for its great scenery, so that’s the route I decided to take.

Leaving Yichang on the G348, there is hardly any transition from the built urban landscape to natural forests and cliffs - it happens immediately. Yichang is already partially built on a hillside, so the road leaving the city turns quite smoothly into a cliffside road following the gorges rising along the river.

As I got on my way, it was 8am and I needed caffeine. Fortunately, at the first scenic turnoff I found a bright teal coffee van with a barista just setting up for the morning.

She told me she owns a brick and mortar coffee shop in Yichang too, but the coffee business is too competitive in the cities and big chains like Luckin are driving prices down too much, so she got this mobile coffee van to snag the tourist traffic on the G348 (notice she has the route drawn on the vehicle windows).

I wholly embrace this recent trend along China’s national roads. It’s easy to find coffee everywhere now - and it wasn’t always the case!

On the edge of a forested river gorge, we chatted about coffee market dynamics in China while she prepared my drink. The conversation turned to Chinese Yunnan coffee beans - becoming more popular every year among coffee lovers.

She told me it’s basically impossible to secure a stable supply of Yunnan beans these days unless you have a direct relationship with a plantation owner there, since they are so popular and demand is so high. The bigger chains are also starting to ride the trend of interest in Yunnan beans and their demand is creating scarcity in the market. Meanwhile, it takes at least 5 years to get new plantations producing, so the supply can’t possibly expand as rapidly as demand.

We laughed over the irony of her having no problem sourcing beans from global coffee regions like Yirgacheffe, Sumatra, and Panama, while struggling to get beans from China.

There was a bike packer with a tent set up behind her coffee van. They hadn’t gotten up yet, but I was impressed how far they had apparently ridden up these mountains on the previous day.

I thanked the barista for my coffee and headed on my way. A few minutes further down the road, I stopped at another scenic valley lookout which had a walkway extending over the lush gorge. It had just finished raining and the mists rolling gently over the forested mountains were perfectly atmospheric for the moment.

I bought some grilled baby potatoes tossed with seasonings from two elderly locals. They told me they've had good business selling snacks and trinkets at this lookout point since the G348 opened two years ago. Their village is down in the gorge I had just been overlooking.

After crossing the ridge, the road dumps out next to the Yangtze for the approach to Zigui County. The view across the opposite bank is dramatic, with terraced fields and houses rising up the side of choppy karst-like outcroppings.

Entering the Three Gorges area proper, I had to register with a police checkpoint, so if you go, make sure you bring your ID and the vehicle permit (I had to ask the rental company to send me a copy of this on WeChat, as it wasn't in the car already). It’s not military territory, but it's critical infrastructure, so you get checkpoints, armed police (武警), and controlled access. Once I registered, I got a yellow temporary certificate to put on my dashboard to clear the checkpoints. But the registration process itself was quick and painless. No photos of that part, for obvious reasons.

And then I was into the dam area. I parked the car, ignored the touts in the parking lot (they wanted to charge me extra for parking, or try to get me to buy a ticket to a free attraction) and headed straight into the tourism center. I bought an inexpensive (but mandatory) touring bus ticket and headed in.

There are also options to pair your visit with a river cruise, or a package ticket with other attractions in Zigui, but I stuck to just the dam this time.

The first scenic spot is an elevated lookout point called Tanzi Ridge Park (坛子岭公园) on an island separating the main dam/reservoir section and the ship lock section. It had just finished rainy, so it was still pretty misty, but it made the environment very atmospheric.

To the southwest, you can see the dam and the reservoir, as well as Zigui County and Quyuan Park on the opposite bank.

To the north, you can see over the ship lock mechanism, with allows for two-way traffic of 10k-ton vessels across five sequential lock chambers, resulting in an elevation differential of ~100m between the reservoir and the river continuing below. I'm not even really a boat guy, but I could have spent quite a while watching the vessels navigate the locks.

Local calligrapher and fourth-gen painter Xi Yongqin (习永钦) has a booth at the gift shop at Tanziling. He specializes in ink finger paintings (指墨画) of the Three Gorges Region.

“I’ve been painting for over 50 years, and painting the Three Gorges for over 30 years” he told me with pride.

He's fairly famous (he has a Baidu Baike page) and is semi-retired now, but still teaches and sells framed pieces while doing quick sketches for tourists. I bought one framed piece and he added another a quick sketch and some postcard calligraphy.

"You've been painting the same subject for 30 years? Don't do you get sick of it?" I asked.

"Only if it's not appreciated by anyone. As long as someone appreciates it, I feel good to paint it. I saw you liked it, so it gave me energy too." He was a super nice guy, happily posing for pictures and asking for a picture together as well.

The tour continued on foot down to the next sightseeing stop - roughly level with the dam itself. The best photo opportunities are here, and you get a chance to appreciate how truly massive the thing is. You can assess the reservoir’s water level with these concrete pillars. When I was there, it was very low, awaiting usage as a sponge to absorb water from upstream.

Next, it was back on the sightseeing bus to cross to the opposite bank to visit the Three Gorges Project museum.

The museum was quite good, tracing the history that led to the decision to build the dam. I learned that the flood-management capabilities of the dam and reservoir are actually the main intended functions of the dam, with power generation only secondary.

Generally, the worst flooding in the Yangtze River basin happened when there was a major storm or spring snowmelt that put the Yangtze at very high levels and left it unable to absorb any more water. Then, when the spring rains came, its tributaries would be unable to dump their water into the Yangtze and end up bursting their banks somewhere else.

The Yangtze’s main stem itself also could flood, but actually the most destructive flood events were more often associated with its tributaries that were unable to drain into the Yangtze when it was full itself.

There hasn’t been a major catastrophic flooding event in the Yangtze region since 1997, a year before the dam started construction, so it’s actually been operating in power-generation mode more often than it was expected to in its concept design. However, they have have been two flood seasons (in 2010 and 2020) that probably would have been much more severe were it not for the successful operation of the dam in water-retention mode.

According to the museum, it ended up paying for itself with power sales 7 years ahead of schedule compared to its original financial model, by the mid 2010s.

The museum also has cool exhibits about hydropower and floods in general and is a great visit for a certain type of nerd (me!)

Notably absent from this museum was any discussion of the social cost of building the dam, the inundation of riverside villages and towns, and the mandatory relocation of 1.3 million people. However, I would see this topic covered in great detail the next day, at the Three Gorges Resettlement Museum in Zigui County proper.

Finally, outside the museum is a park where you can check out some of the retired construction equipment used, including the massive dump trucks.

You also get a different view of the dam from the riverside parks on the south bank of the river, and again appreciate its mass. It’s nearly 6x wider than it is tall, which really contributions to the sensation of endlessness. Interestingly. the Three Gorges Dam is far from the tallest dam in the world - or even China. There are other dams in China that are more than 100m taller than this one. 3GD is notable because of the volume of its reservoir and the size of its generating capacity. The geography of the river gorges behind the dam maintain a perfect gentle sloping elevation for hundreds of kilometers, allowing for massive retention of water behind a dam that is much shorter than that of other massive hydropower projects.

After completing my site tour, I headed to my hotel in the new Zigui County (the original Zigui County is now underwater in the reservoir) and checked into my hotel on a cliff looking down over the rebuilt town, the reservoir, and the dam itself in the distance. This town was built into a cliffside to relocate the people who were affected by the reservoir without forcing them to move too far from their original location. Next time, I share some of my exploration around the town and talk about the resettlement.

In a strange way, what struck me most about visiting the Three Gorges Dam was how normal it felt.

If you spend enough time reading about it online, it starts to become less a piece of infrastructure and more a representation of an idea, or even the embodiment of an ideology. Depending on who is talking, it’s one of the greatest engineering achievements in history, an ecological travesty, a grim monument to reckless authoritarianism, a military vulnerability that’s going to drown millions downstream when it gets bombed, or an engineering disaster waiting to happen.

None of those idea-backed narratives were really present when I was standing it front of it, which was even kind of disorienting. The dam was just…there. Tourists are taking selfies and buying magnets at the gift shop. Employees are going to work. Boats are moving through the locks and the reservoir sits quietly behind the dam and it was all just normal. The concrete doesn’t know it’s the object of someone’s ideology.

I really enjoyed my visit, and I’d encourage everyone to go see this enormous thing too. But you very quickly normalize it, even in its enormity, and then it just…is. Reality is a lot less ideological than the conversations that are had about it. The world’s most famous hydroelectric project has basically completed the transition that all infrastructure should sooner or later: It has become boring.

And that’s good.

I preface my essay with this commentary to establish a shared understanding: when I say something in China is secretive, I really mean it. A secretive Chinese project in my eyes means it’s hard for anyone to find credible information about it, even in Chinese.

That is the case for the now under-construction hydropower mega-project on the Yarlung Tsangpo River in eastern Tibet/Xizang, also called the Medog Hydropower Station. Formally under construction since the middle of 2025, this behemoth is budgeted at ~1.2 trillion CNY ($160-170 billion USD) and expected to eventually boast 60GW of electricity generation capacity. That’s more than the current installed power generation capacity of Sweden or Argentina, and roughly 3x the generation capacity of the Three Gorges Dam, which is already the largest hydropower facility in the world. Yet for all the fanfare about the project’s ambitious scale, the public knows very little about what kind of facility it even is, never mind details like the construction milestones, power export destinations, effects on the surrounding environment, or expected impact on human populations.

Although information transparency is never going to be the Chinese government’s strongest suit, it’s still uncommon for the public to lack even basic details about the engineering concept for a project like this. Will there be a dam? How high will it be? Where will it be? Is there a reservoir? How large will THAT be? Even today, these seemingly mundane details about the Yarlung Tsangpo project aren’t publicly stated anywhere, contrasting even with the history of other large hydropower projects in China. Understandably, the lack of information has also created worries among China’s neighbors.

Natural abhors a vacuum, and this applies to information too. In the absence of transparency, conspiracy theories and insinuations swirl. Unfortunately, rarely is the information that fills voids of particularly high quality. With Himalayan mountains and insinuations about Chinese infrastructure development plastered all over the news thanks to the recent mudslide/flooding tragedy at the Nepal/Tibet border, despite being 1000+ km away, I suppose it’s as good a time as any to assemble all currently known or inferable information about the Yarlung Tsangpo hydropower project, in English, in to one place, in FAQ format. I’m not typically much of an OSINT guy, but I do like big infrastructure projects, so I decided to make an exception this time.

I envision this as being a kind of long-term project, with this essay updated as often as there’s new, interesting information to share, but we’ll see how committed I’ll be.

What is the Yarlung Tsangpo?

The Yarlung Tsangpo is the 8th-longest river in China, the highest major river in the world, and the longest in Tibet. Its headwaters arise from the Angsi Glacier on the Ganglung Kangri Mountain, not far from Mt. Kailash. It flows steadily east for 2000km, growing in size along the way, before making a sudden change in direction to navigate around Namcha Barwa Peak and head south, where it crosses the Himalayas, leaves China and enters India’s Arunchal Pradesh state.

In Arunchal Pradesh, it is briefly called the Siang River before it enters Assam state, where it picks up its most well-known name, the Brahmaputra. From there, it crosses northeastern India in a southwesterly direction and enters Bangladesh, where it becomes the Jamuna River and again turns south. Shortly thereafter, it merges with the Ganges (aka Padma) east of Dhaka and then the Meghna River, before draining into the Bay of Bengal, ~2900 km downstream from the Angsi Glacier.

In the Tibetan language, Tsangpo means “great river” and Yarlung is a reference to the Yarlung Valley, regarded as the cradle of Tibetan civilization.

The Mandarin Chinese name is 雅鲁藏布, or “yalu zangbu”, a transliteration of the Tibetan name. For clarity, the word for “river” (江) is usually affixed to the end of the name, since the typical Chinese speaker wouldn’t recognize zangbu already means river, so they end up calling it the Yarlung River River (this happens all the time with river and mountain name translations across languages). It’s often shortened to just 雅江 (Ya River) in colloquial speech.

Where is the Hydropower Project?

There are several small-to-medium hydropower projects further upstream on the Yarlung Tsangpo, but the one that we really care about is quite far downstream on the eastern stretch of the river where it does that big loop (the Great Bend) around Namcha Barwa Peak (南迦巴瓦峰) and then heads south.

Administratively, this area is part of Tibet’s Nyingchi (林芝) Prefecture, with its prefecture seat indicated by a small dot on the map above.

That dot also corresponds to Bayi District on the map below. I’ve traced out the Yarlung in blue, showing how it enters Nyingchi in Nang County, threads its way through the center of the administrative area of Mainling City (米林市) in a east-northeast direction and then makes that big loop down into Medog County (墨脱县) and eventually India.

Don’t be misled by the existence of “districts”, “counties”, and “cities” here. This region is extremely remote, which you can tell clearly from the satellite view. Actually, it’s essentially all mountains and glaciers, with most of the area surrounding the mountain being pristine alpine wilderness.

The Great Bend of the Yarlung Tsangpo circumnavigates the 7,782m Namcha Barwa Peak (labeled Namjagbarwa Peak on the map), the 27th-tallest mountain in the world. From 1967-1992, it was the tallest unclimbed mountain the world. Since its first ascent by a joint Sino-Japanese team, it retains the distinction of being the tallest peak to only have been summitted once. It is the tallest of the east Himalayas, the third most-prominent, and the world’s easternmost 7000m+ peak.

As the river meanders on its ~200km sightseeing loop around the mountain’s ravines, it does something very interesting: it loses more than two kilometers of of vertical elevation. By the time it reaches the bottom of the Great Bend, it has succeeded in crossing the Himalayas, descending from an elevation of ~2950m in Mainling City to just 600m in Medog County.

This 2350m drop between the upper and lower segments of the river’s great bend around the mountain is the basis of the concept design for the hydropower facility. Hydraulic engineers know what a river’s huge elevation loss over a short distance means: a lot of potential energy to be harnessed by a hydroelectric facility.

What Kind of Hydropower Project is It?

To oversee the project, Beijing created China Yajiang Group (中国雅江集团) in 2025, making it China’s youngest centrally administered state-owned enterprise (央企). Construction and engineering work is expected to be carried out by China's major hydropower and infrastructure firms, with PowerChina the likely lead contractor.

Officially, the Medog Hydropower Plant is a “run-of-river water diversion” facility. The project will bore a long underground tunnel as a headrace between the upper section of the Great Bend, near Mainling City’s Pai Town (派镇) at 2950 meters, and the lower section of the Great Bend, with Medog County’s Xirang Village (西让村) at ~600 meters.

Water from the river at the top of the great bend will be diverted into the ~40km tunnel, where it will flow downhill at a continuous grade of ~5-6%. This will produce an enormous hydraulic head, sufficient to drive five separate powerhouses along the way, each one at ~12GW, for a total of 60GW. At the bottom of the tunnel, the water will rejoin the original course of the river. Chinese planners have described this approach as “zero net consumption of water”, with all water diverted at Pai Town returned to the river’s course at Xirang Village before it flows into India.

This tunnel diversion concept is not new. Hydropower engineers have employed similar principles since the earliest generating schemes at Niagara Falls more than a century ago, diverting water through artificial channels to take advantage of elevation differences before returning it to the river. China itself has already demonstrated a modern version of this approach at the Jinping II hydropower project on the Yalong River, which uses a 16km tunnel to bypass 150km of the Yalong River’s natural course. What makes the Yarlung Tsangpo project remarkable is the unprecedented scale.

When the project is completed (est. 2033), the generated power will feed into a network of UHVDC lines and transmitted across the country. As of 2026, one dedicated export corridor is already clearly identified: the 800kV Tibet-Greater Bay Area UHVDC line, rated at 10 GW. However, this line alone can only transmit a fraction of the project’s expected annual output of ~300 TWh. Additional UHVDC export corridors will certainly be required, but are not yet known.

Crucially, the water diversion scheme in Pai town will NOT use a high dam. It can’t be a high dam, because there’s no room for a reservoir behind it. From the image above, you can see Nyingchi’s main airport in Mainling City lies just 30km upstream on the riverside only ~20-30m above the water level. There’s only a 10m elevation drop between the river bed at the airport and at Pai Town and the airport only has a small clearance over the river, so any reservoir deeper than 20m or so would endanger the airport.

Local media in Mainling City mention a “low dam” (低坝) at the Pai Town site, with “backwater extending to the airport”. This updates earlier thinking on this structure, when there was confusion about whether it would be an actual dam blocking the river, or more like a weir to direct the water into the diversion tunnel. It is now confirmed to be an actual dam - a small one, with a small reservoir, suitable for hourly and daily regulation of the water supply to the hydroelectric facility.

But if there’s no large dam, and no large reservoir, how can the project possibly serve as a stable and reliable producer of electricity on an annual basis across seasons? A 20m reservoir created by a small dam isn’t going to get you far.

That’s where one of the biggest confusions of this project arises. Because there IS a high dam for this project; it’s just not at Pai Town.

Where’s the Damn Dam?

This has been the point of greatest confusion - whether the Medog Hydropower Plant will have a high dam, what kind, how tall, and where it will be. Nearly every Chinese online post about dam structures for this project over the last 2 years has dissolved into arguments and name-calling over this point. Fortunately for you, my reader, out of all the people spending hours of their lives researching and debating this topic, I have the least regard for the value of an hour. And thus, I can present to you the most accurate up-to-date perspective on the question of this damn dam.

The high dam associated with the Medog Hydropower plant will be located over 100km upstream to the west of Pai Town, past the airport and even Mainling City itself, in Yusong Village (玉松村) of Lilong Township (里龙乡).

Lilong Township is administered by Mainling City, and so this dam will be called the Mainling Dam (米林大坝) or the Milin Dam, if using the Chinese name for the city. I will use Mainling Dam for this essay, although the names are interchangeable.

The annotated map below illustrates the relative location of the Mainling high dam, the low dam/tunnel intake point at Pai Town, and the exit point at Xirang Village on the other side of the Great Bend.

This location appears to have been chosen for a high dam because it offers more space for upstream backwater accumulation without submerging larger settlements. While there are a few small towns and villages along the river further west upstream of Yusong Village, there’s certainly nothing as critical as an airport.

This location for the dam is aligned with the best piece of official evidence we have: a CCTV Xinwen Lianbo broadcast from August 2025 covering the official launch ceremony for the construction project, which was held near the dam's planned construction site and attended by Premier Li Qiang.

In the image below, taken by Zhihu user 梵华, an black arrow in the center of the image points to the confluence of the Nyang River and Yarlung Tsangpo, just downstream of Nyingchi. Upstream of that (to the west) I’ve drawn my own circle around where you can see a faint red line drawn across the river, just after where it looks like a small tributary joins the river.

In reality, this tributary isn’t a permanent river so much as a small stream coming down the ravine, carrying seasonal glacial melt. It’s not even named on maps. This is more apparent when looking at a Google Earth close-up of this location.

From both the satellite view and the August 2025 Xinwen Lianbo broadcast, we can see the land next to the river is fairly flat here. The launch ceremony appeared to take place directly in the riverside fields of Yusong Village, just across from where the small stream emerges from the ravine and enters the river, as Zhihu user 梵华 identified below by cross-matching the broadcast view to satellite images:

According to these images, the actual location of the dam would be immediately after the confluence, with the reservoir stretching back up the main channel of the river, as well as the side ravine. This is my speculation here, of course but perhaps it would look something like this:

In the image, I’ve drawn my speculated location for the dam in red and illustrated in blue-green where the reservoir’s backwater would begin, inundating both the Yarlung Tsangpo’s valley to the west, as well as the ravine to the north.

If my assumptions are correct, then the first settlements that will be submerged behind the reservoir are Deji New Village and Yusong Village on the south bank, and Tashi Rabten Township (扎西绕登乡) on the north bank. Any smaller villages along the banks of the river upstream will also be inundated.

So, How Large is the Dam?

The Mainling dam's precise technical parameters like crest elevation, normal pool level, and reservoir depth have not been officially disclosed. This is where things get tricky, and where my view differs from Chinese OSINT analysis.

The height of the dam determines the normal water level of the reservoir, as well as how far upstream will be inundated. Here, Chinese netizens have identified at least one hard constraint: the Lhasa-Nyingchi railway, which makes its way through the mountains north of the river at an altitude of 3070m. Also, recent procurement records for road construction in the area have indicated that the roads must go no lower than 3070m, seemingly a floor level related to the hydropower project. I agree with these figures as the peak possible height for the reservoir.

The river’s water level at Yusong Village is ~2944m according to Google Earth satellite data, which would give the dam’s designers ~130 meters of reservoir depth to work with (putting aside that there’s a margin of error with satellite elevation data in mountainous regions that we just have to accept and work with). Chinese OSINT users thus assumed the Mainling Reservoir would be 130m at its deepest point, with the dam crest ~150m high.

I think these OSINT assumptions have some problems. To stress-test them, I introduce an additional source of evidence that earlier analyses either did not consider or did not yet have: local resettlement records. For several years now, well before the official construction launch, local officials have been working to relocate villages that will eventually be submerged by the reservoir, and they usually make a point of celebrating when resettlement work has completed. Local media also likes to document when officials visit the resettlement zones to inspect and oversee work.

So, I searched local government and state media reporting for references to reservoir migrants, completed village relocations, new settlement sites and replacement town centers. I then compared that paper trail against the potential inundation zones indicated by the terrain data to determine the likely upstream extent of the planned reservoir.

My evidence trail:

First, the administrative center of Tashi Rabten Township just north of the planned dam is clearly moving. As early as July 2025, Tibet’s Party secretary inspected the new “Tashi Rabten town construction site”, where officials briefed him on its overall planning, infrastructure and supporting facilities. The visit was explicitly framed as settlement construction in service of a “national major national project”.

Furthermore, in June 2026, the Tashi Rabten government announced that it was vacating its existing premises and temporarily moving its government offices and public service center to Milin City. State media has also reported on the successful relocation of Kangsa Village (康萨村) and Senbo Village (森波村) upstream on the north bank, both part of Tashi Rabten.

We also have good evidence for the relocation of Lilong Township (里龙乡) further upstream. This emerged in April 2026, when a local government Wechat account covered Milin Party Secretary Yan Shihui’s inspection of the “new Lilong township site”. The report noted the need to “carefully and patiently” manage work with “relocated residents” and described a visit to the “resettlement site of Deji New Village”, including temporary housing and a kindergarten, strongly suggesting that relocation of at least part of the existing Lilong Township is already underway.

From these clues, we know the reservoir will at least stretch upstream to the point where it will inundate the Lilong Township seat.

In the satellite view below, which is oriented facing east, we can see the current seat of Lilong Township at 2983m is just ~23 meters higher than the river’s level at 2960m. This area will undoubtedly be underwater. Notably, some of Lilong Township’s constituent villages on the cliffsides of the south bank sit much higher than this, and may be unaffected by the reservoir.

The really crucial evidence comes at the next town along, about 45km upstream from the hypothetical dam site. This next town is Wolong Town (卧龙镇), the last town belonging to Mainling City heading west before entering Nang County.

There’s currently no indication that Wolong Town is preparing to be submerged.

Perched on cliffs high above the river below, Wolong sits at a lofty 3050-3080m of elevation, some ~80-110m above the river’s level at 2970m. Moving upstream from Lilong, the riverside terrain gains elevation quite rapidly, while the elevation of the river itself rises much more slowly, forming a dramatic gorge.

Now, not only does Wolong Town show no signs of being resettled, but there are even news reports and procurement documents saying Maicun Village (麦村村) in Wolong has been identified as a potential site for a new resettlement community related to the construction of the hydropower station. In 2024, project tenders were issued for geological hazard assessments relating to the proposed Maicun resettlement, including for rockfalls coming from the steep mountainous gully located directly above the proposed site in Maicun.

These documents suggest that Maicun is being evaluated as a location to receive relocated populations, NOT slated for relocation itself. However, the documents do not specify where the residents assigned to this settlement will come from.

Maicun Village broadly sits between 3100m and 3000m, with the highest point on the terraced fields behind the village rising to 3140m.

So, if the Wolong Town seat at 3050m shows no signs of being relocated, and Maicun Village at 3000-3100m is being considered as a resettlement destination, we have a fairly revealing indicator for the maximum height reservoir is expected to reach, and it seems to be a good bit lower than what was believed previously.

Contrasting the earlier conclusions of the Chinese OSINT writers, the normal water level of the reservoir cannot be as high as 3070m, else Wolong at 3050m would be submerged and Maicun would be unsuitable as a resettlement destination. This still doesn’t give us a definitive reservoir level, but if we use Wolong as the limit and assume a 10m safety margin, then it would imply the reservoir level can rise to no more than ~3040m. Based on the river level at the dam site sitting at 2970m, this implies a maximum reservoir depth ~70m, and thus a dam height in the range of 90-120m, significantly lower than the 150m figure circulating in OSINT analyses.

Of course, even if Wolong Town is safe on its cliffside, the reservoir will still continue on quite a ways further upstream until the elevation of the river bed itself meets the normal reservoir height. The next county over - Nang County - notes on its government page that the lowest point in the county is the river surface at its border with Mainling City at 3016m, indicating the reservoir must extend into Nang County too. If we assume the reservoir’s normal water level is 3040m, then it will not meet the elevation of the riverbed until another 70km upstream from Wolong, in Nang County’s Dongga Town (东嘎镇). Many smaller villages upstream along the river banks between Wolong and Dongga that are not perched safely on a cliff like Wolong Town will likely also be submerged.

This is all reflected on my map below, which marks the locations of the affected towns and townships, along with the hypothetical reservoir area in blue-green. I have not marked the extent of the other reservoir created by the low dam at Pai Town as it will seemingly be quite modest, but will apparently stretch back to the airport.

How Many People Will be Displaced?

There’s not yet any official estimate from government authorities on how many people will need to be resettled or relocated. However, the number will necessarily be much lower than a project like the Three Gorges Dam, which saw 1.3 million people displaced by the massive reservoir that formed in the densely-populated regions behind the dam. In Tibet, the populations are much smaller and much more dispersed.

According to records from local gazetteers, Tashi Rabten has a total population of just 3552 people. Lilong has another 2038, and Wolong has 4258. Meanwhile, Dongga Town has just 2489. There’s also likely to be some displacement downstream in Pai Town to make way for the dam structures in that area, although that’s distinct from reservoir inundation. But even if EVERY person in each of the affected townships had to be resettled to make way for the project, the entire resettlement effort would be on the order of 12-15k people. That’s likely a high estimate, since some of those villages far above the riverbed (like the seat of Wolong Town itself) will not be displaced at all. Certainly a very different situation from the Three Gorges Dam.

In addition to settlements, the project will also affect agricultural land. In some cases, a village may not require full resettlement, but part of its land must be requisitioned, such as riverside terraces or cropland located near the future reservoir shoreline. In 2024, Milin City published a schedule of compensation rates for agricultural land requisitioned by the state.

The document itself is a general compensation schedule and does not explicitly connect the villages to the hydropower project. However, it’s likely no coincidence that virtually all the villages identified in the schedule are located along the river corridor, near either the Pai low dam or the Mainling high dam reservoir. Back in October 2025, Milin media platform cited Tashi Rabten Township head Awang Banjiu saying that resettlement compensation funds of 400+ million CNY had already been fully disbursed, which covered both the hydropower project as well as legacy compensation issues from railway construction.

What are the Biodiversity and Ecology Concerns?

Both the high dam and the low dam will be subject to mandatory “minimum ecological flow” for water release targets, which are defined at the national level. For hydropower projects in China, this minimum flow level is typically set at “not less than 10% of the river’s average flow”, although it can vary by province. For the Pai low dam, the implication is that once the mandatory 10% ecological release is satisfied, the remaining 90% of the river’s flow is, in principle, available to be directed into the diversion tunnel. This will make the Pai-to-Medog stretch of the river (i.e. the Great Bend section) a greatly reduced flow, although not an entirely dry one. The flow in this section will also be supplemented by convergence with other rivers and streams along the way, including the Parlung Tsangpo (帕隆藏布), the Yarlung Tsangpo’s largest tributary.

This same stretch of the Great Bend is within the Yarlung Tsangpo Grand Canyon National Nature Reserve, the only place on Earth to host four big cat species (Bengal tiger, snow leopard, clouded leopard, and leopard) in a single river valley. It is also home to China’s only confirmed wild Bengal tiger population, which were only captured on camera in this area for the first time in 2019.

What the proposed reduction of the river to ~10% of its natural flow could mean for these cats is a matter of ecological inference. The reduced-flow sections of the river overlap with the mid-elevation forest belt that ecological surveys show is exactly the habitat of the big cats, including golden cats and the extremely elusive clouded leopard. A major reduction in flow could alter the microclimate, reducing the clustering of prey and degrading the dense forest cover these species depend on. Whether the mandatory ecological release will be sufficient to maintain the canyon's habitat’s integrity is unclear, at least publicly, as the EIA has not been released.

The Chinese Felid Conservation Alliance (CFCA) has written extensively about these issues, warning that the project will cause “habitat loss and fragmentation” in the Great Bend’s forests through construction of tunnels, access roads, and a huge population influx (expected 100,000+ project workers vs. Medog’s 14,934 permanent residents), with “irreversible impacts on the ecology.” They have also expressed concern over the lack of transparency, mentioning they requested to review the EIA and were rejected on the basis that the project is classified as a state secret.

At the upstream Mainling dam, a different challenge is posed by fish. The Yarlung Tsangpo is home to a number of endemic high-altitude fish species - including some nationally protected endangered species - whose life cycles depend on moving up and down the river to reach spawning grounds. Dams interrupt those migrations, forcing developers to find ways to reconnect the habitats.

At the 100m+ Zangmu Dam on the same river higher upstream, engineers were forced to build a ~3.5km fish passage system (a “fish ladder”) to allow native fish to bypass a dam far higher than what fish ladders were traditionally designed to handle.

Monitoring data to date suggests fish DO use the facility at the Zangmu dam, and the dam operator has supplemented it with hatchery breeding, fish stocking and other habitat protection program. But fish ladders are ultimately an imperfect solution. The fish can only benefit from the ladder if they find the entrance, successfully navigate the passage and continue their migration on the other side. The Mainling dam will almost certainly incorporate similar measures, but the long-term question is still unclear as to how effectively they will be at preserving the river’s unique fish populations.

Is the Region Seismically Active?

Yes, extremely. The Yarlung Tsangpo Great Bend sits in one of the most seismically active regions on Earth. The area lies at the collision point of multiple tectonic plates with active faults and rapidly forming/deforming mountains. Since 1900, the Nyingchi region has recorded more than one hundred earthquakes of magnitude 5+, including multiple quakes of M7 or higher. The most catastrophic was the 1950 Assam-Tibet earthquake, which reached M8.7 and remains the largest earthquake ever recorded in China and one of the strongest earthquakes in modern history. More recently, the 2017 Mainling M6.9 earthquake hit directly within the project area, with intensity maps placing Pai Town near the core intensity zone. To put it bluntly, the Yarlung Tsangpo dams and diversion tunnel are being built in an environment where major earthquakes are a predictably recurring geological feature.

Chinese planners are well aware of this problem. The project developed has noted that key project structures are being designed for seismic fortification intensities of 9 to 10 degrees and are expected to withstand earthquakes rated to M8.5. Besides the dams, the project’s underground tunnel system will cross several major fault zones and dozens of subsidiary faults, requiring extensive reinforcement and continuous monitoring. However, the detailed design considerations remain classified, so it is impossible to assess how these risks are being managed.

What Do China’s Downstream Neighbors Think About the Project?

The Yarlung Tsangpo enters India through Arunachal Pradesh, flows across Assam as the Brahmaputra, and then crosses Bangladesh before reaching the Bay of Bengal. Unsurprisingly, the hydropower project has attracted significant attention downstream. The most common concerns include the possibility that China could reduce water flows, manipulate releases for strategic purposes, disrupt ecosystems and sediment transport, or create flood risks. However, many of these concerns are discussed only in generalized and highly politicized terms, making it difficult to distinguish between plausible risks and more broad speculation.

Some concerns appear more credible than others. For instance, claims that China could simply “turn off” the river do not align with the project’s apparent purpose as a hydropower facility, which requires the water to flow, and returns water to the river after generating electricity. But more plausible concerns involve things like coordinating the timing of water flows, since India has its own substantial hydropower ambitions for the river. Bangladesh has also raised questions about sediment transport, as reservoirs are likely to trap some sediment that would otherwise flow downstream. This is important because sediment deposition plays an important role in sustaining the river delta and combating erosion. More broadly, the greatest uncertainty seems to stem not from the project itself, but from the very limited amount of public information available about its design and operation.

Despite the prominence of these concerns in public debate, the official positions of India and Bangladesh have generally been more measured than media coverage might suggest. Bangladesh’s Foreign Adviser Touhid Hossain stated in August 2025 that Bangladesh saw no “reason for concern,” while also pragmatically acknowledging the limits of their influence, remarking: “We cannot stop it… We have to see that we are not harmed.” This pragmatism + caution helps explain Bangladesh’s continued emphasis on transparency and access to hydrological data. India’s official stance has likewise focused primarily on practical issues: obtaining technical information, maintaining dialogue, understanding the project’s engineering details, and ensuring that existing mechanisms for transboundary river cooperation remain functional - not calling for the project to be stopped.

And contrary to some media narratives, there ARE active channels for coordination on these topics between the neighbors. Since 2006, China and India have maintained an Expert Level Mechanism on Trans-Border Rivers, a forum for exchanging hydrological data, discussing flood management and addressing river-related issues like the dam. Both governments continue to reference this channel, and recent reports indicate that meetings and data-sharing arrangements via this mechanism are still active, although some data sharing was suspended in recent years following border tensions in 2022.

To editorialize a bit here, the key disagreement may therefore be less about the absence of communication, and more about the format in which it takes place. China traditionally prefers technocratic exchanges conducted via expert and bureaucratic channels, while many Indian politicians, commentators and analysts appear to anticipate a greater degree of public transparency around the project’s design and operation. The challenge for both sides thus would be determining whether a mechanism originally designed for hydrological cooperation like the Expert Level Mechanism on Trans-Border Rivers is a actually sufficient platform for a project of such unprecedented scale. But such matters are probably above my pay grade.

Conclusion (for now)

Well, there we go. All the main FAQs I think there are to know about this controversial and notably mysterious project so far. Hopefully it’s a little less mysterious now than it was before.

I recognize that for large parts of the essay, all I did was replace the speculative OSINT of others with my own, slightly less-speculative OSINT. But at least it’s as up-to-date as possible. And for the other sections, hopefully you’ve found some value in my efforts to summarize what’s up with this project as reasonably as possible. But do let me know if I’ve missed any key FAQs that I could address in future edits.

Thanks for reading!

I've visited central China’s Hubei Province many times. I am, after all, a “Hubei son-in-law” (湖北女婿), as locals would say. Until recently, however my experience with Hubei had been limited almost exclusively to eastern Hubei, in the greater “Wu-E-Huang-Huang” region (referring to the capital city Wuhan and the nearby cities of Ezhou, Huanggang, and Huangshi).

My prior writing has featured Ezhou quite a lot in particular, as I am most specifically an “Ezhou son-in-law”. But as you can see from the map below, the area that I was most familiar with was really just a small corner of a much larger province that was much more mysterious to me.

But on my most recent trip to visit the Three Gorges Dam, I finally made it further west to the city of Yichang (宜昌) which has been on my target list for a long time.

To reach Yichang from Wuhan, you must cross the Jianghan Plain, which lies directly directly west of Wuhan. It’s a massive stretch of farming flatlands - like Kansas with more rice paddies. It has been one of China's key breadbasket regions for thousands of years. In June, the fields were bursting with luxuriant green crops, dripping in the plum rains, a lush contrast to the modest grey concrete village houses squatting in small clusters across the landscape. Of course, the high-speed rail makes short work of the traverse, covering it in under an hour.

The Han River winds east-west across this plain, moving roughly parallel with the Yangtze 90km to its south before they finally converge in Wuhan proper. The train route crosses it several times as we head west. The Jianghan Plain is a floodplain, and before the Three Gorges Dam was built, the Yangtze River basin's most destructive floods were often associated with this region. When rivers here burst their banks, they can submerge fields for hundreds of square kilometers.

Yichang itself isn’t so famous as a tourism destination, but I’d long wanted to come to visit what is undoubtedly the most famous modern megaproject in China: the Three Gorges Dam. The 3GD is not in Yichang proper, but in Zigui County, 50km further west up the Yangtze into mountain country. I would visit the next day, but first I would spend a night in Yichang.

Yichang City lies along the Yangtze on the very western edge of the Jianghan Plain, the last flatland outpost of central Hubei as you head west, before the terrain starts to rise into the dramatic river gorges and cliffs that define Hubei’s real mountain regions like Enshi and Shiyan.

The scenic areas of western Hubei aren't as well-known as those of western Hunan (like the world-famous Zhangjiajie and its “Avatar Mountains”) but are no less stunning. They’re just a few hundred km apart too, so they also share the same geology, climate, and ethnic minority cultures.

And as I’ve mentioned before, I’m a sucker for Chinese mountain vistas, so my camera was having a good time…

Hubei is a classic archetype of a central Chinese province, economically and culturally dominated by its capital, with broadly weaker secondary cities. This model applies to province like Hubei, Hunan, Henan, Shaanxi, Shanxi, and Jiangxi, to name a few.

But if Hubei had a second economic pole beyond Wuhan, Yichang would certainly be in the conversation. In 2025, Yicai classified Yichang as a 3rd Tier city (ranking #88 out of 337 cities tracked across China). [What’s a Tier? What is this Yicai system anyway? I break it all down in this post.]

With a total administrative population of 4m and an urban core of ~2.5m, Yichang is not massive, but it still has the key characteristics that make it a Real City - and not just an administrative region. I think this is a key characteristic of all the T3 cities in Yicai’s ranking - that they are “Real Cities” (tm).

So what do I mean by “Real City”? Let's talk about the T3 designation for a second.

China's T3 cities are a bit tricky to define for me, perhaps because I don't visit them very often. But they ARE different from Tier 2 or Tier 4 cities. You can feel why T2 cities are T2 cities when you visit, because they typically have some built-up downtown areas indistinguishable from T1 cities, substantial skylines, multiple large anchor industries, concert halls, museums, universities, hospitals, etc.

I’m talking about places like Fuzhou, Dalian, Nanchang, or Xuzhou. If you’re in the right part of town in a T2 city and squint a bit, you might say: “hmm okay, this could be somewhere in urban Beijing”.

By contrast, Tier 4 cities are mostly just huge population sinks. Half their population resides in remote counties spread out over a huge prefectural area. They usually don't have much of a skyline in the “city proper” because they don't have the type of industries that would fill out a high rise skyline (like finance, insurance, real estate). They have schools, but more often they’re just teacher-training schools and vocational institutes. They'll have a few good hospitals and an economic development zone with a handful of pillar industries, but most employment will be in service industries, traditional manufacturing, or the public sector. Everyone you meet there is basically from there, or one city over. And you’d hardly ever find a part of town that you could confuse for a Tier 1 city when you squint. Linfen, which I visited last year, is a good example of this.

So now that I’ve defined the bookends, perhaps we have a better way to define a T3… cities like Ganzhou, Zhenjiang, Yichang, or Guilin?

In my view, T3 is the lowest tier of city that still functions as a true regional hub of urbanity, a place that can absorb populations, services, and economic activity from other prefectures.

This is at least true for the top half of Yicai’s T3 list. There are 70 T3 cities, and when we get to the bottom half, I’m not sure there’s really much distinguishing them from the T4 cities. In other words, I’d say the Yicai T3 bucket is a little too large to be rigorously descriptive, because there’s too much of a gap between the cities in the top half and the bottom half. This is a complaint I will be sure to lodge with the appropriate departments at Yicai, when I get the chance. :D

Putting that aside, my conclusion is that I should define T3 Chinese cities more by their proximity to T2 cities than by their proximity to T4 cities.

Yichang demonstrates this logic well, sitting comfortably in the top half of Tier 3.

1. It has a genuine industrial economy, dominated by power (Three Gorges) and chemicals (it sits on one of China's largest phosphate bases). But it also has a healthy equipment manufacturing sector (especially hydropower-related) and Yantgze shipping/logistics sector. T4 cities rarely if ever have this.

2. It has a comprehensive 4-year university (China Three Gorges University) offering a broad range of disciplines but especially known for strength in electrical and hydropower engineering. Not all T3 cities have a comprehensive university, but the strong ones do. CTGU isn’t elite per se, but it’s a real regional institution, with credible research output and industry ties, putting it well above the types of vocational schools you’ll find in T4s.

   

3. It has seven 3A-ranked (三甲, which is the highest tier) hospitals, including a major hospital ranked in the national top‑150 (Yichang People’s Central Hospital). It is basically the primary healthcare node for the 1000km gap between Chongqing and Wuhan. It serves not just the local city but a wider hinterland of western Hubei and rural Chongqing.

4. It is a human capital hub. Despite being far smaller than T1/2 cities, it still attracts net worker inflows. The service workers I spoke to were from all over Hubei and even neighboring Chongqing. So it’s a place people move to for better urban opportunities - not a place that people leave for better urban opportunities.

So, that’s Yichang. Nothing amazing, nothing standing out really incredibly so, but a real city nonetheless, and a credible second economic pole for Hubei (contested with Xiangyang). I didn’t find it unpleasant. It's a perfectly functional small city - a “real” one - and a clear step up (or two) from a place like Ezhou, where I had just spent the three days prior to arriving in Yichang.

I’ve only visited Ezhou once or twice a year for the last 7-8 years (for holidays) and I can already navigate it without GPS, and have already tried what feels like almost every restaurant in town. No way I could do that in Yichang.

Now remember, I only spent one day and one night in Yichang on my way out to Three Gorges, so I’m not claiming to know everything, just what I picked up from vibes, backed up by the context of a lot of Chinese inland travel.

So I arrived, stayed in a nice new hotel for 350 CNY a night, walked the streets, got some street food, walked some more, ate some dinner, snapped some pictures, chatted with a few locals, and called it a night. Don’t sue me.

Although Tier 3 is still above the median city urban development level in China, Yichang is still much more representative the "average" urban Chinese experience than any T1/T2 city. You will learn far more about the Chinese urban condition spending time here than in any number of visits to Beijing, Shenzhen, Hangzhou, or Chongqing.

Yichang is not trying to be (nor can it be) a mini-Wuhan, but it’s also much more than just a large Tier 4 prefecture town. It’s Tier 3, and I think I know what that means now.

Thanks for reading.

If you find this as interesting as I did and would like to see Yiwu for yourself, I will be leading a weekend immersion tour to Yiwu at the end of October in partnership with the team at Baiguan. Check more details about that here.

About Yiwu

Located in central Zhejiang just a few hours south of Shanghai on the high-speed rail, the small county-level city of Yiwu is the small commodities wholesale procurement capital of the world. Despite that, it’s still basically unknown outside of a few sectors.

Originally, I visited and wrote about Yiwu assuming that it would become part of my “Off the Beaten Track” series. But then I wasn’t so sure. Yiwu IS actually pretty well-known in the supply chain and sourcing sector. And to be fair, most Chinese people and even some non-Chinese will have heard of it too.

Sure, it’s not famous like Chongqing or Wuhan, but you can find many videos on domestic and international video platforms of people visiting the city, and the international media likes visiting Yiwu quite a lot as well, especially the Yiwu International Trade City (YITC), which is the largest light goods wholesale market in the world. So, I wondered: is it really off the beaten track?

In the end though I decided it indeed qualifies, after mentioning it to a few non-Chinese and getting blank stares. I guess it really isn’t known outside of the procurement and supply chain segments. Frankly, this is a shame, because if you have any kind of interest in international economics, trade, or globalization, Yiwu is one of the most educational places you will ever visit.

Administratively, Yiwu is part of the prefecture-level city of Jinhua, but it has a very strong identity of its own. I was told locals quite resent being categorized as part of Jinhua. This is simply an arrangement of administrative convenience.

Perhaps it’s because Yiwu was an independent city for most of its history until a restructure in the 1920s finally made it part of Jinhua…

…or perhaps it’s because Yiwu is the true economic anchor of the region, much more so than its “parent” city. Jinhua doesn’t even have an airport. It’s in Yiwu…

…or perhaps it’s because the population of Yiwu is larger than the population of Jinhua itself. Jinhua administers a total population of 7 million people, but “Jinhua Proper” - i.e. Wucheng District - has just shy of 1m. Yiwu has 1.9m…

Well, whatever the reason, there’s a lot of Yiwu pride, and zero affinity for Jinhua.

Yiwu today is famous as the small commodities (consumer goods) capital of China. Historically cursed with poor soil and subpar growing conditions, Yiwu was a commerce and trading hub long before the establishment of modern China. It was perhaps then naturally primed for its modern-day role.

Walking the YITC

Following the economic reforms of the 80s, Yiwu re-grew itself into a light goods manufacturing and distribution center of unparalleled proportions under the savvy guidance of a forward-thinking County Party Secretary. This was also the origin of the Yiwu International Trade City (YITC), where most of Yiwu’s vendors are concentrated.

The city is filled with factory shops on every street, but the centerpiece of commercial activity is surely the YITC, a massive shopping mall sprawling across 5 huge buildings (called “districts”) each with 5 floors. It’s a 7km walk from end to end.

From stickers to sneakers, from helmets to hammers, from bicycles to birdcages, if it’s a small commercial commodity, today it’s distributed in the YITC.

Tacticool camo? YITC has it.

Hands-free headlamps? YITC has it.

Kids’ bicycles? YITC has it.

Treadmill for your home gym? You’d better believe YITC has it.

The usage of the word “city” is not an exaggeration. There are roughly 75,000 vendors present across the YITC, selling every small household good imaginable.

Most of them are wholesale only, which means they won’t sell you anything in-store (although some vendors offer retail sales too, and have signs saying so). Think of it more as a massive exhibition hall, representing tens of thousands of factories.

On a typical weekend at any point during the day there will be well over 100,000 people inside the YITC, including the shop vendors’ children, who run around shouting and treating the open areas of the mall as a huge playground.

The districts are well-organized, with floor maps indicating what type of goods can be found in each area. For example, Floor 1 of District 3 has writing tools, paper products, and office supplies. Floor 3 of District of 2 has lights, batteries, and small electronics.

An entire section of District 4 is just zippers and buttons. Hundreds of shops, thousands of choices for zippers and buttons.

Of course, some categories are too specific to have their own section, so they get stuck together with other goods. Things like old-timey radios your grandfather would love, ugly knick-knack clocks your grandmother would love, or Confederate flag memorabilia that the weird classmate you haven't seen since high school would love (by the way, the vendors at the shop have no idea what the Confederate flag is, what it means, or even what country it’s from).

Somewhat famously, there's a large section in District 1 specializing in holiday and party supplies, with Christmas-, Halloween-, and birthday-themed sections. Here, it’s always Christmas, all year long.

My mother has collected these little wooden nutcrackers for years. I think she perhaps has enough to open her own wholesale business too.

Launching My Sourcing Side Hustle

You can wander for hours here, but your eyes will start to glaze over quickly and everything starts looking the same. To entertain myself, I decided to play at “sourcing” some Chinese goods, according to requests from my friends.

My friend Glenn at Read Write Invest told me his kids are into Rubik’s Cubes right now, with the Gan 356 model the most popular option for serious speed-cubers (people who compete to solve the cube as quickly as possible).

It’s going for 20 bucks on Amazon:

After a few minutes of searching the toys section of District 1, I found a vendor who carried this exact model. This is a “high-quality cube for people who like to play professionally”, the vendor told me. The price? 15 CNY each (about 2 USD), with a minimum order quantity of 30 pcs.

But if you don’t need a high-quality one for professional play, a cheapie unit will do you just fine, she told me. The unit price on the cheaper ones could go as low as just 3 CNY each (or 0.42 USD).

The vendor told me she’s been selling Rubik’s Cubes for 20 years, but a recent downturn in global demand for the classic toy has forced her to diversify into other educational toys. “I only want to sell toys with educational value, so I have started carrying some new products, all related to children’s learning. As long as it helps children develop their minds, then I can sell it. But the market for Rubik’s Cubes is shrinking.”

I suppose there are few people on earth more qualified to offhandedly summarize the historical ebbs and flows of Rubik’s Cube demand. Like many vendors in the YITC, she splits the stall with another vendor, to share the hefty rent of 180,000 CNY/year (around 25k USD).

Her stall-mate sells rubber chickens.

After finding the Rubik’s Cubes for Glenn, I continued on my volunteer product sourcing duties. Next, my friend Lei wanted to know about pet supplies - specifically cat nail clippers.

I used the convenient YiwuGo app to search what section of the market carried pet supplies. It pointed me to Floor 2 in District 2, where I quickly found stores carrying these products. I took the wide photo of the product display above and shared it back to my “client”.

Lei soon spotted a familiar item - these Gonicc branded nail clippers which he had recently picked up for 10 bucks on Amazon.

These exact clippers are at the top of the display in my picture above:

At this wholesale vendor, these clippers were selling for 3 CNY each (42 cents) with a minimum order quantity of one box (which holds 240 pieces).

The vendor told me that almost all the pet products hanging are produced in the same factory, even ones with different packaging, or showing different brand-names. “We make all of these…the branding and colors can change…this is just to show you options. We can use any packaging material you like and put your own brand on it.”

New Energy Products

After this, I found an area I was keen to source products for myself: the New Energy Products Market, built just a few years ago.

The products in the New Energy Products Hall are basically a giant celebration of the huge array of things you can stick solar panels onto, once they become as cheap as dirt, like they are in China.

I found solar-powered fans with built-in batteries for 100 CNY (14 USD) per unit . "On a full charge, these fans operate for 16 hours" the vendor told me hopefully. "The minimum order quantity is just 10 pieces".

Another vendor had foldable off-grid solar kits (180 CNY, MOQ 20 pcs) along with inverter-battery combo units offering capacities between 500-4000 watts, for 2-4 hours.

The 500-watt batteries cost 750 CNY; the 4000-watt batteries went for 4000 CNY.

“I guess these are for…camping?” I asked the vendor.
“You can use them for camping” she replied. “But most of my customers are from underdeveloped parts of Africa. They use them in places that don’t have power yet, or where the power supply is not stable.” Yep, you’d best believe I was embarrassed.

The type of international buyers throughout the YITC ranged from Middle Eastern, to African, to Central Asian, but in the New Energy Products Hall especially, they were mostly sub-Saharan African. Small groups of buyers with backpacks in small groups roamed the halls with notebooks and I typically heard a mixed patois of French, Arabic and English as I passed them.

At first, I thought they were carrying backpacks to help carry sample products. But later in the afternoon, I realized the backpacks’ true purpose when one of the daily prayer times arrived. In synchronicity, groups of foreign traders migrated to the central plaza areas of the mall, carefully picked the direction of west, pulled out mats from their backpacks, removed their shoes, and began to pray.

Besides the solar products, I also noticed a lot of African traders procuring basic electrical equipment, including things like cables, wires, switches, and outlets. If you’re working on off-grid power solutions, I guess it’s quite convenient to have a one-stop shop to procure all these other things too.

Outside on the street of Yiwu, many shop signs are in Arabic and English. The same is true for signs in the YITC. Some vendors have a gimmick where they stand outside their shop and shoot their sales shot by trying to guess where the passers-by are from and greet them in the correct language.

For me, they guess Turkish, then Arabic, then English, in that order. Not an unreasonable series of guesses, I suppose, depending on how recently it’s been since I’ve shaved my face.

Some of the shops in the YITC are even owned by foreigners. I stopped by one selling off-grid solar solutions managed by an Armenian man. His business partner was from West Africa. He told me it’s his third business in Yiwu, after he got a start in cosmetics and perfume. Their customers are mostly from Central and West Asia, as well as Africa.

The shop next door in the New Energy Products hall knew exactly who their target customers were, and was promoting their very specific message to that very specific audience.

Buy our solar and battery products, and you too can be the hero who brings power to your dusty savannah village. Six-pack abs not included.

On a recent work trip to HK, I forgot to bring my power adapter and my hotel didn’t have one to borrow. I ended up having to buy one from a vending machine in the nearby mall for (an exorbitant) 200 HKD. If only I had been in Yiwu. Here, I could have bought 20 of them for the same price (they wholesale for about 8 CNY each).

Yiwu the City

The YITC vendors are pretty much all private enterprises, as there’s very little participation from SOEs in the light manufacturing segment these days. These were all the first sectors to privatize back in the 1980s.

There is at least one notable state player here though: Yiwu City itself. The YITC was built, operated, and maintained by Yiwu City Investment Group 义乌城投, or “Cito”.

They also operate the bilingual YiwuGo app, which provides interactive maps for the YITC, as well as a product search directory.

I noticed Yiwu City Investment Group has their fingers in many pies across the city, from rideshare bikes and gas stations to tourist sites and road infrastructure - a classic Chinese municipal LGFV. Their skyscraper headquarters dominates the Yiwu skyline.

And yes, Yiwu has a skyline. In fact, Yiwu has a comically large skyline for the small county-level city it is. Yiwu has a downtown financial district built up along the Yiwu River that could put many larger cities to shame.

The names on the skyscrapers aren’t the ones you’d expect to see – no HSBC or Citigroup. Instead, it’s local companies – Yiwu-based banks, development companies, and of course hotels. With traders from all over the world arriving in Yiwu to procure goods, it seems the most thriving industry besides light goods sales is hospitality and hotels.

It certainly makes sense how Yiwu was ranked #7 in the nationwide ranking of strongest counties in China in 2025, with a YoY GDP growth rate of 7.5%.

Besides being wealthy, Yiwu is also the most broadly diverse city I’ve seen in China. According to its municipal statistics, 1.4m of its 1.9m people are not from Yiwu.

The city has about 20,000 foreign residents, but also sees over 300,000 foreign visitors a year, many staying for weeks at a time, so there’s likely 30,000-40,000 foreigners (or about 2% of the city’s population) running around downtown Yiwu on any random day, making them highly visible.

The foreign resident population is sufficiently large for entire neighborhoods to become international enclaves, reminiscent of Chinatowns in other countries.

Before I came, several people had told me that the best Turkish food in China is found here, and now I believe it, having seen dozens of Turkish restaurants and coffee shops.

Over my two-day trip, I ate local Yiwu cuisine just once; my other three meals were Afghan, Turkish, and Georgian.

Perhaps because of the considerable presence of people from the Middle East (and the resulting proliferation of halal cuisine), Yiwu also appears to be a gathering point for Chinese Muslims. I've never seen them so concentrated in a city that isn't explicitly an ethnic minority region.

I saw Chinese men in taqiyah and women in Chinese-style headscarves everywhere, and also noticed quite a few international couples comprising a foreign Muslim man and a Chinese Muslim woman (it seemed to be mostly Hui women), which I've not seen anywhere else in China.

The Duantou Night Market is one of Yiwu’s 4 major night markets, and the only halal one, featuring Chinese Muslim cuisine from all over the country, including Mongolia, Gansu, Ningxia, Qinghai, Xinjiang, and Yunnan. I've traveled a lot in China and was still seeing new things.

Visiting Yiwu holds huge educational value, both to understand the mechanics of global wholesale distribution of light commodity goods, and, perhaps more interestingly, to help imagine a future where China becomes an immigration destination, and how those immigrants may live.

Overall, Yiwu was revelatory, a novel experience in a province I’ve recently had to admit I’ve been seriously underrating. The city was young, dynamic, diverse, and of course, cheap. There’s entrepreneurial spirit, vigor, and lots of wealth. Good vibes. Definitely worth a visit!

That’s all for this time. If you’re convinced by my description of Yiwu as a place worth visiting, and want to get more into Yiwu as an economic template for successful public-private partnerships empowering the world’s most efficient manufacturing, trading, distribution, and logistics hub, consider a visit on a private guide tour with Baiguan, hosted by yours truly. More information and sign-ups here.

By 2022, however, the numbers had improved drastically, with only 14.3% of Chinese workers aged 65+ not having completed primary school.4

What happened? Did tens of millions of Chinese elders go back and complete primary school in their old age? Of course not. Sixteen years had passed since 2006. People aged 65+ in 2006 were now 81+, and many of them had aged out of the population entirely, while those entering this cohort were much more likely to have received a primary school education. The structure of the group itself had changed.

To have been 65 in 2006 means you were born in 1941 into Nationalist China at war with Japan. Your primary school years were roughly 1946-1952, tumultuous years of civil war in which the Nationalist government was overthrown by the Communist Party and the nascent People’s Republic established.

On the other hand, to have been 65 in 2022 means you were born in 1957 into Mao-era China. Your primary school years overlapped with the Cultural Revolution, which disrupted education throughout the country, especially in 1966-68. However, the worst disruptions were for secondary education and universities, while primary education enrolment actually expanded during the period.5 This China saw substantial gains in basic literacy and primary school access, even as secondary school completion rates plummeted - both of which are visible in the demographic record decades later.

These two Chinas are already very different. But neither remotely resembles the China of 2022, where only 0.3% of people aged 16-20 had failed to complete primary school and almost 50% of young people went on to post-secondary education.6

What happens if all three of these population cohorts are compressed into a single national aggregate statistic of educational attainment? Which demographic is the most representative? Are any of them?

The Compressed Demographic Transition

These questions illustrate one of the more obvious consequences of what Chinese demographic researchers call the “compressed demographic transition” 压缩式人口转变.7 Over the last seventy years, China’s economy and society have changed so much, and so rapidly, that constituent cohorts of its population grew up in functionally different countries. Accordingly, they reflect welfare outcomes from very different stages of development. For national aggregate statistics, these cohorts are averaged together. This yields averaged indicators that are statistically accurate, but often fail to describe the conditions experienced by any single generation particularly well.

Educational attainment illustrates the mechanism clearly because it is straightforward to assess in population surveys. But the same logic applies to indicators as diverse as height, civic behavior, or wealth. Just as you usually cannot go back to attend primary school in your retirement, a retired person cannot benefit from accumulated wealth they never had the opportunity to earn, nor can they draw pension income from decades of contributions they never had the opportunity to make. Experiences in early life cast a long shadow.

Consider the current 65+ generation in China, comprising an estimated 200 million people. Someone 70 years old today was born in 1956. Their childhood nutrition was heavily influenced by the conditions of the Great Leap Forward. They were 10 years old when the Cultural Revolution began and higher education was disrupted (basic literacy rates continued to rise, but fewer than 5% went on to secure a high school diploma).8 They spent their 20s and part of their 30s in the late planned economy and early reform era. They turned 45 in 2001, just in time for China’s WTO accession, at which time China’s GDP was growing rapidly, but still slightly smaller than France’s. As urban employees, they may have retired around 2011-2016.9 However, if they are from rural areas, they are possibly still engaged in informal/agrarian labor today.

In other words, they grew up amid varying degrees of famine conditions, missed out on most of their formal schooling, spent their early working years in a sputtering late-stage planned economy, and then dedicated their most economically productive and mature years to a poor, early-reform China. They contributed to and enjoyed a taste of the country’s spectacular post-reform economic and social development near the end of their formal working lives, then retired - if they could afford to.

Consequently, many entered retirement with fewer assets, lower pension entitlements, and much less financial security than younger cohorts can expect to achieve. Aside from their modest personal savings, they are mostly reliant on their families - or the state - for healthcare, eldercare, and general welfare, which is why they are also the object of a great deal of domestic Chinese research assessing their needs.

Because these cohorts remain numerically substantial, their life outcomes continue to shape national averages. They have lower incomes. They have less wealth. They received less schooling. They’re shorter. They represent the statistical outcomes of the past, and there’s surely methodological merit in demarcating their outcomes from those of younger generations. Chinese demographers and statisticians are of course aware of this - that’s surely part of why they report so much welfare data broken down by cohorts.10

Human societies typically contain four or more living generations at once, and China still contains hundreds of millions of people whose childhoods and early working lives were shaped by pre-reform conditions. When one cohort replaces another, aggregate measures of wealth, income, and retirement security can improve through demographic replacement alone. This process will continue for decades in China. The practical and statistical effects of the current 65+ cohort won’t change much until they mostly age out of the data, and the cohort is refilled with the post-reform demographic instead. In other words, national averages will become fully representative of modern China when all living generations were themselves raised in modern China.

It’s Not That Unique: South Korea’s Earlier Transition

This compressed demographic phenomenon is not limited to China. Actually, it can and should appear in any country where economic growth is rapid enough to create large welfare gaps between successive generations still alive at the same time.

South Korea is a useful leading case here. If we date Korea’s modern economic take-off to the launch of its first Five-Year Economic Development Plan in 1962, then South Korea circa 2010 was as far removed from the beginning of its development boom as China today is from the dawn of reform in 1978. But 1978 is not necessarily the most useful starting point for this discussion. If we instead define China’s modern growth model according to the boom in urban employment, private enterprise, and housing wealth following Deng’s Southern Tour, then 1992 becomes the more relevant benchmark. By that signpost, the best demographic analogue for China today in a cohort-composition sense is not South Korea in 2010, but South Korea in 1996.11

The South Korea of 1996, much like the China of today according to the 1992 clock, was a country more than three decades into its modern growth model. Younger generations were already enjoying the fruits of their country’s rapid industrialization, while the country still held large older populations whose formative and wealth-building years occurred in a much poorer society.

These similarities in cohort composition between 1996 Korea and 2025 China are useful, albeit imperfect. On a real GDP per capita basis, China today still lags South Korea in the mid-1990s: Korea’s 1996 GDP per capita was about US$14,900 in constant 2015 dollars,12 compared with China’s roughly US$13,800 in 2025.

On other indicators, China today is already ahead of 1996 South Korea, for example, life expectancy at birth: Korea’s life expectancy in 1996 was about 74.2 years, while China’s was 78 years in 2024.13 The point remains though: despite robust economic growth experienced by younger generations, Korea in 1996 and China in 2025 were both contemporary upper-middle-income economies bearing the statistical legacy of much poorer historical cohorts. 14

Three decades later, most statistical traces of that older Korea have faded from the aggregate data, but not completely. 2023 data showed that elderly poverty rates remained high versus other developed countries, and research explicitly links the phenomenon to generational disparities in pensions, asset accumulation, and exposure to economic growth.15 Simply put, Korea’s statistics haven’t yet entirely shaken off the final lingering cohort effects of its poorer past.16 In a cohort-composition sense, China is still several decades further behind this point.

When the Numbers Catch Up

We started by establishing how today’s aggregate statistics partly reflect a China that doesn’t exist anymore. Aggregate measures of household wealth or disposable income are arithmetically correct, but remain lagging indicators of past developmental conditions, averaged across populations that faced radically different opportunities over their lifetimes. The Korea comparison suggests it will take decades for these cohort effects to disappear entirely - at least four generations by my reckoning. Even wealthy South Korea will probably need another decade or two before those effects fully fade from its aggregate welfare indicators.

It’s thus quite reasonable to be optimistic about what welfare indicators like average wealth, income, lifespan, or consumption will look like when the average retiree belongs to the first generation raised in post-reform China (e.g., born in 1978, retired by 2038) or even modern-economy China (e.g., born in 1992, retired by 2052). If we imagine China today is roughly where South Korea was around 1996 in a cohort-composition sense, then a meaningful share of China’s remaining statistical welfare gap with richer East Asian neighbors should be set to narrow through demographic replacement over the coming decades - even in the absence of any extraordinary policy intervention.

Of course, growth will still matter most. But as China’s pre-reform and early-reform cohorts are gradually replaced by cohorts raised entirely in the post-reform world, these indicators will improve as a matter of mechanical arithmetic. Indeed, the research I linked above has already observed that Korea’s elder poverty statistics appear to be improving as wealthier cohorts enter retirement - independent from policy action by the Korean state.

In theory, this gap could be closed sooner via state transfers. In practice, that has proven to be fiscally and politically challenging. The role the Chinese state plays here strikes many as overly modest, particularly when it comes to pensions for the rural elderly, which are much lower than those for urban employees. Fiscal transfers can improve retirement outcomes substantially, establishing a floor to help ensure older populations face retirement with a reasonable guarantee of dignity, even though they never had a chance to contribute to a formal pension system in their working lives.

There is of course debate around where that floor should be - and many argue it should be higher. I tend to agree. Even after repeated increases, the national minimum basic pension under the urban-rural resident scheme is only 143 yuan per month, while average benefits remain roughly 246 yuan per month nationwide.17 But frankly, no generosity of fiscal transfers can fully offset a lifetime of unequal opportunities for education, wages, and asset accumulation. Even if pensions are beefed up, members of older cohorts will likely continue to exhibit lower wealth, income, and consumption levels than younger cohorts throughout the remainder of their lives.

This also introduces an interesting counterpoint to some of the concerns around the looming demographic cliff in China - with its old-age population expected to balloon to over 400 million people by 2040. The elderly population of 2040 will be much larger relative to the working-age population than it is today. But they will also enter their retirement with far greater resources than today’s 65+ cohort did. They will be far more likely to own urban housing and have paid into employee pension schemes. They will have spent most of their working lives in a much wealthier economy. The challenges posed by the growth China’s aging population are real, but the needs of its retirees in the future cannot simply be treated as extrapolations of the needs of the 2022 retirees - and certainly not the 2006 ones.

Demographic replacement therefore accomplishes something that redistribution policies cannot. Beyond merely transferring resources to people whose lifetime opportunities were limited, it gradually replaces those cohorts with people that never faced those limitations in the first place.

This will be reflected in the national statistics, of course, and in the practical implications for their families and the welfare state. But most importantly, it will be reflected in the quality of life for the people those statistics describe.

Hainan's beaches are really excellent, and not just 'good for China'. The white sand beaches east of Sanya up the southern coast of Hainan are world-class...as nice as any I've seen.

Probably the very best sand quality I saw was Clearwater Bay in Lingshui (陵水), which is a small autonomous county just east of Sanya. The white sand here was thick, fine, and soft, with clear turquoise water and great views up and down the coast for many kilometers.

Further up the coast, beach quality varied quite a lot, with both white and yellow sand, but rocky beaches were generally rare, sea grass/kelp was minimal, and the water was always very clear. It was easily equal to the best beaches I’ve seen in the Philippines.

Besides beauty, Hainan’s beaches also offer strikingly clear lessons about how coastal real estate development succeeds (or fails) in China. Clearwater Bay has very little development on it, and qualifies as an utterly squandered opportunity in my eyes. Aside from a single luxury hotel (a Raffles) there's hardly any commerce on the 10+ kilometer strip...just a handful of coffee shops and photo-op locations. There's a lighthouse and a marina full of expensive-looking private boats, but not much beyond that. Compared to the dense commercial development of similar beaches elsewhere - even just one city over in Sanya - my impression was that it was all quite bizarre...until I started asking questions.

Cafe Struggles

"We keep applying for new business ventures but we keep getting rejected" the Dongbei owner of a beach coffee shop not far from the Raffles told me.

He and his business partners from Jilin (in northeast China) rented out several blocks of land on the beachfront to develop commercially, but now they're struggling to find something that will pass approval.

"You can't do hot food of any kind, because that makes smoke. You can't use certain kinds of materials because of the environmental impact. You can't exceed a certain height, or a certain floor size area. So many restrictions...we have applied for so many ideas but none of them were approved. We're really making a big loss on our investment here…"

His coffee shop may be directly on the beach, but getting to it requires a 10-minute ride in an electric golf cart, with no direct vehicle access to his location. Along the way, you’ll pass long strips of nicely-manicured gardens, along with a few high-end property developments and villa complexes, but very little commerce or entertainment. The design concept for this recreational beachfront is severely lacking.

Clearwater Bay is currently owned by Chinese developer Agile Property (雅居乐) and they seemingly don’t seem to know what exactly to do with the massive chunk of real estate they control. They have development rights for the entire 10km of coastline, but have mostly filled it with endless rows of residential retirement communities.

I asked the coffee shop guy why the many kilometers of beachfront had nothing except high-rise apartments and luxury villas.

"That's all they know how to make...and that's all the local government wants to see. Tourism sounds nice, but actually it only gives you an economic benefit during three periods of the year: May Day, National Holiday, and Spring Festival. The rest of the year, there's little tourism, except for the Russians. So if you rely too much on tourism, you can't get rich. They prefer to just sell apartments. Or they did, back when the market was still good."

"But can't you charge a lot of money during those three tourism periods and make your money for the rest of the year?"

"You can raise prices a little during the holidays, sure, but there's so much competition in Hainan, so you can't make a lot of money. It's just too competitive."

"Yes, the car rental agency in Sanya told me that daily rates for rental cars here in Hainan are the cheapest in China" (they were, by the way. It was roughly 50 CNY, or about 7 USD a day).

"Of course. It’s too 卷...." (juan - competitive, or involuted)

"Because of these problems, there's very little commercial development along this beach? Isn't Agile annoyed with the local government because of this?"

"Of course. Everyone is annoyed. We rented this strip from Agile, and they want to help us to make money in theory, but they can't force the local departments to approve our project. Lingshui is an autonomous county for the Li ethnicity (黎族, ethnic minority native to Hainan), so oversight for development of coastal resources is very strict. Every application on the beach has to go through approval by village, town, county, and even the province level. In the end, all we can do is put out some chairs and an umbrella and serve drinks."

“So you guys must have lost a ton of money after renting this area from Agile, since you can't even develop anything on it?"

"Yes...very expensive. And it's not just the land. Everything here has to be built more expensively. You wouldn't believe how many millions it cost to build this little coffee shop, because it needs special, reinforced steel to resist typhoon winds and thicker glass. Plus the construction here is so slow. If you hire local people for construction, they work so slowly and not diligently!" He was particularly incensed by this.

"Can you believe, even if you offer them extra money to work overtime to finish the construction faster, they just say no, and go home. They really don't care about working hard. They are too 佛系!"

佛系 is popular slang right now, literally meaning 'Buddha-like' but generally describing someone who is uncompetitive, low‑stress, and goes with the flow, not forcing outcomes.

"Er...kinda sounds like you're calling them lazy. In a subtle way."

He laughed and shrugged, gesturing to the sunny beach and the ocean. "It's not unreasonable to be 佛系 in this kind of environment. But there is a reason why Hainan is still underdeveloped."

He paused and then added: "I've only been here six months and already I am feeling relaxed too. I used to feel like I needed to struggle every day. Now I think it's quite nice to lay in the hammock and swipe on my phone, then watch the sunset...."

The Matter of the Master Planner

Government restrictions and expensive construction can explain part of the story, but not everything. The coffee shop guy told me Clearwater Bay is underdeveloped because of the mismanagement of the site by Agile itself, who isn’t experienced or competent enough to be the Master Planner for beach real estate.

I’m sure it also matters quite a lot that Agile is absolutely flailing right now, had net losses of 22B CNY last year, has defaulted on some of its offshore bonds, and so clearly has no money to put into further developing Clearwater Bay, even if it wanted.

After doing more research, I concluded actually only a part of the blame should go to local conditions or poor policies. I think the competence and financial health of the Master Planner is much more important than both of those, for this kind of project. Whether they have a coherent vision...and are effective at allocating resources in pursuit of it are the key deciding factors for whether the project becomes a lucrative success, or a huge, expensive failure.

We found examples of both in Wanning, the next city east after Lingshui.

With China Overseas Land & Investment (中海) as its master planner, the “Shenzhou Peninsula” recreational beach tourism zone looks like a major success by conventional metrics - and it feels like it on the ground as well. It’s a huge, mixed-use tourist complex that is basically a standalone town in its own right, including residential real estate, malls, outdoor markets, resorts, beachfronts, and even office buildings.

When we arrived, the beachfront hotel/resort slots at Shenzhou Peninsula were all filled by major brands (Sheraton, Holiday Inn, etc.) the high-rises seemed to have sold well, the golf course had excellent reviews, and the commercial district + food street/night market was absolutely packed with people. This is the vision of success for a master planner…fully subscribed across all their verticals. It had to be. Online reports indicated that China Overseas Land & Investment spend billions of USD to secure the development rights for this parcel of land.

Online, I saw quite a few listings for luxury beach villas *by owner*. This means that although the market might be bad now, at least the developer succeeded in moving units initially. From an investor perspective, this is a success, regardless of what happens in the secondary market later.

About Land Reclamation Projects

By way of contrast, we should review the outcomes of a different mixed-used real estate project that ended in disaster. They’re not hard to find in Hainan. Not far down the coast, the Riyue Bay (Sun and Moon Bay) artificial island development has clearly been catastrophic for its state-backed developer Sunac (融创). Originally conceived in the 2010s as a massive thematic artificial island, the project ambitiously planned for two offshore developments in the shape of a sun and crescent moon, linked to the mainland by bridges.

Reality had other plans. In 2017, national environmental regulators stepped in to block the construction of the “moon” island for conducting land reclamation without proper authorization. Later, the sun island finished completion just as the entire real estate market starting heading south. Today, the “sun island” is full of half-finished buildings shells and empty apartments, playing host to dozens of short-term hostels and guesthouses. I’d be surprised if anyone actually lives there full-time.

We thought about staying here initially, drove onto the artificial island, took one look around, and left. With half of the project abandoned and the other half full of empty units, Sunac’s losses must be considerable.

Aside from Riyue Bay, there were actually many poorly-conceived land reclamation island projects off the coast of Hainan in the early 20120s that all got shut down in the first round of stringent national environmental inspections in 2017/18. This included, most famously, Sanya’s new international airport, which was dinged for illegal land reclamation and later earned the project developers an 11m USD fine. Today, Sanya continues to search for a new airport site while continuing to make do with its decades-old legacy airport, which is now obviously straining under a daily passenger load it was never designed for.

But not all land reclamation projects end in disaster. Up on on Hainan’s northern coast, on the opposite site of the island, Evergrande’s “Haihua Island” project in Yangpu Town, Danzhou is a thriving “offshore” community, despite the lack of sandy beaches (the northern coast facing Guangxi doesn’t have nearly the same natural beauty of the southeast coast). It has also faced its fair share of controversies over the years, amid dire claims about its outlook and financial viability. But when I visited later on the trip, it was full of people at least.

Yangpu Town itself is highly industrial (just about the only truly industrialized part of Hainan) and I don’t really understand the charm of retiring to a coastal seaside community in Hainan where the air quality is hardly much better than being on the mainland (and you can see an oil refinery from your window). But the island was packed with people when we drove through briefly. With most people getting around on golf carts and tricycles and the modal hair color being grey, it looks like quite the successful retirement community. Maybe it’s because it’s just really cheap. I saw real estate agents advertising entire 1 BR apartments in Haihua Island for as little as 120k CNY all-in. For context, that’s the price of ONE square meter of real estate in some parts of Shanghai.

Haihua Island seems to highlight a different development logic...not based on tourism or natural beauty, but affordability and internal circulation tailored to retirees. There’s clearly a market for it - certainly at the more affordable end anyway.

The Best Place in Hainan

But on this most recent trip, we didn’t stay on any reclaimed island projects - nor did we ever need to. Actually, the best place we stayed on the trip was in a little town off the coast of Qionghai city on the southern coast called Bo’ao.

Bo’ao is well-known as the host site of the Bo’ao Forum for Asia (BFA), which is sometimes called “Asia’s Davos”. The town is small but comfortable, with many good dining options, pretty beaches, a reasonable level of beachfront development, and very affordable and pretty ocean view apartments in the coastal high-rises.

In terms of beachfront development success, Bo’ao sits somewhere between the failed, arrested development of Clearwater Bay and the obvious roaring success of Shenzhou Peninsula. Charming and very livable, but also not without its share of real estate failures too…it was the perfect topic for my investigatory instincts. But that will be a story for next time!

Introduction

Over the first half of 2026, a new phrase has inserted itself into the vocabulary of trade economists, policymakers, and journalists: “China Shock 2.0”.1 What began as a niche label for China’s dominance in electric vehicles, batteries, and other advanced manufactured goods has now evolved into a broader debate about development, industrialization and trade. Critics of China’s export dominance are naturally worried about the implications for what remains of manufacturing in the industrialized (or increasingly post-industrial) West. But an increasingly prominent version of this argument raises a different concern: that China’s rise is also closing the path that earlier generations of developing countries used to industrialize.

Versions of this argument are now appearing across policy journals and newspaper columns. A recent Foreign Affairs article, “China Is Pulling Up the Ladder Behind It: How Beijing’s Export Strategy Will Keep Poor Countries Poor”, summarizes the authors’ recent working paper at PIIE and presents perhaps the most forceful expression of the argument: By remaining active in lower value-added production (even as it upgrades to high-end manufacturing), China is crowding out poorer countries and closing the traditional export-led pathway to industrialization.2

This Foreign Affairs article is useful not because it is unique, but because it concentrates many of the central assumptions around this narrative into a single, clear thesis. Accordingly, I will use the article as a reference point throughout this essay. The real subject, however, is the broader line of reasoning that has become increasingly common in discussions of “China Shock 2.0.” and its effect on developing economies.

I do not claim China’s economic model poses no challenges for poorer nations. China is a continent-sized economy operating under a single trade, fiscal, and monetary regime designed primarily around China’s own development objectives. Its sheer scale allows it to sustain advantages across an unusually wide range of sectors at once. For countries hoping to build export industries of their own, that reality can be daunting.

My disagreement lies elsewhere: not with the idea that China is a formidable competitor, but with the claim that this amounts to “pulling up the ladder” according to some general theory of how development works. It relies on assumptions that do not hold up to closer scrutiny and treats the interests of a relatively small number of manufacturing competitors as though they were synonymous with the interests of the entire developing world. Ultimately, this thesis relies less on any real-world model of industrialization, and more on a tidy moral narrative of victims and oppressors. Once the actual mechanics of development are examined, and once we ask who really benefits from the proposed remedies, the argument is just not very convincing.

China is Not a Monolith

The first thing the China Shock argument often misunderstands is typically reflected in how China is described - like a single homogenous economic unit. When its proponents emphasize that China’s dominance of both high-end and low-end goods is unnatural, they want you to look at the high-tech industries of places like Hangzhou or Shenzhen and think: why is this country still making toys and t-shirts?

But in reality, it’s not Shanghai labor making cheap t-shirts. It’s low-skilled domestic migrant labor from rural or inland China, supported by mature industrial clusters. China is big enough and its internal development levels heterogeneous enough to specialize at multiple points in the value chain at the same time, the way different countries normally would. Call them China1, China2, China3 if you like. Together these different parts of China can cover a full spectrum of economic activities.

This confusion also prompts the frequent misuse of economic terms. For instance, at one point the authors of the Foreign Affairs article claim China seeks to “retain comparative advantage in everything” - a statement that has gone on to be repeated verbatim by many who agree with the thesis.

But that’s not how the concept works. Comparative advantage is about relative opportunity costs, not competitiveness. The authors seemed to be referring to absolute advantage across a wide range of industries, which is entirely possible. But that’s a different claim and invoking some violation of comparative advantage here only creates confusion. It still exists; it just determines which parts of China do what.

The ultimate claim that emerges is that the continued presence across so many sectors from so many Chinas crowds out poorer countries and blocks the traditional export manufacturing-led development path. If China collectively dominates not just final assembly but also all the steps of intermediate steps of value-add like fabric, buttons, packaging, logistics, etc., then it narrows the pathway for later-industrializing countries, who were “supposed” to inherit those segments. But is that true?

Development is Not a Queue

The argument seems to understand industrial development like a kind of queue: After one country vacates a space to produce a certain good, another country naturally advances into that space to produce that good instead. Each step forward represents an upgrade to industrial productivity and value-add. This is loosely reminiscent of the famous “flying geese” model for East Asia.3 But that framework is descriptive, not prescriptive. It can retroactively be applied to describe a historical pattern of industrial upgrading under specific conditions, but it does not form a general rule that industries must get handed over on a fixed timetable.

In reality, the ability to produce things doesn’t happen just because an empty space opens up in front of a country with low-cost labor. Countries still need infrastructure, logistics, supplier networks, reliable energy supply, sources of financing, and so on. In most cases, the key factor for whether a developing country can move to the space ahead is the maturity of their own production ecosystem, not whether another country is already occupying that space.4

Even if we look at more relatively mature late industrializers, where the “queue” model should theoretically be at its most prescriptive, the outcomes are hardly automatic. Take India for example. Examples abound of firms trying to shift production from China to India (particularly during the Covid pandemic) only to encounter meaningful constraints in supply networks or logistics that unwound the cost savings on labor.5 Most other late industrializers are even less prepared than India to absorb production displaced from China without substantial external intervention.

It’s Not Just About Labor

For this reason, metrics typically applied to measure whether a China has “excessive” low-end export dominance, such as its share of low-skilled exports vs its share of low-skilled labor, are flawed.6 This metric can only be used in that way if the rest of the production ecosystem is comparable across countries, and the only difference between India and China is their shares of low-cost labor. We already know it is not.

Anyone with experience across Asia supply chains could give you a long list of things that matter for China’s competitiveness other than its low-cost labor supply, including automation level, infrastructure depth, energy cost, logistics speed, supplier depth, etc. For Indian producers, Chinese intermediate inputs in particular have proved to be particularly difficult to eliminate. 7 One analysis even bluntly noted that even after years of targeted policy efforts, “dependence on Chinese technology, machinery, and upstream industrial inputs remained largely intact.”8

Today, China continues to account for a massive share of global manufacturing exports although its manufacturing wages are now as much as 5-10x those of India, Bangladesh, or Vietnam.9 However, a divergence between labor share and export share doesn’t automatically demonstrate distortion. It may simply reflect a productive ecosystem that is unusually efficient at turning labor into exports - the potency of which anyone who has sourced or produced products in China can attest to. Thus, treating labor share as the benchmark for when a country “ought” to give up certain industries is not especially helpful.

Historical comparisons of China today to the historical moment when more advanced economies gave up low-end manufacturing run into the same problem. Today’s global economy is not the global economy of years past, when the earlier industrializers underwent their productivity upgrades. Supply chains are deeper, logistics are cheaper, and production is more fragmented than in the past. Chinese production is now operating at a scale and level that has no precedents. It’s not clear that comparing to smaller, more homogenous economies that upgraded earlier would be very informative, or define when China ought to give up certain productive segments.

To their credit, the authors of the Foreign Affairs piece acknowledge in passing that China’s dominance may reflect real underlying competitiveness, not just state distortions. However, they then dismiss the scale of its contribution as “unknowable” due to incomplete Chinese data access, as if Chinese supply chains aren’t some of the most robustly studied production ecosystems in the world. That concession then disappears from the rest of their analysis and is absent in the conclusion. This is typical for the China Shock arguments generally, which focus on trade outcomes and market share while spending little time on the mechanics of industrialization itself.

Exchange Rates Can Only Explain So Much

The common argument here about exchange rates is somewhat more plausible and seemingly straightforward. If the yuan is undervalued, then it can act as a broad subsidy to exports. This has been one of the most consistent international complaints around Chinese trade policy, spanning from the time of the original China Shock to the current day, and it is certainly a logical place to look when trying to uncover why Chinese exports remain so competitive, despite constantly rising labor costs.

But even there, the FX argument can only explain the margins, not the whole phenomenon of Chinese productivity. A weaker yuan helps to bolster cost competitiveness but does not by itself explain how China built supplier networks, logistics systems, industrial clusters, or production ecosystems over decades. Put another way, the exchange rate may affect how much share China retains today, but it doesn’t explain how or why China is able to compete across so many segments of the value chain to begin with.

Whose Interests are Being Represented?

This discussion about the value of the yuan also highlights how the argument changes depending on whose perspective one adopts. We already know that for producers in wealthier nations with mature industrial ecosystems, the weaker yuan is a source of concern because it makes it harder to compete with Chinese exports. This was the basis of the concern around China Shock 1.0, and remains a motivation for China Shock 2.0, extended to the high-end segments China is now competing in.

But by contrast, developing countries are not solely competitors to Chinese industry; they are also importers of Chinese capital equipment like machine tools, power generation equipment, construction machinery, and so on.10 From their perspective, the same exchange rates that make Chinese finished goods harder to compete against also reduce the costs of their own industrialization. The benefits of a stronger yuan would accrue primarily to countries that are already competing with or close to capturing manufacturing activity from China, while the benefits of cheaper Chinese capital good exports are dispersed across a much wider set of developing economies.

This reveals how the “China is hurting the Global South” narrative is blending together concerns about competition and development, picking whichever hits harder in context. Once the focus shifts from consumers of Chinese capital goods to competitors of Chinese manufacturing, the interests of a few parts of the developing world may partially align with those of advanced industrial economies. But the remedies proposed will clearly have limited benefit for countries importing Chinese capital goods and industrial inputs. If the argument is about India, Bangladesh, Vietnam, and a few other countries directly competing for manufacturing activity, then reduced Chinese competitiveness looks quite attractive. But if the argument is about Nigeria, Ethiopia, or Bolivia, then cheaper machinery, infrastructure, and imported industrial inputs is perhaps the more attractive prospect.

What Is China Being Asked To Do?

We’ve now looked at three different commonly-cited indicators used to argue that China’s industrial dominance indicates unnatural distortion rather than real competitiveness. The first two (labor share comparisons and historical international benchmarks) are meant to show that China’s time in these sectors should have already passed. The third (the exchange rate) is meant to explain why is still hasn’t. But each of these arguments is built on a weak foundation: the first overemphasizes labor as the determining factor, the second relies on non-comparable historical benchmarks, and the third overstates the explanatory power of the exchange rate. None of them constitute a binding model for behavioral norms in trade and development.

Furthermore, once we begin asking which developing countries actually benefit from the remedies being proposed, the picture is considerably less clear than the “China versus the Global South” framing implies.

With both the empirical rationale and the claimed constituency fuzzy, what’s left is a set of normative claims about what China ought to do and why. This is where the argument really loses persuasiveness. The demands usually involve importing more low-end goods, giving developing countries duty-free access, subsidizing imports from poorer countries, encouraging Chinese firms to move production abroad, allowing currency appreciation, etc. These all share the same underlying logic: China should give up productive activity so a small roster of other countries can take it.

Won’t this hurt Chinese domestic interests? The authors barely engage with the question, instead skating past it in their Foreign Affairs piece with: “Such measures might have some negative consequences, mainly in hitting lower-skilled Chinese exports”. Okay…then what is the case for why China should do it? Why should Chinese policymakers heed the calls of foreign analysts who are implicitly optimizing for something other than China’s own welfare? The only answer seems to be a normative political one: responsible leadership requires facilitating the rise of others. In other words, China needs to demonstrate leadership through sacrifice.

But even if India, Bangladesh, or Vietnam genuinely benefit from weaker Chinese competitiveness, it's not clear why Chinese firms - or policymakers - would drive this shift if productivity is still high and there are pools of Chinese labor that can be productively employed in lower-skill sectors. If those workers weren't making garments or toys or doing basic assembly, they would not suddenly be making semiconductors. They'd be doing something with even lower productivity - or perhaps unemployed. Ceding those industries to other countries means giving up viable employment for low-skilled workers - a rerun of what happened to manufacturing sectors across the West in decades past. This is not a concern to be casually dismissed by calls for virtuous leadership via sacrifice. And it’s inappropriate to deem this “pulling up the ladder” when much of China is still on the ladder too.

This is an Unprecedented Model

But perhaps you consider the moral mandate to be a defensible argument in and of itself. If you believe that’s what great powers should do, and so China ought to take on a global leadership role that involves charitable transfers of productive capacity to poorer countries, fine. Put that in the open so we can see the argument for what it is.

There is indeed a tradition in international political economy arguing that great powers bear greater responsibility for maintaining international order. Hegemonic stability theory relies on this principle.11 Seemingly, so have calls for China to become a “responsible stakeholder” in years past.12 If critics of China’s economic model believe Beijing should sacrifice some competitive advantage to support broader system goals, they are free to make that argument. But they should make it openly as a claim about the responsibilities of great powers, rather than smuggling it in the backdoor via the language of economic theory or invoking historical norms.

As I have already argued, this is not an argument based in development economics or any historical model of industrialization. There have been policies to help developing countries integrate into trade, and multilateral lending for industrial buildout, but there’s been very little precedent for countries to deliberately exit competitive sectors to create space for others, especially when they are still developing themselves.

This argument is on especially tenuous ground when its proponents try to spur nostalgia for how the West supported industrialization in developing countries in the past, invoking a tradition of benevolence that China is supposedly now violating. In reality, since 1945 only two things have served as effective drivers to transfer industrial capabilities across borders: geostrategic competition and market-driven transfers. Neither were acts of charity.

1. Geostrategic competition: Considerable amounts of Western technology, capital and process knowledge spread to selected strategically important allies during the Cold War to counterbalance the political and economic influence of the USSR.

2. Market-driven transfers: Producers in richer nations more broadly offshored production to low-cost labor nations nations to cut costs, boost margins, weaken labor bargaining power at home, and create opportunities for capital (a trend that has obviously benefited China as well).

These drivers certainly did not extend uniformly across the developing world. Outside a handful of strategically important allies, most Global South countries spent much of the late twentieth century navigating trade restrictions and conditional lending programs that were often far less supportive of industrialization. Neither mechanism was principally charitable, or motivated by a belief that successful economies (or “great powers”) had a moral obligation to surrender competitive industries for the benefit of poorer nations.

Would it Even Matter?

And finally, even if China did step back from low-end manufacturing and revalue its currency as a matter of principled state policy, the China Squeeze argument remains overly focused on matters of trade balances while offering few answers for the harder matters of development. If the constraint in poorer countries is their weak conditions for industrial productivity, then China “making space” isn’t going to do much. Production won’t move to a place not able to receive it.

In practice, China’s external development model via e.g. Belt and Road projects has mostly revolved around overcoming the barriers to industrialization that developing countries face, things like capital goods, logistics chains, and buildout of physical and energy infrastructure. The BRI’s critics would argue that this infrastructure is not developmental in intent. They would likely characterize ports, railways, and power projects built by Chinese actors across SEA or Africa as designed around Chinese commercial interests, whether it be resource extraction, market access, or opportunities for contractors.

There is truth in the observation that Chinese firms benefit substantially from these arrangements. Yet this objection also risks missing the point. The relevant question is not whether Chinese firms benefit, but whether the resulting infrastructure, logistics networks, power systems, and capital goods leave recipient countries with greater productive capacity than they had before. Development has rarely been driven by altruism. The relevant benchmark is not motive but outcome, and the outcome of China’s presence in the developing world is increasingly one of an industrializer.

Ironically, several of the cities cited in the Foreign Affairs piece as victims of the “China squeeze”, including Addis Ababa, Nairobi, and Phnom Penh, have been major recipients of Chinese infrastructure, capital goods, and industrial investment. Their relationship with China is not simply one of competition. If China is truly “pulling up the ladder,” it is choosing an odd way to do it.

As McKinsey summarized in its commentary on China’s 2025 export trends:

“…rising exports of parts and machinery were not tied to replacing China’s lost US sales, rather they supported the expansion of manufacturing capacity in third markets, particularly emerging economies, deepening China’s role as a supplier of production inputs rather than a final-goods exporter.”13

So What?

If we assume that for now, China will continue to hold on to a chunk of low-end manufacturing that other countries wish could be theirs instead, no amount of finger wagging or emphatically argued policy papers is going to change this. Instead, the development task should be to analyze how other developing countries fit into this ecosystem and find something useful to do…or differentiate themselves in a niche.

What lessons can be learned from countries like Vietnam when it comes to learning how to dance with the industrial elephant that is the Chinese manufacturing sector? Considering Vietnam is industrializing immediately within a China-centered ecosystem, it’s unlikely they would describe this process as an “occupying the open space in the queue” maneuver. And while it’s not always unanimously rosy, a common-enough sentiment in Southeast Asian commentary seems to be they are also gaining alongside China, not being crowded out.14 So what policy mix is allowing them to participate and compete, despite China’s scale?

That’s where this China squeeze narrative is at its weakest. After identifying the problem, its proponents often slip into exhortation, as if strong enough criticism will prompt Beijing into repentance for its sins and reorganize its economy accordingly. Rather than identifying strategies for replicating the success of the more successful ASEAN countries, or emerging Chinese partners like Ethiopia, the oft-cited remedies have bare guidance for developing nations, amounting to little more than repeating the call for China to “make space” – i.e., give up parts of its industrial base so others can have them. The beneficiaries of such a strategy are not necessarily synonymous with the Global South as a whole, and developing countries may have very different interests.

The central failing with the “China is hurting the Global South” thesis is that it attempts to transform a complex set of questions about competitiveness, development, and industrial policy into a simple moral narrative. Once the assumed mechanics of industrialization and development are examined more closely, its empirical foundation starts to crack. And once we ask which countries would actually benefit from the proposed remedies - as opposed to being hurt by them - its claim to speak on behalf of the developing world becomes weaker as well. What remains is ultimately a normative argument about how China ought to behave as a major power. That is an argument one can make. But it is not a theory of development.

I don’t usually get into a taxis in China with the intention of discussing a certain topic, or investigating a specific theory. It takes something away from the spontaneity of the conversation, and if I do it too often, I figure my readers are likely to start doubting the authenticity of the interactions. But in special cases, I will go into a taxicab conversation with a bit of intentionality. One case is when there’s some kind of ongoing discourse about what Chinese taxi drivers themselves - partially if the discourse is about what Chinese taxi drivers think, or believe. After all, if I come across a claim concerning Chinese taxi drivers that I’m not sure sure about, who better to ask?

Like the one below, for example:

I happened to see the tweet in the screenshot while on a work trip in Hangzhou, exactly at that moment sitting in a Didi (rideshare) car on my way to a meeting. So I naturally had to immediately ask my driver what he thought.

Hangzhou Driver 1/2: Proud Dad and Urban Maximalist Cao Shifu

“Hey shifu, how many rides do you do in an hour?”

“Uhm. Maybe one. Or more, if they are short. My revenue is about 500 CNY per day. I’ll keep 300 CNY per day, after my expenses.

“So, you’re living on 300 CNY per day?”
“No, what I mean is 300 per day is what I have left, after I pay all my living expenses. I spend about 200 CNY a day on food, rent, expenses, etc.”

“Ah so you own your own car? No car rental?” [many Didi drivers in large cities lease their vehicles, which would cut into his ability to save]
“Yes, that’s right.”

“In what situation would your rides drop to only one per hour? Is it because of a traffic jam, causing the ride to take longer? Or not enough people calling for a driver, so you have gaps between orders?”

"There are too many drivers now. There's basically no entry requirement to become a driver...anyone can do it (“门槛太低了”). As long as you have a car and want to drive, you can. So now we have too many drivers."

"Why so many drivers? What were the drivers doing before this?"
"Oh you know. Building real estate. But now the market for real estate is bad, because it's saturated. There's no demand for houses, because Chinese people basically all own a house. Home ownership is over 90%, you know"

"Do you mean the old houses...the village house?"
"No, I mean at least in the county-level city. Chinese people basically all have a urban house in at least the county-city (县城) now. So there's no demand for it anymore."

He continued: "Except for some people in those places like Xinjiang, Tibet, Inner Mongolia. They don't have the demand for an urban house, because they have a different kind of lifestyle, you know, like nomadic herders. But all the other Chinese people already own an urban property, so there's no more demand for houses. That's why the real estate market is bad. That's how I understand it. So the workers who used to build houses now instead deliver takeout or parcels, or drive for Didi."

“So I guess you have an urban property already too?”
“Yes, that’s right. My parents helped me with the down payment.”

“So now you’re paying off the loan?
“No, it’s already paid. I own my house already.”

“Okay...well I guess you aren’t from Hangzhou. Where did you buy?”
“I’m from Jiangsu...Yancheng City. Property in Yancheng is much cheaper than Hangzhou. Hangzhou property is 30k, 40k per square meter. In Yancheng, it’s just 7k...or maybe 10k in the city center.”

"So if you already have a house, fully paid off, why are you still working in Hangzhou, driving a car for such long hours, netting just 300 CNY per day? Couldn't you go back to Yancheng and relax? The salary will be lower, but your stress will be lower. You can start to tangping.” (躺平, or “lie flat"; leave the rat race and take it easy).

"I can't tangping yet. I have to think about the next generation. If I went back to Yancheng, it's true my life would be much more relaxed. But here, I can save 300 a day to put away for my future, to enjoy when I retire. In Yancheng, my salary would be lower, and probably I could only save 200 per day. I’d have an easier life now, but I want to make sure I have enough savings so I can live a really good retirement, not just passing the days."

"When you say the next generation, are you also thinking about buying a house for your child?"

"No, I don't have to worry about that. My parents have a house, and I already bought a house, so we have two in the family now. That's enough. If we need, in the future I can live in my parents' house and my daughter will live in my house. My daughter is only 16, so I will keep worrying about her until she is married and has her own family. Perhaps I can tangping after she is married."

"Is your daughter in Yancheng? Or Hangzhou?"
"She's in Hangzhou here, with me and my wife. She's in the first year of senior high school now."

"Did you ever have any trouble with sending her to school to Hangzhou...because your hukou is not from here...it's from Yancheng right?"

"Ah, that's not a problem. I have paid many years of social insurance in Hangzhou. But also, you know...honestly...if you are a very good student, even the best schools will accept the student, even if they are from another city."

Here, he let out a proud fatherly chuckle, the only emotion he showed during our conversation.

"If the high school believes the student will definitely go to a top college, and help improve their rankings, they don't care where your hukou is from. They will think of a way to allow them go to school there. heh heh...

...and my daughter is a VERY good student!"

We arrived at my destination, I thanked Cao Shifu, and the ride ended.

That’s part 1. It was a shorter one, so I combined it with Part 2 to continue exploring the same theme of the rideshare economy in Hangzhou. I took this second Didi ride on the very next day (September 10, 2025), kicking off the conversation with the same theme.

Hangzhou Driver 2/2: Rideshare Economics and Jewelry Industry Secrets with Xia Shifu

“Hey shifu, how many rides can you do in a single hour normally?”

“It varies. Sometimes 4. Sometimes 0. Depends what the system gives you.

“So you might have long gaps between rides?”
“During the day, sure. At the evening peak rush, no.”

“That’s right, there are surge periods and flat periods. That makes sense.”

“There are more Didi drivers than passengers during the trough periods in the afternoon. That’s when the price is low and we don’t get a lot of rides. But in the evening peaks, you will accept the next ride even before you finish the previous one. In one day, I can usually make around 500. It’s more in the summer vacation...on those days it should be at least 600 per day.”

“What’s the most profitable time to drive?”
“At night. The fares are higher, and there are fewer drivers. I would like to drive at night to make more money, but my wife won’t let me. She says it’s dangerous to drive long hours at night.”

“This model for rideshare pricing reminds me of my own industry. I work in the electricity industry, and power has a different price at each hour of the day based on real-time supply and demand.”

“Oh, like how charging my EV at night is very cheap. Yes, driving for rideshares is similar. We have peak prices, shoulder prices, and trough prices. But not so granular, like for electricity. It doesn’t change every hour.”

“You know, you could optimize your driving habits. Don’t bother driving when fares are low, and maximize your effort when fares are high.”

“Well, I take all the fares. As long as I can make 500 a day, it’s okay. I own my car, so my expenses are lower. I can normally gross 12k-15k CNY in a month.”

“And you’re driving full-time?”
“Yes, full-time”

“I guess you weren’t driving before this...right?”
“Right. I was driving part-time before, but I had another full-time job. I lost my job, so I became a full-time driver. I was a manager in the gems and jewelry industry. You know Chow Tai Seng?” (周大生)
“Oh, isn’t it one of those Hong Kong jewelry chains...?”

“Actually no. Chow Tai Seng is from Beijing, but the brand’s direction tries to be close to the Hong Kong brands. The 4 famous Hong Kong brands are Chow Tai Fook, Chow Seng Seng, Luk Fook, and TSL.” (周大福、周生生、六福、谢瑞麟)

“What about Chow Luk Fook?” (周六福)
“Actually that’s from Shenzhen”
“Ooookay. I’m confused. They all sound so similar. I only remember seeing these brands in the mall when I was looking for my wife’s engagement ring, and always got them mixed up. So you say you were at...Chow...Tai...Seng?”
“That’s right.”
“What happened?”
“I got fired lah. Management cut half of the staff in the Hangzhou corporate office, from 20 people down to 10 people. The gems industry is not so good right now.”

“When did this happen?”
“This year. Just a few months ago.”

“Are you from Hangzhou? Or did you consider going back somewhere to your hometown?”
“No, I’m from Hangzhou. In the suburbs.”

“Did you expect this was coming, or was it a surprise?”
“It was a surprise! We got one month’s notice, and then I was done. I thought the gems and jewelry business would be stable, because people tend to buy these things when they are unsure about the economy, but I guess it’s not true.”

“What about Laopu Gold? I heard this brand is doing very well right now.”
“Ahh, yes. Laopu Gold is doing very well. They have a very special sales angle. They have done a good job to make the items very fashionable and artistic, and connect them with Chinese culture. So if you can buy it for the art and cultural value. But then it also happens to be made of gold, so it will keep its value as a precious metal. It’s a good business strategy.”

“I’ve always felt the traditional jewelry stores are expensive and not very good value. Because you are paying for their branding, marketing...rent...etc. And you can’t easily find some unique or specialty items.”
“Yes, this is a problem in the gems industry. You are paying a premium price for a mass-market item. We always have to pay for the most strategic sales location on the pedestrian streets, or in the malls, so that’s part of our costs.”

“When I was shopping for my wife’s engagement ring, I directly asked her what kind of ring she would like. I didn’t want to get her something she didn’t want. And it’s a good thing I asked, because I never could have guessed what she wanted. She said she wanted a peach sapphire for her gem. But this is a rare gem - none of the big brands I visited even had one to show me. I had to go to a small custom jeweler to get it made.”

“Wow, that’s very unique. Yes, I would be surprised if all the Chow Tai Seng stores in Hangzhou combined even had one peach sapphire in stock. Sapphires are becoming very popular right now - and desire for diamonds is decreasing. Everyone realizes now that diamonds don’t actually have any value, but also lab-made diamonds are pushing the price down. On the other hand, even a lab-made sapphire is still very expensive, so sapphires will keep their value better. Especially a rare color like peach.”

“Yes, I guess they’re only from one place? Sri Lanka? that’s what it said on my gem certificate.”
“Yes, Sri Lanka.”

“So what did your other colleagues do, after your company cut staff?”
“Someone of them went back to their hometowns to find jobs. Some got rehired back into the stores. They are young, so they can still go work in the retail locations. One of them is now selling Teslas.”

“What about you? What’s your plan next? Just keep driving?”
“I’m 40. I’ve spent my whole career in the jewelry business, but I can’t go back to work in the retail locations now like the young kids. I could go work for one of the other big chains...I should be able to find a job. But they wouldn’t pay me as much, as an outside hire. So I’d be doing the same work I did before, for less pay. I wouldn’t be able to accept that. So for now I’ll drive full-time and see if something else appears. Driving is not bad for a short time.”

“And you’ve already bought a house?”
“We bought in 2019. It wasn’t at the very top of the market, like in 2021, but it was close. The house is worth less today than when I bought it. But it’s okay...not too bad. I’ve only lost about 500k CNY.” (70k USD)

”Losing 500k CNY is not too bad?”
“Some people who bought in the Future City area, next to Hangzhou West Station, at the peak of the bubble...their prices went from 50k/sqm to 30k/sqm. Their losses could total 2m CNY or more (280k USD).

“So your driving salary is enough to cover your home loan, take care of your kid - you have a kid right? - and your family’s lifestyle?”

“Yes, My child is 6 years old. It’s okay. My wife also works, so that takes some burden off me. She makes 8-9k per month as a makeup artist. She does the makeup for the brides at weddings...did you know this is so expensive? Doing makeup one time, you can charge 2-3k CNY.”

“Wow. But I think driving can’t be a long-term option, right? You want to spend time with your kid...”
“Actually driving hours are flexible, so that’s an advantage. Before, when I worked for Chow Tai Seng, I frequently had to go on business trips. And it was normal company policy to choose one day...Saturday or Sunday, to come in and work anyway. Even though our contract said we had both days off, in reality we were expected to work one of them. But it’s true...I don’t want to drive long term. It’s not good for your body. For the last few days, my foot has been feeling weird...it frequently feels numb. I will go to the doctor and get it looked at.”
“Oh yeah...you should. Maybe you need to change your sitting posture. Your body isn’t used to driving full-time...”

We’re arriving at my destination.

“Well personally, I think your former colleague who went to sell Tesla vehicles had the right idea. If you want a future-proof industry, you can look for something related to clean technology, such as EVs. They need corporate managers too.”

“Okay, I’ll try that thanks. Take care!”
“Bye!”

Over years of taking taxis and Didi rideshare vehicles, I’ve talked to a lot of drivers, slowly learning through many conversations what’s most important to them, and how they see the world. They like to complain. They complain about real estate prices. They complain about the city (especially Shanghai). They complain about how tiring their job is. But one thing they don’t complain about is low wages - and the frequency of orders (one per hour or otherwise) didn’t seem to be a major consideration for the two Hangzhou drivers I asked. It’s likely different in smaller cities, where the ridership demand is smaller, but then again the drivers are fewer as well.

Typically, driver wages are fairly in-line with median wages for whatever city they’re in, and almost always higher than the other low-skill work options available to them, like factory jobs, retail, or F&B. Indeed, if you can drive, and especially if you have your own car, driving a rideshare vehicle might be just about the highest-earning option available in the gig economy (better than delivering takeout). Perhaps that’s why it seems to be a more common choice for Chinese knowledge workers who have lost their job (or struggled to find one as graduates) and are looking to maintain a certain income level while they try to figure out what comes next. In that regard, not so dissimilar from driving for Uber in the US I suppose…

If you’ve spent any time in the China economic development space, you’ve probably heard this specific “600m” factoid tossed around.

Sometimes it comes up in earnest conversation about Chinese development outcomes. Other times it’s used more disparagingly. Either way, it’s usually delivered with total conviction, because surely the Chinese Premier knew what he was talking about, right?

Right?

Well…not exactly. It’s not that Li didn’t know what he was talking about. He was citing real data from the National Bureau of Statistics (NBS). But the statistic has been widely misunderstood, and at the time, unfortunately framed in a way that made it quite easy to misunderstand.

Let’s break it down…

Where does the “600 million” figure come from?

Back in May 2020, then-Chinese Premier Li Keqiang famously said in a speech:

That quote immediately received a lot of attention, both within and outside of China. How could Chinese incomes be that low? That’s below minimum wage in most places. Did Li inadvertently reveal some hidden reality about widespread poverty wages?

No.

The main issue is that Li was not referring to wages at all. He was citing a cohort of household incomes averaged across people, not wages earned by individuals.

This figure comes from the bottom 40% of the population (the lowest two quintiles). It is calculated from household survey data, where total disposable income within the bottom 40% of the population is divided by the number of people in that group.

Imagine ranking every Chinese household by income, then dividing each household’s income by the number of people in it to get a per capita household income. At the bottom ranks, you’ll get many households where income is being spread across children, elderly, and other non‑earners, pulling the per‑capita figure down.

Now take just the households in the bottom 40% of the ranking. Add up all their disposable income and divide by the total number of people. That’s your number.

So why is that so confusing?

Because this measure includes everyone, not just wage earners. It includes:

• Children (zero income, by definition)

• Rural elderly (often minimal or no pension income)

• Non-working dependents

• People outcome formal employment

All these people have little or no income, pulling down the per-capita average.

When Li said “600 million people’s income is less than 1000 CNY/month”, it sounded like he was describing wages, In reality, he was citing a statistical average for a population cohort that includes tens or hundreds of millions of people with no income at all.

No wonder people misunderstood.

In fact, the confusion spread so widely that the National Bureau of Statistics had to address it directly. Its spokesperson responded to the controversy in a press briefing, clarifying what the figure actually referred to: this is a statistical average for a segment of the population, not a count for wage earners.

The Economist later wrote an article referencing this episode, which included the subhead: “China is not as poor as Li Keqiang thinks”.

They included a useful illustration:

The Economist example is helpful in principle, although the subhead framing is unkind to Premier Li. I think he knew the meaning of the statistic, but erred by sharing it in a way that was likely to be misunderstood.

Compared to the Economist example, the effect is even more pronounced for China, because of the sheer number of people with little or no formal income.

Li’s wording didn’t help

So why WAS it so widely misunderstood? In my view, the biggest driver of confusion was Li's framing comment:

With this line, he likely meant to accentuate the hardship of being in a lower economic quintile. But his framing implicitly invites the reader to interpret the number as individual income vs. cost of living (i.e., a wage).

In reality, Chinese household incomes include multiple earners, rural families have subsistence resources and often don’t pay rents at all, and disposable income excludes in-kind benefits and social transfers. So this comparison really invites the assessment to go in the wrong direction…

[Also, as an aside, but you can rent for 1000 CNY a month in lower-tier cities. Not in Shanghai or Beijing, obviously. But not impossible.]

So what’s the actual number?

If we instead ask precisely: “how many individual people actually have disposable income below 1000 CNY/month”, we’d have to make an estimate:

• According to NBS data, in 2020 the lowest-income 20% of the population had an average monthly disposable income of 655 CNY/month. Almost the entirety of this cohort of 280m people falls under the CNY 1000/month threshold.

• Next, the lower-middle 20% had an average of 1370 CNY. Let’s conservatively say that half of them are below 1000 CNY/month. So that’s another 140m people

We don’t have within-quintile distributions, so any estimate here is necessarily rough. But this estimate yields ~420 million people (30% of the population).

What about today？

Looking at the 2024 NBS data:

• By 2024, the lowest-income 20% was still averaging just CNY 795/month. The growth is appreciable, but it’s still low. Low enough that I’d think about everyone in this quintile is still below the 1000 CNY threshold.

• Over the last five years, disposable income growth for the low-middle 20% has been the fastest across all 5 quintiles. By 2024, it reached 1800 CNY monthly, implying far fewer people under the 1000 CNY threshold. So how many exactly…?

Again, same caveat about not having within-quintile distributions. But if we try to keep our guesses conservative, guess perhaps another ~70m people in this quintile crossed the 1000 CNY monthly barrier. So in 2024 there were around 350m people (or ~25% of the population) below 1000 CNY/month. Still a large number, but not the common “600 million wage earners living on poverty wages” interpretation. Importantly, most of them are children, elderly, farmers, and informal workers, not formal wage earners earning below the minimum wage.

What should we take away from this?

Chinese wages aren’t as low as the meme suggests. And this cohort-based averaged per capita statistic isn’t even meant to describe absolute incomes. It’s most useful as a measure of inequality and relative development levels. It shows us:

• Large parts of the country still have very low formal incomes

• Income distribution is uneven, evidenced by cross-quintile gaps

• Even as it grows rapidly, China remains a developing country (which of course was part of Li’s intent in sharing this statistic in the first place)

In my view, the most important structural issue highlighted here is the low rural pensions, which affect ~180 million elderly people and obviously drag down the disposable income averages substantially. Likely the most meaningful policy pathway to improving low incomes in the bottom two quintiles is just to bolster rural pensions. Not that this is necessarily simple or straightforward to do in practice, but I hope it illustrates the kind of thinking you’re able to participate in when you grasp what the statistic is actually measuring - and what it means for Chinese stakeholders.

Understanding this makes all the difference.

I'm heading to a meeting in Pudong (the other side of the river in Shanghai). Mrs. Mi picks me up in a new GAC Aion Y and confirms my phone number.

Her Chinese accent sounds like mine in Chinese class 15 years ago. Mandarin is clearly not her first language.

"You...you're not Chinese, right?" she asks.
"No, I'm American. Do I look Chinese?"

"You look Arabic, or like you’re from Afghanistan. But you sound Chinese"
"I've been here a long time"

"How long?"
"13 years"
"Oh, longer than me"
"How long have you been in Shanghai?"

"Over one year"
"Ah, welcome to Shanghai. I can hear your accent...you're from Xinjiang right? What part?"
"Yes. I'm from Kashgar"

"Oh, Southern Xinjiang. I have friends from Korla and Yili, but I don't know anyone from that far south."
"Yes, that's Northern Xinjiang. Different from us."

"Do you drive Didi full-time?"
"Yes, this is my full-time job"
"So I guess you drive all day? 14 hours?"
"Normally 15 hours. I drive in the morning, rest for 2 hours, then drive again in the evening"

"That's very tiring. But you should make good money"
"Not bad, about 700 yuan a day"
"How many days do you rest each month?"
"Just one..."

"Wow, so you're making ...700 yuan times 30 days... Then subtract the lease for the car...which is what...like 7k a month...?"
"It’s a little over 20,000 a month. I don't rent this car...we bought it. So I have to pay back the car loan."

"Ohhh...you bought the car. Are you by yourself in Shanghai?"
"No, my husband is here with me. We are both driving. We drove here together from Kashgar to make money in Shanghai.”

"Does he also drive 15 hours a day?"
"No, less, because he has to share his car with another guy. So he only drives for half the day"

(She's grossing about 2x the median Shanghai salary, and about 3-4x a typical F&B salary, e.g. a server in a restaurant)

"So the two of you are making several tens of thousands a month. What is the plan? Go back to buy a home?"
"Yes, we will go back to Kashgar to buy an apartment. Our child is there now, with my parents. We don't want him to grow up without his parents, so we will go back to buy a house in Kashgar."

"How much is real estate there?"
"For one square meter...you can find houses for 3000, 4000, 5000 RMB. You can get a house for 100k, 200k, 300k RMB total. We plan to buy a 70sqm house for 300k. How much is that per square meter?"

I input the numbers on my phone. "That's 4.2k RMB per sqm. That's pretty cheap. I guess you'll be able to go back soon"
"Yes, we'll go back before the end of the year"

"It would take a long time to save up for this if you worked in Kashgar?"
"Kashgar...the jobs are bad. The most I could make in one month is 3000. And my husband could make 4000 or 5000. Too little."

"So it makes sense to come here, work like this two years, then go back. But it's a burden."
"Yes, it's a burden for us, for our child, for my parents. It's hard for everyone" (大家都辛苦了呗)

Now she wants to ask me questions.

"Are you really American? You don't look American. You don't have blonde hair..."
"Not all Americans have blonde hair. Actually most of them don't. There are all kinds of Americans."

"But in the movies they have blonde hair."
"Do all Chinese people look like Chinese actors? Of course not. American people don't look like American actors either"

She's not convinced. "You don't look American. You look like...you look like one of ours..."
"From Xinjiang?"
"Yes, you look like some of our people from Xinjiang"

"haha, some Chinese people told me that before, but I don't see it. My ethnic background is closer to Mediterranean. Italian and Jewish. You think people in Xinjiang would believe I'm from there?
"Yes, they would"

"Well, I don't speak any Uyghur, so they would know right away I'm not. I only know one phrase: "qoy shish kawap, pish dey?" Can you understand me? I probably said it wrong..."
"It means 'are the lamb skewers ready?'"

"Yeah, haha. A Uyghur guy at the BBQ restaurant I used to go to taught me how to say this. Is the Xinjiang food in Shanghai good, in your opinion?"
"It's okay. There are some good restaurants."
"Would you consider them authentic?"

"Not bad. You can try one in Shanghai, it's called Baigelige. I think it's good. We go there with friends."
"Are there many people from Xinjiang like you, working in Shanghai to save money and go back to buy a house."

"Yes, a lot. I know they are here to make money. But I don't know their destination...to buy a house or something else."
(amusingly but endearingly, she uses the wrong Mandarin word here, mixing up the words for "objective" and "destination", which is something I did too for years)

Is there Xinjiang food in America?"
"Yes, I think so, in some big cities. I've heard you can find Xinjiang food in cities like New York. But I've never tried it."

"Do Americans eat pork?"
"Yes, mostly. Unless they are Jewish or Muslim"
"Oh, there are Muslim people in the USA?"
"Yes, many. There is every kind of person in the USA"

"What do you eat in Shanghai? Is there authentic American food? Have you adapted to eating Han food?"
"There are some good American-style restaurants. But I eat everything...Western food and Chinese food..."
"Oh, okay..."

We're nearing my destination, so I pull open the Didi app to check her name.
"Okay thanks for the ride, Mi Shifu. Eh...your surname is Mi? Like rice? I guess that's not your whole name"

"No, we have longer names. I choose a shorter name so it's more convenient. Mi is just the first part. My whole surname sounds like "Mi Ya Gu Li" （米娅古力)."
"Oh, I understand. I have this problem too. My surname in English is too inconvenient to say in Chinese so I shorten it. Okay, thanks Miyaguli Shifu. Bye!"
"Bye!"

The content of the conversation wasn't so different compared to conversations I've had many times before in taxis and rideshares in Shanghai. Regardless from where people come, be it Anhui, Henan, Dongbei, or Xinjiang, they generally plan to bust their ass for a few years in Shanghai, working as hard as their bodies can endure, and then go back to their hometowns to slow down and buy property. That’s the modern pathway to fast-tracking first home ownership these days. I have a lot of taxi conversations that are just repetitions of that same theme, so many I don't bother sharing anymore, because it would just be the same story.

What made this time notable was my driver was a Uyghur woman from Kashgar. Female taxi drivers are already pretty uncommon in Shanghai (not sure why, because they're common in Beijing) and migrant female taxi drivers even more uncommon. Thinking back, this might even be my first time. In the few times I had a female driver in Shanghai in the past, it was always a Shanghai local, driving part-time.

Her Mandarin was pretty choppy and it was apparent that she wasn't super comfortable with it. The basic chatting we were doing was probably just about her limit. She was late 20s if I had to guess, and I suspect she barely spoke Mandarin before she got to Shanghai (that's been the case for my other friends with similar backgrounds).

Anyway, I wish her the best. I’m finally posting this essay on my Substack nearly a year after I met Mrs. Mi in Shanghai. I hope she’s not in this city anymore. I hope she executed her plan on schedule, and she’s back in Kashgar, where she’s already bought her apartment with her husband and they’re raising their little boy or girl and not thinking anymore about those 15-hour days driving a car in Shanghai.

If you want to visit the Uyghur restaurant in Shanghai that she thinks is worthy of being called "pretty good", it's this one: 百格里格 Kashgar Flavor. After she recommended it so heartily, I had to try it, but I thought it was just okay myself. But what do I know? Although a lot of people think I could pass for it, I’m NOT actually from Xinjiang. :)

You’ve heard of Chinese “nail houses” - those local residents whose unwillingness to sell their property block the aspirations of real estate developers, highway builders, and city planners. Contrary to much popular imagination, Chinese property (and land) owners enjoy fairly robust legal property protections and it can be quite difficult to dislodge a resident who is determined to not leave.

The most famous stories are often in rural areas, where standalone nail houses are striking and obvious, requiring highways or railways to make awkward detours to avoid the holdout property. But anyone could become a holdout, including someone in an apartment in the Shanghai old city.

When I say “Shanghai old city”, I’m referring to the eastern part of the modern-day Huangpu District south of Renmin Road (strictly speaking, the part of Huangpu that used to be Luwan District before they were merged in 2011). This roughly the neighborhood where the ancient Shanghai city wall used to stand. Looking at a map these days, you can still trace the oval shape where the old city wall once stood, following the routes of the modern Zhonghua and Renmin Roads. Even today, neighborhoods in this part of the city still refer to “gates” that were torn down 100+ years ago, but beyond this, scant other evidence of this city wall remains.

There's also now very little left of the old residential neighborhoods in this part of the city. With the South Bund development growing to the east and Xintiandi expanding to the west, these areas have been nibbled away bit by bit over the last decades until now almost nothing is left.

But you won’t find me waxing romantic about the renovation of these neighborhoods. As I’ve said before on here, these weren’t really nice places to live. They were indeed the closest thing to slums you’ll find in downtown Shanghai, and relocation to a new place in the suburbs is usually an upgrade - one usually welcomed by the former residents.

But not always. What happens if the locals don’t want to move?

On a recent walk through one of the condemned neighborhoods, I was surprised to find the gates to an apartment community still open, although all the shops and apartment windows were already boarded over/bricked up. I figured I could just walk in and check it out.

After a minute of wandering around, I realize there are actually still people living in here...a handful of holdouts living down an otherwise eerily quiet, desolate alley where the only apparent movement is the furtive scurrying of a dirty stray cat. From behind a few bricked-up doorways, I hear voices…but saw no people.

Finally, deep in one of the side alleys, I spot a human being, a guy just seemingly just a few years than I, in flip-flops and a t-shirt, leaning on a building and swiping on his phone.

He nods and smiles as I walk past, so I try to chat with him.

"Hey, what's going on here? Will this all be torn down?"

"Yeah this will all be destroyed"
"When?"

"I don't know...still a while I guess. No rush."
"Where is everyone else?"

"They already left...they moved away"
"But you're still here? How many people are still here?"

"Ah there are still a few. Here...2 households. There, 3." He points to the next alley.

"There are also some people living here secretly. They moved out, at first but then they came back here to live after a while when they say that the demolition hadn’t started. It's not legal, but no one minds."

"Yes, I heard some voices from apartments that I thought were sealed. So they're inside? Do they have power and water?"
"No, those are disconnected when they move out. It's mostly elderly people, who miss their old home."

"I guess more than 90% of people have left already? So why haven't you left? Nostalgia?"
"No, the developer did not give me a suitable offer"

"Right, they have to compensate you. Give you a new apartment...in the suburbs somewhere like Songjiang. Or give you cash. Right?"

"Yes. But if it's an apartment in the suburbs, it should be in a reasonable place. Not too far. And not smaller than my current home. And if it's cash, it should also be reasonable."

"So you're still here...I guess you don't think it's reasonable?"
"No, not reasonable."

"Almost everyone else left already. So they got some offer that they think is reasonable, but you haven't? Are you hoping for more compensation?"

"At the beginning, the developer promised everyone would get a good compensation, but they didn't define how it would be calculated."

“Yes it’s true that the Communist Party gave me this house, but now they say I should give it up. I’m not against it in principle, but it must be reasonable.”

“Oh, your family was allocated this house back during the property distribution in the 1950s?”
“Yes, it was taken from the landlord and given to my family. The landlords owned many houses. You can still see their big houses around here too.”

“Where I live over in Jingan District, we have many old mansions that used to be owned by wealthy people, industrialists, that have been divided up into multiple apartments that still have many units today.”
“Yes, some of the landlord houses around here are also divided into multiple apartments”

"So why do you think the developer's offer is unreasonable?
"They are only offering 60,000 per square meter. That seems too low...if they build apartments here, they will surely be worth at least 200,000 per square meter. But worse than that, they aren't even calculating the area of my home correctly"

"They say your house is smaller than you think it is?"
"Exactly. We've lived here 70 years. My family made many upgrades to the house over time, expanded its size. But they only acknowledge the original design drawings"

"The original blueprints...? Like from 1950?"
"Yes, basically like the original blueprints. 60k per square meter perhaps is ok if you credit my entire house. But you can't just offer me something based on old drawings...that isn't even what my house looks like!"

[I want to soberly point out the renovations performed on his house over the decades were probably illegal or non-permitted construction activities, else the drawings on file WOULD reflect his home's current layout and this exact issue wouldn't exist].

"So what will you do? Do they send some people to threaten you?"
"Ha, kind of. The developer doesn't dare to do it directly. They hire a subcontractor who hires a subcontractor who will hire some guys who will come and harass you, push you to accept the offer and move."

"So they're thugs? Are they violent?"
"No, not anymore. In the old days, they would be violent. But now there are cameras everywhere, even here" (he points). 'So they are just annoying to me. They knock on your door, ask you to come out and talk, even at night."

"What about the local government? What do they say?"

"Hah. The officials can't do anything. You know unless they are high level in Beijing, or in the Shanghai city committee, the local officials are basically just regular people, laobaixing like you and me. They come here to try mediate the problem...well actually they can only say some useless things like 'Ah...you are all reasonable people. Discuss this well! Discuss it well!* I'm sure you can find a way.' And then they leave. They are helpless."

(*”好好谈一谈， 好好谈一谈， 肯定有办法”)

He continues. "The government used to have more money. The property developers had more money too. If they needed to destroy your house, you could get rich...buy 2 or 3 houses. Now they have no money...so they’ve become stingy."

"So it's impossible to resolve? You need more money. They don't have it."
"Yes"

"Do you know what they plan to do with this area? It's actually a really good location...close to the Confucius Temple...good tourist value. Also...isn't this a shikumen community? I thought those were protected in Shanghai."

"Yes...it's shikumen. But it's not protected like the ones in Tianzifang or Xintiandi. It's just a regular community. However, the Confucius Temple is just across the street...it just finished its refurbishment. I should visit...I haven't yet."

"I would think they would choose to refurbish it, instead of destroy it. But I guess this area is indeed going to be destroyed? It's 拆迁, not 旧改, right?"

"Yes this is 动迁. Destroy the house and move the people. If it is 旧改, they will refurbish and the people move back later."

“Well...good luck I guess. I’ll keep walking around now”
“Okay, good to meet you friend. Can I add your WeChat?”
“Uhm…sure.”

He unexpectedly switches to English, reminding me that I’m not in a random tier 4 city in central China…this is Shanghai, where English is common.

“My name is Jarvis. Nice to meet you.”
[Fascinatingly, Jarvis has a slight stutter in Chinese that disappears when he speaks English]

“Oh you speak English! Nice to meet you Jarvis. Uh... Jarvis, huh? Like in that movie...uhh...”

“Yes, like in Iron Man. But I choose this name a long time ago, when I was a student. Before the movie came out.”

“Hah, okay. Well, nice to you meet you Jarvis. Stay in touch”

I took the picture below just in front of Jarvis's apartment. The signs on the wall and chair warn that this is private property, not to touch anything, not to knock on his door, and to basically leave him alone. I guess they're for the benefit of the annoying people who come by to harass him about moving.

A few reflections

The overall disruptive power of nail house holdouts in Shanghai is surely more impactful than in rural areas, but with the property market in its current state, I suppose the real estate developer is no in particular rush to find a resolution to people like Jarvis. Actually, there doesn’t seem to be any obvious legal channel to “fix” an issue like this in a way that everyone will deem fair. When discussing the issue of the excessively low compensation, Jarvis speaks with the confidence that comes from truly believing justice is on his side; as far as he’s concerned, his house IS larger than what is shown on the official drawings, and it’s simply improper to withhold compensation for square meters of his home that he can clearly point to the physical existence thereof, regardless of what it says on some dusty blueprints from the early Mao era.

And yet, he is only in this situation because the weaker institutions of a bygone era allowed his family (likely his parents’ generation) to get away with making construction modifications to their home without a permit, or without formally filing new construction blueprints. Maybe the process didn’t exceed at the time. Maybe it was too complicated. Maybe they thought it was just too much trouble. For whatever reason, Shanghai’s stronger civic institutions of 2025 now explicitly mandate his compensation match whatever square meterage is documented on the blueprints, making those illicit modifications performed in the “old China” of just a few decades ago (a China characterized by much flimsier institutions than today) a detriment rather than a boon.

Institutions are expected to get stronger as the country itself develops. Property compensation standards are an example of one such institution - one that is working to Jarvis’s disadvantage in this case. However, on the flipside, Jarvis’s legally protected right to squat in his home until the developer offers him a suitable sum is yet another. In that same bygone period when verboten modifications were performed on his home, a frustrated land developer blocked by an unwilling homeowner would likely have chosen to resolve the issue in a very different - and more aggressive - way versus the bureaucratized channels of the modern day. Such contradictions are also the nature of modernizing institutions.

At the time of posting here on Substack in March 2026, this conversation took place nearly a year ago now. Maybe I should go visit Jarvis and see how he’s doing these days. That might be an interesting follow-up someday…

In Nanning, (the capital of Guangxi) I hop in the rideshare car for a quick ride across town. Mr. Huang helps me load my suitcase into the back of his sub-compact EV, where I notice a yellow construction hat and a greyish mass of what looks like melted plastic in the trunk.

"Oh don't worry about that" he chuckles when I ask about it. Mr. Huang is a short, slim man who smiles and laughs easily. He has a close-cropped receding hairline and his Mandarin has the soft, choppy accent of someone whose mother tongue is likely Cantonese. "It's just some adhesive from my work that dripped and then dried. It won't get on your suitcase.
"Oh okay. Why do you have a construction helmet?"

"Because I work in construction." [guess I could have figured that out on my own]
"Huh, really. I've never had a rideshare driver who works in construction. I didn't know you guys had time to drive."
"It's the start of the national holiday, so the all worksites are already on break. I have nothing to do, so I figured I'd drive and make some extra money."

"Are you from Nanning?"
"No...I'm from Yulin." (玉林)

“Oh, Yulin. In eastern Guangxi, close to Guangdong [that explains the accent]. I know of it. Why didn’t you go home for the holiday?”
“My wife is pregnant...due any day now with our second child. We don’t want to use the hospital back there, and she wants to “zuo yuezi”* in Nanning. So we stayed for the holiday.”

[*zuo yuezi - 坐月子, the traditional month-long postnatal recovery and rest period in China]

"Ah, I see. I'm sure your family will accept this reason to not return home for the holiday. But Yulin is a big prefecture-level city too. There must be some good hospitals and yuezi facilities there too, no?"

"Well...I'm not really from Yulin City. It's Bobai County, governed by Yulin. It's a rural place."
"Ah...I see....Bobai County...let me check where it is on the map..."

As is typical for my taxi drivers, my showing interest in his hometown prompts him to volunteer more information about it.

"We eat dog there, you know."
"Uh. Oh. Yes. I know. Yulin has that festival..."
"Yes, the Dog Meat and Lychee Festival. When I was young, they didn't have this festival...it didn't exist. But everyone has heard of it now. Actually most people don't eat dog now. But some still do."

"Do you like to eat dog meat?"
"Sure, I like it. I think it's tasty, and it's not strange if you are used to it. But now people also have dogs as pets."

"Do you think it's contradictory, having dogs for pets and also using them for meat?"
"No contradiction ha. You can have dogs for eating and also dogs for pets."

"Do dog meat restaurants serve meat that was specifically raised for eating?"
"It's hard to tell where it came from. Or what kind it is after it's cooked."

"Many people are critical of Yulin because of the dog meat festival. What do you think of that?"
"Traditions and habits are different everywhere. Some people keep pigs as pets, but pigs are also food. They don't eat cows in India. Some people like to raise parrots as pets and teach them to talk..."
"But do people eat parrots?"

Mr. Huang shrugs and laughs. "Where I come from we would.* In the past, we ate everything. Some people ate cat, snake, wild rodents..."
"Wow I heard the joke before that people in Guangdong eat everything. Actually sounds like it should refer to your hometown instead."

[*I'd like to point out all species of parrot in China are nationally protected and trade/consumption is illegal.]

"In the past, when people were hungry, they would eat anything to fill up their stomach. If you are starving, a snake could be quite lovely. Now they don't have to eat like that anymore. We have chickens and pigs and ducks. But some people miss that old flavor. They prefer it."
"Isn't it illegal to eat wild animals now? Especially since covid, it's become very sensitive?"
"Oh yes, the government bans this. But some people still do it secretly. I guess people don't eat this in Shanghai?"

"Shanghai doesn't have this kind of restaurant. It's illegal to eat wild animals. It's not officially illegal to eat dog in Shanghai actually...but there are no restaurants. At least, none selling it publicly. I checked before. In Shenzhen eating dog is officially illegal. But I think it's the only city."*
"Oh yeah. I heard that, when worked in construction for a while in Shenzhen."

[*Actually, Zhuhai has banned it as well.]

The conversation peters out and we drive along in silence for a minute, and then Mr. Huang volunteers a new piece of information.

"But I really like dogs you know. I owned a dog before. A purebred male Labrador retriever, raised from a puppy. I fed him milk from a bottle and trained him. He was very smart!"
"Um. So you even raised your own dog as a pet, but you also like to eat dog. Wouldn't you feel weird eating your own dog?"
"Of course. I wouldn't eat him. I raised him, and I had a lot of affection for him. He was a very smart dog. I truly like dogs." He points to the two stuffed puppies curled up on his dashboard to prove his point.

"So...do you still have your dog?"

"Aii. No.” his tone suddenly turns sad. "When my wife gave birth to our first child, her parents were very worried that the dog would get something dirty on its paws and touch the baby, or transfer some disease or bugs. Actually, this kind of fear is not reasonable, but old people have their fixed ideas and its very hard to change them. So I had to give up my dog. At first I thought about selling him, and some people offered me a lot of money...a few thousand CNY. But I could bear to say goodbye. So instead I paid my friend 1000 CNY to drive here to Nanning, pick up my dog, and bring him back to Bobai to take care of him."

"Do you still visit your dog?"

Mr. Huang's tone gets even sadder, and also takes on a twinge of hesitating embarrassment.
"No...a few months later, the dog disappeared...My friend called me and told me couldn't find him anymore. I guess he was stolen..."

"Stolen? For what reason?"
"Yes stolen." Huang laughs bitterly. "The reason….well, everyone back home knows what it means when your dog suddenly disappears. He probably was eaten."

We have arrived at my destination, and so I do not get a chance to ask Mr. Huang whether he thinks his appetite for dog meat could have contributed to the market demand that led to the abduction of his pet. But that might have been my next question.

I thank him for the ride. He helps me unload my luggage, wishes me a happy new year, and drives off in his EV, on his way to make a few more yuan before his wife gives birth, together with his hard hat and his stuffed dashboard puppies.

Thanks for reading. Here are some final notes, as I am aware this is a sensitive topic:

Please remember this is one conversation with one driver; it’s not a survey of Yulin or Bobai County, where dog meat is *not* consumed regularly by most of the population. In the future, I would be very interested in doing a trip to Yulin to chat with people about their feelings on the the infamous festival. After all, it is now the unambiguous defining feature of the city, undoubtedly to the great chagrin of the locals, most of whom do not partake (it was started by dog meat traders in the city in an effort to revive flagging sales back in the mid 2000s).

For an actual fieldwork-based survey and interesting sociological assessment of dog meat consumption in Yulin, I suggest checking out this 2023 paper by Zhejiang post-doc researcher Liu Ying: "Wu Masculinities, Traditional Men-Dog Relationships and the Defence of the Dog-Meat Festival in Modernizing Yulin China."

In 2020, national agricultural guidance in China explicitly described dogs as companion animals rather than livestock for the first time, which muddied the legal and administrative basis for commercial breeding and trade.

Some Chinese research or media have described “edible dog” breeding operations, typically framed as small and informal rather than a proper industrial sector. You can find promotional videos for these operations on sites like Rednote. But beyond that, it's difficult from the consumer's perspective to assess the provenance of the meat - and evaluate whether dog meat enjoyers might be consuming someone's stolen pet. As Mr. Huang put it, once it's on the plate, “it’s hard to tell where it came from”.

While visiting Xiamen in July 2025, I grabbed an opportunity drive up to Shishi City for an afternoon. Shishi is a small Fujian coastal city, mainly known for its large garments industry - with huge business in both production and distribution. But I had a different reason for my visit: I saw it on a map and thought it had a funny name.

Shishi is the name of the city, and the word “city” in Chinese is ALSO “shi” (市), so its entire proper name is shi2 shi1 shi4 (石狮市) all with different tones. Its name literally translates as “Stone Lion City.” Reminds me of a tongue twister. That’s pretty much the only reason I need to visit a random small city like this…I notice it, think it’s funny, and then I decide I have to go there.

Shishi is governed by Quanzhou City and has a total population of about 700k. Because it’s a trading and distribution hub for the garment industry, about half the population is migrant workers - uncommon for such a small city where usually the inhabitants are locals.

Quanzhou has a pretty odd administrative structure (see image below). The large western counties (light green, blue, orange) are very rural - basically in the mountains. Meanwhile, the coastal districts and cities are relatively tiny, squeezed into the 50km of coastal plain tucked between the ocean and the mountains. Shishi City is the small coastal blob in dark green below, completely surrounded by Jinjiang City in lavender. It only takes a 1 hour’s drive north and west from Xiamen to reach Shishi.

Despite being small, Shishi is not a poor city. Its bustling garments industry and strategic coastal location have done wonders for the city's development.

Indeed, I wasn't surprised to find it in the Top 100 list of China's strongest counties- at #72 on the 2025 百强县 ranking. This list is headed up by more well-known wealthy small cities in the greater Shanghai area like Kunshan, Taicang, and Yuyao. I will surely write a separate post about this “100 Strong Counties” list…someday…

Besides its relative wealth, something making Shishi notable is the large foreign population (for a city of its small size). There are many traders here from the Middle East especially and it was common to spot shop signs and restaurants in Chinese, English, AND Arabic.

When I arrived in Shishi, the first thing I did was to visit the city's namesake - because yes, there IS a stone lion after which the city is named.

Shishi's Fengli Temple is quite small and very old. The derpy stone lion guarding the temple entrance is nearly 1500 years old - and looks it.

One cool thing about historic sites in smaller cities in China I've noticed is they are less likely to be renovated or refurbished for tourism, which help to maintain their authentic - albeit slightly weather-beaten and grungy - characteristics. If this lion were in Shanghai, I think it would have been "fixed up" a long time ago.

Sometimes, a renovation or refurbishment is just what is needed. The image below is the Shilang Arch, commissioned by the Kangxi Emperor in 1698 to memorialize the achievements of famed admiral Shilang who led the Ming’s Empire’s naval forces in Fujian in the late 17th century.

It stood unmolested for 300 years after that until local villagers encased it into a stone wall for a housing project in the 2000s. Then in 2011, when the renovation of this neighborhood began, the wall was torn down and the original stones of the Shilang Arch were discovered and retrieved. Finally it was rebuilt at the entrance of Bagua Street in 2021, still using the original materials, just in front of the Fengli Temple.

Bagua Street was still under renovation, with its shops mostly empty. But further down the street, I came across one open venue: another temple - Shishi's City God Temple.

Fujian has a LOT of temples, more than any other province I've visited. Buddhist, Daoist, and family shrines are seemingly littered on every street. Matching the proliferation of the temples, Fujian is stereotypically one of the most superstitious provinces in China.

In the ornate and heavily-incensed City God Temple, I even saw a woman doing “zhi jiao” 掷筊 - throwing bamboo blocks on the ground to ask deities questions and poring over the arrangement of how they landed on the floor to seek an answer.

After Bagua Street, I made my way over to the Shishi Clothing City, or SCC, centerpiece of commercial activity in Shishi. SCC comprises several huge, multi-story malls, with hundreds of shops serving as distribution points for fabric, clothing accessories, and completed garments.

The first mall I entered was the accessories and fabric mall. I wandered, looking for someone who didn’t seem too busy who’d chat with me. After not too long, I found two guys sitting on the ground outside their shop which sold drawstrings (for pant waistbands).

One of them was operating a digital cut-to-length machine, pulling braided cords from three separate piles at once and producing a pile of trimmed cord. The other was meticulously and painstakingly cutting all the aglets (the plastic cap thing) from a bunch of completed cords.

We chatted for a little bit - typical stuff like where I'm from, why do I speak Chinese, how do I like China, etc. They both joked about how proper my Mandarin sounded compared to theirs. The native local language of Shishi is quite similar to other Min Nan dialects spoken in Xiamen, Quanzhou, Zhangzhou, or Taiwan, and native Min Nan speakers often have heavily accented Mandarin, especially if they don't use it often. Indeed, "Fujianese people have barely-comprehensible accents" is such a well-known (and often true) trope that it can be used as a punchline in Chinese TV and movies. See, for instance, the character of the mycology expert Professor Ling in the show Minning Town (although apparently the non-Fujianese actor’s efforts at mimicking a Fujian accent yielded poor results, according to experts on the internet).

I asked them how business had been recently.

“Ah, alright” the one operating the cutting machine responded. “Some reduction recently, because of the trade tariffs. But only for the export orders. The domestic orders are still fine.”

“Last time was worse.” the other one added. “The first tariffs from Trump last time hurt business more. Right now I have more domestic orders.”

I had to go do some internet research to remind myself that Chinese textiles and apparel were indeed tariffed pretty aggressively during Trump 1, from 2018-19, before a deal was reached. It’s hard to keep track of it all sometimes.

“Is this how you normally meet orders? Working a cut-to-measure machine yourself?”

“No, this is for a small order. If it’s a big order, it will be done in the factory. But right now I have some time, so I’ll cut the lengths myself.”

“How long will it take?”
“Not long...I’ll finish these three piles within an hour.”

“And then you still have to add the aglets on the end?”
“Yes. We have different ones for different products. Plastic and metal. These will be used for the strings inside pants, so it will be plastic.”

"What about you?" I asked the other guy. "Why are you cutting the aglets off the ends?"
"Er...it was an mistake. We weren’t supposed to use this one, so I have to remove it."

"Whose mistake? Let me guess...your mistake? Whomever fucked it up has to fix it…that’s the logic, right?”

"Aiii….." he hung his head while the other guy roared with laughter. " Ha! You nailed him! Hahaha.”

I started feeling hungry, so I asked them for a food suggestion and they pointed me to a parking lot between two of the malls where the workers grab lunch. It was already past 2pm, so once I headed over, I found just one food cart still operating. He was offering various braised meats served over noodles.

I already knew from some internet research that braised beef intestines are a specialty of this region, and so although they're generally not my favorite, I ordered a bowl of intestines with noodles from Chef Wu. He cooked the noodles in the braising juice, which was a nice touch.

The noodles were served and I discovered to my very happy surprise that the beef intestines were absolutely goddamn excellent. Chewy, springy, beefy, and with none of the off-flavors one associates with poorly prepared offal. Just delicious.

The guy sitting next to me had ordered a bowl of the same and we started chatting. He was a local guy and told me he's very happy living in Shishi. When he was younger, he worked in Hubei for a while but didn't like it and came back. There's plenty of work locally, and more importantly, it's home, so he doesn't want to leave. He mostly does odd jobs, driving, logistics, loading and unloading trucks, stuff like that.

I asked him how important is the garment industry to the local economy, but he wasn’t sure how to answer, so instead I asked him how many of his friends or family are locally employed directly or indirectly in the garments or textiles business.

He thought about it for a second, then replied: "all of them".

I told him the shopping center I had just been wandering through was pretty big, but I hadn’t seen any foreigners...were there more foreigners around here in the past? He shook his head.

“Just not you were walking in this one, right? That’s just the accessories and cloth market. The export market is this next one, over here. And behind that is the shopping center for the domestic market.”

I suddenly realized at that time truly just how large the SCC is. I had been wandering around what I felt was a huge mall when I met the guys at the drawstring shop, but actually that was just one of three similar-sized shopping centers.

“We’re a small city, but we’re still one of the most important ports for the clothing industry. I think we should be #3 in China, after Guangzhou and Xiamen”. He told me there are still many foreigners living and working in Shishi, especially from the Middle East, which is why there are so many shops with Arabic and middle eastern restaurants, but I wouldn’t see them in the shopping center I’d just been exploring.

Online, I saw there's also an Arabic cultural center with a mosque and three Christian churches. Shishi has been a trading hub along the Maritime Silk Road for hundreds of years and has the unexpected cultural diversity (for such a small city) to prove it.

I chatted for a bit more about life and work in Shishi with my temporary lunch partner, until he had to get back to work. Before he left, I confirmed with him that he thinks the beef intestines are excellent too. “They’re the best I’ve had” he told me. “Chef Wu cleans and prepares them himself every day, with fresh ingredients - you can’t buy this in a store. You can’t let offal sit for more than one day...it will get gamey. That’s the key to cooking intestines....clean it very well, and use very fresh meat. Okay, I have to go. Take care friend. If it is fated, we shall meet again.” (“有缘我们再见面”)

“He comes every day for lunch” Chef Wu told me once he’d left. “He’s a very loyal customer”. I too had finished my noodles, so I paid the bill and headed on my short drive back to Xiamen. I never would have guessed a small street-side shop in a small stone lion city in Fujian would change my mind about braised beef intestines….!

Anyway, that was Shishi. I sure wish I had more than an afternoon to spend here. Chef Wu’s intestines certainly were a nice final impression of Shishi…even if Chef Wu rarely puts down his cigarette as he cooks - ha.

Mr. Le picked us up at the entrance to the Zeng Cuo’an tourist area in Xiamen. We were going to the ferry terminal and it was the middle of the day, so we had a bit of a ride ahead of us. I was immediately struck by his unique vehicle. (guess I should have been looking where I was going, amirite? *ba-dum tiss*)

Actually, what I meant to say is I immediately noticed his distinctive vehicle. It was a BYD EV, but I’d never seen one like it before. It was shaped like a smaller SUV crossover but had sliding doors and a somewhat boxy roof that offered lots of headspace like a minivan.

“Hey shifu, what is this car model? I’ve never seen it before. The BYD what?”

Mr. Le responded enthusiastically. "We just call it “Little Green” (小绿). BYD designed a car for rideshare drivers, as a partnership with Didi. It sold very well in Xiamen."

He seemed to have a lot of energy for a guy who spends all day just sitting. Great, someone to talk to!

“Really? We don’t have this car in Shanghai. At least, I don’t think so. I’ve never seen it. Didi drivers just drive regular cars.”

“I chose this one because it’s got a good design for passengers. The sliding door is safer to let out guests on roads with many bicycles. And it’s more comfortable for tall passengers to not bump their head.” His eyes flickered up to the rearview mirror.

“Like you...”

“And I guess it’s pretty cheap?”
“Yeah, not bad. About 150,000 CNY. And it has a good range for a small city car…about 370 km.”

Later I found the car model pretty easily after a quick search. It’s the BYD D1, purpose-built by BYD as a vehicle for rideshare drivers. Although I had never seen one, it’s probably just because I wasn’t paying attention, because two days after I got back to Shanghai, I saw one…I think. Or at least a vehicle that looked very similar:

Le Shifu went on: “I remember when BYD released their first car models. Wow, they were ugly! No one bought them. But their first NEV vehicle…it was so great. That’s already 10 years ago. Now everyone knows they are attractive, good value…and cheap.”

(Here, I assumed Mr. Le was talking about the BYD Qin, which was released in 2013, and wasn’t actually their first NEV, but it was the first one anyone cared about).

"Yes, BYD is cheap. Actually, BYD is so cheap that it’s creating problems for other EV manufacturers in China. They have to lower their prices when BYD lowers its prices, or they can’t survive. The EV sector is so juan." (卷, meaning hyper-competitive, sometimes destructively so].

"BYD is the most 卷! I heard they are not hiring any graduates this year. Yesterday, some engineering students in my car said they hoped to apply to BYD this year, but they are not taking anyone. They also mentioned applying to some company in Ningde, but also too hard."

"Ningde? I guess they mean CATL. It’s also a new energy company…that makes batteries. The battery industry is very hot right now, but also very 卷. Maybe they should just apply to join the civil service."

"Civil service jobs are even more competitive! But in a different way. For the civil service, it doesn’t even matter how good of a student you are…if you don’t have the right connections you won’t get a job. You have to be good at sucking up…following orders...good at that style of working. The civil service is not suitable for everyone."

"Well, I entered the clean energy industry 10 years ago, so I don’t envy the students finding jobs in this competitive environment now. I analyze some of those 卷 companies, but my own field of work itself is not so 卷. There are only a few companies that do what I do."

"Do you want to go back to the USA?"

"For my industry? No way. The policies can change completely overnight…from supportive to not supportive...a complete reverse. The entire strategy for what is most important in the USA can change every 4 years if there is a new president. Maybe my industry would suddenly have no support because of some political reasons. If that happened, it would break my heart. I have some big ambitions for clean energy. In China, I can always get closer to my dreams. In the USA, I would probably have my heart broken"

"Americans don't use EVs?"

"They do, but some people don't like them for culture reasons. And they are always worried…about their range, about safety, about availability of charging stations. Do you remember when China didn’t have a lot of EV chargers at the beginning?"

“Yes, you had to wait in line to charge your car. Very annoying.”

“To build a charging station requires a lot of capital, and you are hoping enough people will buy EVs and eventually use the chargers to recover your investment. You need a lot of trust…or the government should have some policies to remove the risk from your investment. But for that to happen, it means the government really wants it to happen. China really wants this, so you see so many EV chargers recently.”

"Why does China like EVs so much?"

"Well, it was part of your national plan, to use more EVs and reduce ICEs, reduce gasoline and petroleum usage. China likes to make long-term plans."

"Oh yes, China is good at planning. This is a gift from Chairman Mao."

We arrived at the ferry terminal, where I was going to take a boat to Gulangyu Island , so we said farewell to Mr. Le and headed inside. But I did ask if he could stay parked for a moment, so I could take a picture of the BYD "Little Green" you saw at the top of the essay.

Fun extra fact:

Le is a very uncommon surname in China. The character is 乐, which as a surname usually pronounced “Le” like in 快乐 (kuaile, happy), not its other pronunciation “Yue” like in 音乐 (yinyue, music), but apparently even this can also vary by region.

I believe this is the first person I've ever personally met in real life with this name, however, there is a well-known Chinese entertainer and TV host with this surname: Mr. Le Jia (乐嘉). Baidu Baike says its prevalence in the Chinese population is 0.015%, meaning there should be just 210,000 people in China with this name. A rare name indeed.

In the north of Xiamen's main island in Huli District, just west of the airport, is Dianqian Community, one of Xiamen's last urban villages (and its largest). Urban villages are called 城中村 (literally: village in a city) or sometimes 村子.

Urban villages can be found in large cities in southern China especially, and are often described as China’s “ghettoes” or “slums”. This is not quite correct in my opinion, and the topic deserves a separate thread. But they are indeed generally home to people with lower incomes.

Dianqian has gained fame in recent years on social media as a place of pilgrimage for aviation enthusiasts visiting Xiamen. It lies immediately beneath the final descent path of airplanes arriving at Xiamen’s Gaoqi Airport, offering unique photography opportunities.

When I visited, Qingshui Temple was a particularly popular spot for photos. An elderly man held down the temple entrance, selling incense and advising tourists when the next plane was about to arrive from a schedule he has on his phone. Throughout the day, planes roared overhead every 5 minutes or so.

Nearby was the main commercial street of the urban village, choked with motorcycles, full of cheap food and daily goods. Vendors hawked melons and Nikes from trucks. It was crowded and loud. The air was humid and dense with odors of cigarette smoke and overripe fruit.

Heading into the side alleys off the commercial street, I found the residential quarter of the urban village. The multi-story apartment buildings are called “handshake buildings”, constructed so densely only a bicycle can pass among them. They are dark and humid and claustrophobic.

In some sections of the village, only a sliver of daylight is visible looking straight up. The constant dripping of AC units and the lack of sunlight ensures the alleys are permanently damp and mildewy. But the alleys are still buzzing with people, smoking, sitting, walking.

Nearly every building had one or several signs advertising rooms or apartments for rent - youth apartments, worker housing, and cheap hostels. I saw many delivery drivers and construction workers - this is clearly the part of town to live in if you're hoping to save on rent.

I also saw small fabric sweatshops - in the literal sense of the word (I don’t think I saw AC units, just fans).

I know this isn’t typical of China’s fabric or sewing industry - it looks like *very* small scale businesses taking advantage of low rent in the village.

I’d really wanted to chat with someone about their life in the urban village, but it seemed most people were busy with somewhere to be...

There were a few elderly people sitting around I could try to talk to, but so far I’ve had very limited success chatting with older people in Xiamen because of the language barrier...I can barely understand their heavily Minnan-accented Mandarin. So, I decided to use one of my typical plays to find someone to talk to: I stopped in a small grocery store, bought an ice cream, and complained a bit about how hot it is and how tired I am (it was 35 degrees out and a million percent humidity, so this was real enough).

After accepting the seat and the invitation to “rest for a while” that was promptly and courteously offered to me, I now got to chat with the shop owner for at least for the duration of one ice cream. Never fails.

It’s a family-owned business. There’s a wife and husband proprietor duo, along with their daughter who looks to be in her 20s, bouncing a baby on her lap.

“So you live here, right? You don’t rent?”
The wife answers
“Yes, we’re residents”

“Is it cheap to rent here?”
“There are cheap places...and more expensive places. Why...do you want to rent?”
“No, I’m just visiting. I’m just curious about the price. I guess it should be cheaper than other parts of the city”.
“Yes, it’s relatively cheap to rent a small room. It can be very cheap, if you don’t have a lot of requirements.”
“You mean like...single room, without AC?”
“Ah...most apartments should have AC...”

An airplane roars overhead.

"How can you sleep or rest with the noise of the airplanes? I came to take pictures of the airplanes, and that's fun, but I don't think I could live here"

The daughter smiles. "I'm used to it...I grew up here. It feels comfortable to me. I can't sleep in other places...it's too quiet..."
"Do you think Dianqian could be demolished in the future? I saw online many of the old urban villages in Xiamen are being demolished or renovated."
"Yes, that's true, but not here yet."
"Do you know why? Is it because it's too close to the airport, so it's less attractive for redevelopment."
"I don't know..."
"Or maybe it's intentional, to ensure there is a place for low-income workers in Xiamen have a place to live? I saw a lot of delivery drivers in the alleys outside."
"Yes, maybe. It's very cheap and convenient to live here, even though it is a little loud."

A small boy emerges from the back room and sits at the counter. He looks about 5, eating something, and playing a game on an iPad. He glances up at me, back down at his game, and then does a double take to stare at me.

“Hey, is he learning English already in school?” I ask.

"Yes" his grandmother answers. "Children learn English in primary school now. His English is better than his mother's already."

Her daughter laughs self-consciously.

"Hey" I call out to the boy in English. "How old are you?"
He looks up at me, then smiles nervously back into his iPad, avoiding eye contact.
"I'm six...years...old."

"Wow, six years old! What's your name?"
He continues smiling shyly, pausing for a long moment as he carefully gathers his thoughts and stammers them out, almost whispering.
"My...name...is...Adam..."

"Nice to meet you Adam! My name is David."
"Nice to...meet you..."

His mother and grandmother beam proudly.

"How old are your children?" the mother asks, continue to bounce her second child on her lap.
"Oh...I don't have kids yet."
"Really!?" She seems surprised. "Are you married?"
"Yes, married. But no kids. Are you surprised?"
"I mean...how old are you?"
"I'm 35."
"Oh, not too late. That's okay then."
"Ahh, I know why you were surprised. You think I look old! And you are surprised to hear I'm only 35!"

"Ohh no noo. That's not what I meant. I mean...we get married and have children relatively young here in Fujian. I was born in 1998. But you are not too old."
"Oh come on, I know Chinese people think foreigners look older than we really are. It's okay, I'm used to it."
"Haha well...foreigners do tend to look...mature. But I don't think you look old."

A delivery order of drinks arrived, and they got busy, plus I'd finished my ice cream, so I decided it was time to leave. I thanked them for the time, and the ice cream, and headed back into the alleys. I was 10 minutes back to the metro station, and then 45 minutes back to downtown.

I previously lived in Shenzhen, which also has urban villages, but Dianqian felt quite different. In Shenzhen, the urban villages may be right in downtown, and although they're also messy and crowded and full of low-income tenants, the real estate is also extremely valuable. By contrast, Dianqian is located far from Xiamen’s downtown, next to an airport that limits land value and redevelopment potential. At the end of the day, it's still basically working class, so not really a "slum", but it's perhaps the closest you'll get in urban China.

Checking online, I found single room rentals in Dianqian go as low as 600 CNY/month (and 300 for the first month) which could be attractive if you're a migrant worker arriving in Xiamen with nothing in your pocket, looking to minimize expenses and maximize what you can save. You'll pay 2x more just one subway stop away.

And if you go online and look up discussion of urban villages, commentary on their destruction is often quite negative. People don't like the disappearance of super-affordable housing, even with its drawbacks, and worry what happens to the locals. A topic worthy of nuance.

Thanks for reading!

"Hey, American! Come take pictures of this. See how our women pick potatoes!"

Early in the morning after checking into my guesthouse in Jiuzhaigou, I decided to explore the dirt path up the mountain behind the Tibetan village - the way to the old hamlet I'd heard about the night before.

It was breezy and the sky was blue with wispy white clouds. The air still held a touch of the night’s chill, which yielded as the sun rose and lit up golden corn in the fields. I cannot imagine or describe a more perfect late summer/early autumn morning.

After 15 minutes of hiking, I heard a babble of voices from a field somewhere above me, but there was a steep soil embankment blocking my view, with a wooden ladder leading over the embankment. As I made my way past the ladder, a woman’s head popped up over the embankment...

"Hey, there's a foreigner!" she called out. Like a cartoon sketch, three more heads popped up over the embankment to gawk at me as I passed, wheezing slightly, thanks to the 2000m of altitude. I smiled and waved at them, but kept moving. Ahead, I rounded a bend and spotted weather-beaten grey stone-and-wood houses.

There were only about 15 houses in the hamlet and most of them were already in pretty bad shape. These are the houses that are left behind after poverty relocation efforts see the former residents moved to more modern accommodations.

The buildings were very simple, with stone foundations and walls, wooden doorframes, and an enclosed courtyard space.

A few of them were in better shape though, showing evidence of some recent renovation work done during Covid - as I had found out while talking to the hotel desk attendant the day before. If I owned a property here, I'd definitely want to keep it in decent shape too. To enjoy the views at least…

I circled through the hamlet and found myself back on the fields again, but this time I'd emerged above the group of locals I’d passed before, whom I could now see were digging potatoes. It might have been the most aesthetically attractive potato patch I've ever seen.

They sat to take a break and one guy spotting me standing there waved me over.

"Hey, where are you from?"
"I'm from the USA. I'm traveling here. Is it okay if I hike in your village?"

"Yes of course, no problem. Do you want to pick some potatoes?"
"Pick potatoes...? No thanks..."

“Do you have potatoes in America?”
“Yes of course. Actually I grew up on a farm, picking potatoes. I think I already picked enough potatoes for my lifetime!”

“How do you pick potatoes? With a rake like this?”
“Yes, we use a similar rake. I also used my hands to dig potatoes.”

“Your hands? The dirt is too hard!”
“We pile the dirt up into mounds before planting the potatoes, so the dirt is looser and easier to move in the fall, when the potatoes are ready. We are worried about piercing the potatoes if using the rake too aggressively.”

“Oh I understand. We’re going to start work again now. Hey American. Come take pictures of this. See how our women pick potatoes.”

Break time was over and everyone got back to their tasks. Younger women swung two-pronged rakes to pull up the hard dirt, revealing long root systems and plump yellow potatoes. The older woman sorted and packed them into canvas bags, while the men took turns hoisting full bags of potatoes onto their backs and slowly staggering uphill to where I assumed their storehouse was. They told me they’d been out here since 6am, right after the sun rose.

I heard a mix of Tibetan and southwest (Sichuan) dialected Mandarin, so I guessed it was both ethnic Tibetan and Han people out, here digging potatoes on the slope of the mountain today.

I left them to their work and continued on the uphill path through the hamlet as their voices faded out behind me. Within a few minutes, I had passed the last house and was looking down at the hamlet, the valley below, and the mountains across the valley.

While most of the houses seemed abandoned, there were still a few signs of life. This house had chili peppers drying in the open air on its second floor. The cuisine in Jiuzhaigou is Sichuan mixed with Tibetan, and it features a lot more heat than the Tibetan cuisine in Gansu Province, from which we had just arrived the day prior.

Despite the awkward location, the hamlet still WAS connected to electricity, and I could see the power lines continuing even higher up the mountain as well, perhaps connecting to even more-remote hamlets.

The mountains on the other side of the valley were also covered with power lines, suggesting more small villages hidden in those hills as well, probably populated only by a few elderly stragglers at this point.

Up behind the village was a small religious area, a grove of trees covered with prayer flags flapping in the morning breeze with a small Buddhist shrine (chorten) made of rock. Passing through, I continued hiking uphill another 30 minutes, with frequent pauses to catch my breath...

I had heard that much higher up the mountain was another old village site that is now completely abandoned and hoped to explore it. I was planning to climb up as long as it took to see the old houses, but then I met some locals who told me yak herders had already pulled down the old abandoned houses to create more grazing area...

Denied the opportunity to see the old houses, and with hunger and caffeine cravings starting to catch up to me, I decided to head back down for breakfast.

On the way down, I found the group of potato pickers en-route to another field, gathered around a tree in the hamlet, trying to knock down apples for a snack.

I offered my 1.85 meters to assist, and successfully snagged a few apples for everyone - and myself too. The guy chatting from earlier pressed something into my hand.

“Have you ever had fresh raw walnuts? Try them...they’re good. You can use a rock to open them.”

Walnuts and apples become my morning hike’s bounty:

My short peaceful morning hamlet excursion with the potato pickers ended up being a highlight of my time in Jiuzhaigou. Jiuzhaigou is undoubtedly touristy and commercial, but if you're looking for something a little more untouched, just ask, and you can surely still find what you seek.

Thanks for reading.

Part 1: Meeting Boss Liu Over Yak Hotpot

On our last night in Sichuan Province’s Jiuzhaigou scenic area, we decided to try the restaurant at our local guesthouse. Actually, it wasn’t a restaurant…more like a common room with wooden tables, benches, and a huge Tibetan cookstove that was used for warmth, meal-prep, and warming butter tea.

That night, the guesthouse boss was finally back from his trip to Chengdu. The front desk attendant with whom I had I previously chatted is his nephew. His surname was Liu (that is, his adopted Han surname).

Boss Liu had one of those personalities that fills up the entire room. He had a short buzzcut and was loud, boisterous, and infectiously good-natured. He was an uncle’s uncle, ever cheerful and full of tall tales the wise listener knows to enjoy but never take at face value...

We told him it’s our last night in town, and that we’ve been wanting to try Tibetan hot pot here, but we’ve been put off by the touristy restaurants in the town, so we decided to eat at the guesthouse tonight.

“Well then, you came to the right place! We are an authentic Tibetan family and we make authentic Tibetan food! You need to order the yak hot pot! This one, here!”

His finger stabs at the menu, indicating a massive copper pot mixed with fresh meat, cured meat, and vegetables.

“Will it be too much?” I ask with concern. “It says it serves 2-3 people. Recently in Xiahe County we ordered too much at a Tibetan restaurant. The waitress didn’t stop us…we even asked if it was too much and she said it was ok. As a result, we over-ordered, and we wasted a lot."

Boss Liu is aghast. "They let you order too much and waste food?! That’s not right!! You are traveling, it’s not convenient for you to take leftovers home"

"Yes, that’s right. I felt very bad about wasting food. I guess they only care if we order a lot…not if we eat it all."

"Well, they definitely weren’t Tibetan!” He’s bubbling with indignation now.
"If they let you waste food, they’re not true Tibetans! Don’t worry, this is the right amount of food. Order just this and a barley flatbread (青稞饼) and it’s enough! Are you drinking?"

"Yes, I’ll have a beer. The highland barley one."
"Okay. Do you want your hot pot to be spicy?"
"Uhm…is it normally spicy? I want authentic flavor. Not spicy for the sake of it."
"Then it can be a little spicy, but not so spicy like in Chongqing. Actually, Tibetan hotpot is usually made with clear broth, but since this is Sichuan, we have the habit of making it a little spicy here. Come make your dipping sauce for the hot pot.”

He motions to a sidebar that’s set up with condiments.

"How would you prepare the dipping sauce?" I asked. "I think this is what shows the local characteristics in China. Many places have hot pot, and the preparation is largely similar with small variations…but the key lies in how you prepare your dipping sauces."

"Ah, you understand!" he roared with approval. "You are a true food lover, like me! You’re right…I’ll make one for you that is perfect for Tibetan yak hot pot. Some chillies…some cilantro…also some sauce…and then some of the liquid from the pot." He ladled the simmering broth into my bowl.

"Eat it like this. Eat the beef first while it’s tender…it will overcook if you leave it until the end."

We tucked into our food (which was delicious, of course) while he joined us at the table and peppered me with questions, full of energy and apparently delighted to chat.

"Hey, your Chinese is great! If I close my eyes, I can’t tell you’re not Chinese!"
"That’s just because China has so many different accents already.“ I responded. “If I use the wrong word or tone, people might just think I am from a place in China with non-proper Mandarin, like…Sichuan!” I tease him.

“No, no, your Chinese is really very standard, with a Beijing accent. At first I didn’t want to talk to you, because I was afraid you wouldn’t understand. But it's very easy to communicate with you. Last time I had foreigners staying here. Two girls from Germany. I had to use the translation app for everything. It was so tiring! When they arrived, I told them I needed their passports for the public security registration and they didn’t understand why. They were so suspicious, ‘why do the police need my passport??’ and I couldn’t explain clearly. The more I explained, the more confused they were."

"Haha, I guess you were the first place they visited in China.” I replied. “If they had been anywhere else, they'd expect it. Actually, in the places I just visited in Gansu, several hotels didn’t know how to register me, because I was their first foreign guest, but they took me anyway."
"Yes, Gansu has fewer tourists compared to here. But it’s a good place to go. Their Tibetan culture is very pure there. You can go to see the horses. You know this is Amdo…you know Amdo right?"
"Yes, I read about Amdo online"

"Right, there are three Tibetan regions, each with their own of three treasures. The treasure of U-Tsang is religion…the religion is very deep there. The treasure of Kham is the people…they are very strong and hardworking. And the treasure of Amdo is the horses…we have the best horses."

As a side note, I can’t find any evidence of this particular breakdown of “three Tibetan treasures” in my online research. There are “three jewels” in Tibetan Buddhism but they refer to something else entirely. This might have been Boss Liu riffing a bit…an entertaining but perhaps not fully reliable narrator.

“Yes, we saw horses in Gansu. I haven’t seen many here. I guess there are no grasslands...”
“The herders ride horses. But there are not many herders here. You have to go to Ruoergai County to find grasslands.”
“That’s where we came from…we took a dangerous mountain road."
"Oh, I know that road. Many rockfalls! We can’t have a highway coming through this area, because this is a UNESCO site. One of the conditions for its inclusion in the UNESCO registry is that we must protect the natural environment, so it’s not possible to build a highway. Now, you can drive an hour through the mountains to the new highway being built in Jiuzhaigou County, but we’ll never have a highway here."

I didn't know it at the time, but I would end up taking this highway the next day, after I missed my train and had to drive to Chengdu!

I’d been stuffing my face all this time, but it seemed the volume of food in the pot had barely budged. I was getting a bit worried about whether we could finish, and I told him I was worried about waste.

"No problem, take your time! Eat as much as you can. Hey, after dinner, do you want to see my house?"
"Your house? Oh no…we couldn’t. I don’t want to intrude."
"No problem, I take anyone who wants to see Tibetan culture to visit my house. All the tourists like to see my house. I built it in the traditional Tibetan style to show them. It's very beautiful!"

"Okay, sure then..."

Part 2: The Secret Tibetan Culture Museum

And so...after dinner (I ate all the meat, but I'm sad to report I couldn't finish some of the vegetables) we took the short walk over to Boss Liu's house, the stately mansion next to the guesthouse, full of anticipation, hoping to be impressed and amazed.

..and boy did it deliver! Entering the front door, this was the first thing we saw:

It turned out, Boss Liu had turned his huge personal home into a museum, a veritable monument to Tibetan art, culture, and craftmanship.

Not a single square inch of wall or ceiling space had been wasted. I hardly knew where to look, because everything was so interesting. His main living space on the first floor featured another huge stove like the one in the guesthouse, with dozens of small niches around the walls to hold dishes and ornaments.

“Come here, to the traditional living room. This is where the family gathers for holidays, when we’re all back at home. Men sit in this corner, women, sit in that corner, and children sit on that side. But we also have another, more modern living room on the second floor.”

"When we're sitting together, we can all have tea and chat. See this traditional teapot…that’s green bronze. It’s handmade…all of the decorations in my home are handmade. I am still decorating it…it’s not done yet, even after 20 years."

"When did you build this house?"
"It was relatively early... in 1998. I've been decorating it slowly, for 20 years. All these murals are hand-painted…

..and all the wood carvings are done by our master woodworkers. I called the best masters from everywhere to make art for my home."

I started to suspect Boss Liu might be a little more than a regular village guy. This house is a little too ornate. I decided to try to find out…tactfully…

“So…this was a very big house that you built, all the way back in 1998. And these decorations are not cheap either. Even at that time, you had a lot of resources..?”

Liu answered carefully. “Well...I have some background. My father was an old soldier. And I am retired from the PLA too. So…our family has some resources….”

Ah. This makes more sense now. Judging from the ornateness of this house, and his uncharacteristically cagey response about his family background, I’m going to guess dad wasn’t just regular PLA…he probably had some rank. I’d bet Boss Liu did as well…

It also makes more sense now why his guesthouse lobby is filled with an impressive collection of old political posters like Mao shaking hands with Tibetan leaders and a framed declaration of the anniversary of the establishment of the Tibet Autonomous Region. Boss Liu appeared to come from old “red” stock.

In the main living area, I noticed a shrine with an eclectic mix of photographs: Mao speaking at the 9th National Congress in 1969, an image of an unfamiliar man on a light blue background I assume was Liu’s father, and a curious photo on a dark blue background of a religious leader unfamiliar to me.

After some sleuthing, I found this person is Lungtok Tenpai Nyima, former spiritual leader of the Bön (苯教) Tibetan folk religion who died in 2017. On Wikipedia, it says Bön is “the indigenous animist and shamanic belief system of Tibet...revolves around the worship of nature and claims to predate Buddhism.”

I went down a rabbit hole on Bön while writing this essay, and I’m not going to lay it all out here, but you are free to do your own research. According to official materials from the Tibetan Autonomous Region, there are around 130,000 practitioners of Bön in the TAR. I couldn’t find any details confirming whether these numbers include other Tibetan regions like Amdo, where they are supposedly more concentrated. But any way you look at it, it’s definitely a very small minority religion, and quite the surprise, to discover my guesthouse owner is apparently a practitioner of a minority Tibetan folk religion. Apparently, there are considerable overlaps between Tibetan Buddhism and Bön, but they ARE distinct belief systems. Supposedly the religious belief systems of China’s Naxi and Pumi ethnic minorities are closely related to Bön. I guess the reason it was surprising to me is Tibet and Tibetan Buddhism are so closely linked in my mind that I never really considered the possibility of a Tibetan folk religion that predates Buddhism still having adherents today.

Back to the tour. Standing in a corner of the living room, Boss Liu adopted the air of a stage performer. “Now, are you ready for the surprise?” he asked theatrically.

He pushed on the wall and suddenly a secret passageway appeared!

"This is the way to the third floor, where the shrine is. Come upstairs. Get your camera ready to take pictures...and videos! You must put these online and show Americans our Tibetan culture!"

The third floor turned out to be a elaborately decorated (but unfinished) shrine, with luxurious carvings, murals, and hardwood floors.

Boss Liu scampered proudly around the room, instructing me where to take pictures, and giving me the backstory of each feature and its details.

This carving was done by a master woodworker from Lhasa...
That painting was done by a master painter from Qinghai...

These alcoves in the wall will eventually hold 1000 images of Buddha...

This wood carving is still unpainted, but he has invited the master to return soon to finish the work…

On and on, a thousand small details to be admired and appreciated. Even the curtains on the balcony were special and noteworthy, hand-braided from yak’s hair. (“let me tell you, the yak is a treasure! Every part of its body is a treasure!”)

His house is truly a museum. Boss Liu even insisted I return to the entrance of the room and film a slow walk-though to capture the space. I was narrating, but unfortunately my phone's mic messed up, so it's silent. But you can get the vibes anyway...

Our tour ended, and Boss Liu brought us back to the entrance, where he demanded we admire one last thing: his wife’s plants.

"Look how special these are! My wife specializes in succulents. Ask anyone who knows about succulents and they’ll tell you these are amazing!"

“She’s been raising some of these for 20 years! Just ask anyone….they’ll tell you these are very impressive! We even had some professors from the agricultural university come here to study her succulents!”

I don’t know anything about succulents, but I’ll take his word for it.

Anyway…what a guy. Big Boss Uncle Liu, living his best life.

An aggressively proud Chinese (retired PLA with a lobby full of red paraphernalia) AND an aggressively proud Tibetan (has a literal museum of Tibetan culture in his house) with seemingly no identity dissonance between the two, not that I detected anyhow.

I also thought it was cute how proud he was of his wife's plants. I've never seen a man so excited about succulents before. I wish I had gotten to meet her

I’m not qualified to comment how common this view is across places I haven’t visited, but you have to imagine this is Beijing’s ideal vision for a Chinese cultural identity that is both unified and diverse at the same time - without clashes or contradictions. Oxen of two colors, plowing the same earth, so to speak (Liu used this metaphor on the tour while pointing out this mural below).

So if YOU go to Jiuzhaigou, I say...skip the hotels and stay in a village house. Chat with the locals. Hike to the old mountain hamlet. Eat the yak hot pot. Drink the butter tea. Ask what they think, ask how they’re living, and then listen. It’ll always be worth it.

That’s it for this one! Thanks for reading. I’m trying to quickly get caught up posting all these old essays from the last 12 months so I can start the year with a clean slate, so to speak. So…expect a lot of content coming in the next week or two!

Exploring Rural Linfen

I began Day 2 in Linfen as auspiciously as possible: with a late breakfast of a spicy local specialty - Linfen beef ball noodle soup. Then, I set off to rural Ji County in the far west of the city’s administrative area. Compared to downtown Linfen, Ji County (吉县, lit. “auspicious county”) has a much stronger economic identity, with both a signature local product - apples - and a powerful tourism draw: the Yellow River Hukou Waterfall. Contrasting the uncertainty I concluded with for the development of Linfen City, Ji County’s future gave me more reason to be optimstic.

The waterfall is the only national 5A-rated tourist attraction in Linfen, which is actually a Big Deal. There are only 358 5A-rated tourist attractions in China. Think of them as you would think of National Parks in the USA - a powerful tourism draw often sufficient to economically sustain an entire region via dining, accomodations, and other consumption spending. While some large cities may have three or more, (Beijing has nine) most cities don’t even have one. Hukou is the largest waterfall on the Yellow River and the second-largest in China.1 Here, the Yellow River forms the border between two provinces, so if you visit the waterfall from the Shanxi side, you’ll also see tourists in Yichuan County, Yan’an City, Shaanxi on the opposite bank.

For the 90-minute drive to the waterfall, and to conveniently explore more of the county at the same time, I decided to rent a car. Like most small cities in China, the cost of services in Linfen is generally calibrated to the lower local wages, which means they’re very affordable in my eyes, coming from Shanghai. My daily car rental from a national chain (Shenzhou) came out to just 108 CNY (15 USD) including fees and insurance.2

Leaving Linfen to the west, the flat terrain continued for about 20 minutes of driving on the G22 expressway. The primary crop here is corn, but the harvest was already complete, so only rusty, withered stalks remained in the fields. Combined with the overcast skies, the atmosphere would have been quite gloomy, but for the bright pops of yellow from the ginkgo trees planted along the road - and as windbreaks dividing the fields - a single spot of vibrancy in an otherwise somber landscape.

Suddenly, mountains appeared ahead of me, starkly jutting up over the broad plain. I entered a long tunnel, and when I exited, I was now in the mountains, following a small river along the gorge it had carved into the terrain. The fields disappeared, replaced by steep hillsides and rocky slopes. My GPS told me I’d entered Xiangning County, a coal-mining region of Linfen. The car’s temperature gauge showed the outside temperature had dropped by five degrees.

The terrain showed the evidence of decades of mining. Even from the expressway, I could see many muddy dirt trails disappearing into the hills, undoubtedly leading to mines. Deep channels formed by rainfall pouring down the hillsides made it look like the mountains - the earth itself - had been scratched to the bone by giant claws. There were few actual villages, other than mining camps and coal processing facilities, half of which lay abandoned and rusting…

Tourism Booms in Hukou Town

Emerging on the other side of the mountains in Ji County’s Hukou Town, the terrain dropped rapidly once again, descending into the Yellow River valley gorge. The sides of the road were stained black from coal falling off trucks, which I had seen before in Ordos, but the town was active and lively. I saw signs for the waterfall everywhere, but also posters advertising a major local event: the 6th Annual Linfen 520 Yellow River Heroes Challenge, an offroad rally racing event hosted in the town that had just kicked off. Hukou’s streets were clogged with modified off-road vehicles, both driving themselves and tugged on trailers.

As I drove through, it seemed like every restaurant was full of people, and surely every hotel room in the small town was booked too. Online, the event organizers claimed 500 racing teams from around the country were registered for the two-week event, competing for a total prize pool of 1.6 million CNY. Coming from a rural part of the world myself, I appreciate keenly the economic importance of these kinds of local festivals - and felt an unexpected pang of nostalgia and kinship, observing the sound monetization of good ol’-fashioned rural redneck fun in such an unexpected place.

At the Hukou Waterfall, the riverbed of the Yellow River suddenly drops 20 meters, which isn’t that much, but at the same time the river abruptly narrows from 300m wide down to just 50m, compressing an enormous volume of water into a relatively small area. The result is the waterfall, a booming wall of water, roaring and surging down the river bed, tossing spray and foam into the air. I visited in the middle of the week during the off-season, so most of the tourists here were elderly people on package tours, although I spotted a few families with younger children. I wondered why they weren’t in school…

“Dad, take a picture with me!” A young girl who looked about 10 years old pulled her dad into position for posing, while mom waited with the camera. “Like this, hold your fingers together in a heart.” Dad complied, but he was doing it wrong. “Nooo, like this! Finger and thumb!” Dad tried again, this time apparently sucessfully, and the picture was taken. “Okay, now we make a heart with our hands together. I make one half and you make the other half.” Dad tried, but he offered the wrong hand.
“Daa-aa-aad. You’re doing it wrong!” Dad laughed haplessly in confusion and mom had to get involved to position his hand in the correct position to form the other half of the heart. Finally this picture was taken too, and the girl squealed in delight at the (apparently acceptable) result.

After staying about 15 minutes, I made my way towards the exit. The waterfall was fine, but unless I wanted to take a picture with a donkey, there was not much else to do. A middle-aged lady approached, phone in outstretched hand, and I braced myself to obligingly take fake cheerful selfies with a stranger, but she surprised me. “Hi…” she said haltingly in Chinese. “I’m…also traveling by myself. Would you…take some pictures of me in front of the waterfall?” Pleased that she wasn’t asking for a selfie, and doubly pleased that she asked me in Chinese, instead of exaggeratedly miming the action and forcing a few words of broken English, I was more than happy to oblige.

The National Highway to Ji County

Leaving Hukou Town, I took the National Highway 309 towards the Ji County seat. The route followed the Yellow River gorge, and the views were great, but I needed to concentrate on the road, which was clogged with heavy trucks carrying coal, straining up the hills, which made overtaking on the winding two-lane road a dubious prospect, except for the brave - or the foolhardy.

Along the way, I found something quite unexpected sprouting out of the cliff faces alongside the road: cave homes. I knew cave homes were still a thing in the loess plateau regions of Shaanxi and Gansu, and in retrospect it made sense that I’d find them here too, but I still wasn’t expecting for my day trip to morph into a cultural site visit as well.

The pathway up to this cluster of three caves was overgrown and they had obviously been abandoned for many years. I noted they previously had electricity, and while the accomodations were humble, they seemed basically similar to any other small home in these rural areas - except of course that they were caves built into a cliff face. The door to one house was open, and the style of decorations suggested the house had probably been abandoned for 20+ years.

There were a few scraps and bits lying around the house still, and I saw a traditional wood-burning stove often used in rural kitchens (not sure where the smoke goes). Interesingly, the original inhabitants had apparently been Christians - at least assuming from the pictures of Jesus on the wall flanking the faded wedding photos of the cave’s previous occupants. As recently as 2010, an estimated 30 million people still lived in traditional cave dwellings like these across north-central China. Unfortunately, more recent statistics eluded me.3

Continuing towards the Ji County seat, the chunks of coal alongside the road gradually disappeared, replaced by something else clearly falling from trucks instead: apples. Ji County is famous for its apple production, with up to 80% of its arable land dedicated to orchards. Although Shaanxi (not Shanxi) is the largest producer of apples in China, Ji County is directly on the border with Shaanxi and enjoys a very similar soil composition and climate. This, combined with its relatively small population, (fewer than 90k people) means it has one of the highest apple production rates per capita in the country.4

The end of the apple harvest season had arrived by the time I visited, and so it seemed just about every other building on the outskirts of town had converted into a temporary receiving area for bruised apples. I stopped the car by one massive pile of apples to survey the mound, identifying several unique cultivars. Up close, I could see that a few apples were completely rotten, but most of them looked to be just bruised - probably windfalls. I even found a few undamaged apples and tried a bite of one. It just tasted like a regular apple - crisp and sweet.

An older woman emerged from the nearby building and suspiciously asked what I was doing, making me immediately self-conscious. She didn’t seem particularly friendly, and I suppose that was quite fair; after all, I was the foreign stranger snooping around her apple pile taking uninvited bites. In my defense, I was assuming the apples were intended for animal feed... I told her I was just a tourist passing through and tried to ask a few questions.

“Hey auntie, what will you do with all these apples? Are they to feed to pigs?”
”No, they’re for juice. Someone will come to collect them and press them.”
”Ohh. Heh. Er…I thought they were for animals. Sorry…I took a bite of one. So you buy leftover bruised apples from local farmers - the ones they can’t sell to grocery stores?”
”Yes”
”You pay by weight then?”
”Yes”
”How much do you pay for a kilogram of bruised apples?”

She peered at me with the utmost skepticism. “Why do you want to know this?”
”Just curious.”
”Curious! Hah.” She shakes her head.
”It’s not convenient to tell me?”
”No, I can’t tell you.”
“Is it a secret? She continued to shake her head.
”Okay. I won’t ask for any secret information then. Sorry, auntie!”

I returned to my car and continued onward, laughing about the awkward exchange. What’s so secretive about the purchase price of bruised apples? What if I actually had a bag of bruised apples to sell…?

Exploring the Ji County Seat

Just a short way further up the road, the Ji County seat appeared. Only slightly larger than a big town, it really only comprised a few main roads and some smaller alleys. School was getting out and the roads were clogged with children. A plaza in the town center hosted several pushcart vendors, attracting throngs of children. I parked the car on the side of the road and proceeded on foot, attracting wide-eyed stares from the kids. Not a lot of foreigners make it out here I guess. On the side of the main road, I discovered an old, seemingly abandoned village house, completely blocking the sidewalk. This is a nail house, or dingzi hu (钉子户) in Chinese - someone unable to reach an agreement with authorities on an appropriate compensation level for their home and thus remaining a holdout, resisting the tide of urban development. Fortunately for the municipal authorities, this one was was only blocking the sidewalk, not the road, so it wasn’t terribly disruptive and seems to have just been left there.5

I found a grocery store and stepped in to assess the offerings. While it was obviously smaller in scale than something I would see in Shanghai or Shenzhen, the layout, options, and design were all very similar to grocery stores in larger cities. I’d invite anyone inclined to claim modern or clean grocery stores are only a feature of large cities to visit Shenglong Supermarket of Ji County, Linfen!

I didn’t really need anything, but after wandering aimlessly around the supermarket for a few minutes, (and attracting many more stares) I decided to buy some winter dates, (冬枣) which I knew are local products. Of course they were much cheaper than they would have been in Shanghai - just 7.6 CNY per kilogram.

Back on the street, I decided to find a local fruit vendor to procure some local apples. From online research, I had seen Ji County produces many different apple cultivars, so I was looking forward to trying something other than the ubiquitous Red Fuji apple, standard across supermarkets nationwide and surprisingly popular with Chinese consumers despite its (to me) overwhelming - even cloying - sweetness. I was particular keen to try something a bit more exotic and tart, either the Xiangfei apple or Ruixue apple, which I knew were grown in Ji County6. Unfortunately, when I asked a fruit vendor if she had any Xiangfei or Ruixue for sale, she laughed in my face.

“Come on, you think we could sell those apples here in this town? Those cost 40 CNY per kilo! No one here is paying 40 CNY for a kilogram of apples. Those products are only sold in the big cities. Here, I have…Red Fuji and Striped Fuji. They’re both good, from our own orchards.”

Well. I’d come all this way. So I guess I’d better buy something. I got a kilo of Red Fujis and continued chatting with the apple vendor.

“This town is so small. I guess everyone here is local?”
”Yes, we’re all from Ji County. There are not many people from outside.”
Does everyone grow apples?”
”If they have land, then they probably grow apples. As long as it’s flat land, you can grow apples.”
”And it’s all kinds of apple products?”
”Fresh apples and juice mostly. Starting from the last month of August, we will have a different kind of apple reaching maturity each week, ready to harvest. We have 15 kinds of apples in Ji County, all ripening at different times.”
”I see. But now, all you have is Fuji?”
”Yes. They ripen later, and they also can be stored for a long time.”
”Oh. As far as I can tell, it seems Ji County is really only this one big street…”
”There’s two streets total. This street has all the stores and hotels, and that street has all the restaurants. It’s really a small place.”
”How do you feel about living in a small town like Ji County? Is it comfortable?”
”Very comfortable. It’s not like the big city, where going out is so inconvenient, with the traffic. Everything is much more affordable here compared to big cities. From here, if I drive a car for one hour, I can reach Linfen, Yuncheng, even Shaanxi or Henan. In Taiyuan, if you drive a car for one hour, you can only reach the Taiyuan suburbs!”7

She finished weighing out my apples and I paid, but I still had one more question. I pointed across the street to the nail house. “Hey do you know about the nail house? What’s the situation there? Why didn’t they agree to move?”
She rolled her eyes and clicked her tongue “啧啧. I heard they demanded 10M CNY in compensation. Who would give them so much?”

It was getting dark, so at this point I decided to wrap up my exploratory trip through Ji County and head back to Linfen city for dinner. A perfect day of exploration was concluded with a dinner featuring perhaps the most famous dish from Shanxi that isn’t a noodle: 过油肉, or quick-fried pork (lit. “passing the oil pork”). Stuffed and happy, I returned to my hotel to scribble notes and reflect on the day.

Final Thoughts

In my last essay, I concluded Linfen City overall probably wasn’t the best representative of a national median city, but could be taken as a decent proxy for the “post-industrial northern Chinese city trying to reinvent itself” archetype. Importantly, and highly educationally, that model and that observation doesn’t apply to Ji County at all. Ji County is thriving! It never had a really strong coal or steel industry, so it’s not feeling the sting of their decline. Ji County economy revolves around its signature apple industry, which is likely to see increased production as Chinese agriculture continues to modernize and mechanize. It is also very fortunate to have a powerful 5A tourist attraction to drive tourism consumption - along with some clever local festivals.

Furthermore, Ji County doesn’t appear to suffer from the inflated real estate prices afflicting downtown Linfen. I saw brand new high-rise apartments in Ji County listed for as little as 4,500 CNY/sqm, which is well within a normal range for small cities, and half the price of a new apartment in Yaodu District. Used apartments are surely even cheaper. Overall, my trip made me feel that while Linfen might suffer from an uncertain future, Ji County has a lot more too look forward to.

Footnotes:

I just spent several days exploring the city of Linfen (临汾) in southern Shanxi. Linfen could be the most average city in China - at least that was my hypothesis before arriving. But what does that even mean? And why does it matter? In this episode of Off the Beaten Track, I went to Linfen to try to answer those questions.

The Most Average City in China?

Haven’t heard of Linfen? Don’t worry…you’re not alone. I’m a geography buff who’s been living and traveling in China for the past 14 years, and I hadn’t heard of it either - nor had virtually any of my Chinese friends with whom I shared my travel plans in advance. Straddling the Fen River, Linfen (literally: “close to the Fen River”) is a city virtually defined by its unremarkability.

In the annual Chinese city tier rankings released by Chinese financial magazine Yicai, 337 prefecture-level cities are ranked by their overall commercial competitiveness across a wide array of evaluation criteria. Naturally, booming metropolises like Beijing, Shenzhen, or Chengdu score highest, up in the 1st Tier, while mountainous backwaters in undeveloped regions like Tibet or Qinghai score lowest, down in the 5th Tier. The 4th Tier city #169 is the median - theoretically the most middle-of-the-road city in China. In 2025, that city is Linfen.

Linfen is an ancient city - and perhaps was even one of the first capitals of China (if you subscribe to the popular legend that the mythical Emperor Yao located his capital here in prehistoric times). In modern times, however, Linfen’s stature is a little more modest. With a downtown population of 1m and a total administrative population of 4m, Linfen is neither large nor small, neither rich nor poor. Linfen comprises a fairly large chunk of real estate. Like all Chinese prefecture-level cities, it encompasses both an urban downtown area (“Linfen proper”, which in this case is just Yaodu District) plus many distant satellite counties and cities.1 Thanks to its mountainous terrain, traveling from its northwest tip in Yonghe County to its southeast extremity in Yicheng County is a 300 kilometer drive requiring 3+ hours.

Arriving at the airport, I told my taxi driver I was an academic, coming to Linfen to study the development of China’s smaller inland cities. It’s a half-truth at best, but it certainly easier than explaining my odd interest in trying to understand China’s development via social ethnography. “I’m here because I think Linfen is an extremely typical small city in central China - not better or worse than any other cities, but just representative. Do you think this is a reasonable idea?” He chuckled. “Very reasonable. There are many Linfens in China.”

Downtown Linfen indeed has a very typical basic Chinese urban structure. There’s an older city quarter where small shops and rows of haphazardly parked scooters flank narrow, claustrophobic streets, but also an Economic Development Zone with broad sidewalks and broader streets tracing out a forest of glittery glass-paneled office towers. The park-filled greenways lining the Fen River are tidy and new, and the bridges light up with rainbow LEDs at night. There are creaky old department stores with brands you’ve never heard of, but also a new Wanda shopping center with a Starbucks, Huawei, and Li Ning. There are wet markets that stink of overripe fruit and fish guts, and clean modern grocery stores with imported French cheese and bubbling tanks of live Alaskan king crabs…

The streets heave with a frothing stew of e-bikes and gas-powered farm tricycles lugging burlap sacks of apples, but also BMWs and new electric mini-buses. There’s a (restored) Drum Tower, a (restored) section of ancient city wall, and a (restored) God of Wealth temple. You’ve never been to Linfen, but it doesn’t matter, because if you’ve visited a medium-small Chinese city in the last decade, then you kinda already have.

The Invisible Industrial City

“Hey, don’t take this the wrong way…” I told barista Zheng at the trendy coffee shop next to the God of Wealth Temple. “…but I think Shanxi might be the most invisible province in the country. And Linfen is an invisible city in an invisible province…”2

He laughed. “No, you’re right. Shanxi is invisible. No one in the whole country thinks of us at all until the winter comes, and then we get a bit of attention, because then everyone needs our coal. But besides that, we may as well not exist.”

Like many cities of central-north China, Linfen is rich in coal resources - coking coal specifically - which originally attracted the steel industry in droves. At one point in the late 2000s, Linfen boasted as many as 200 steel foundries within its city limits, along with hundreds of coal mines. These enterprises brought economic development - but also severe pollution to its water, soil, and (especially) air. Exacerbating the issue, Linfen lies in a geologic basin surrounded by mountains, which serve to trap the blanket of smog. By 2006, the city suffered from a near-permanent haze, and the World Bank named Linfen one of the most polluted places in the world.

Unfortunately, although the pollution has improved, this is mostly what it’s still famous for today - if you’ve heard of the city at all. “In those days, the sun didn’t really rise,” Zheng told me. “The sky would get light, and you could see the light source moving across the sky, but the haze was so bad you’d never actually see the sun.”

Embarassed by the negative attention on its choking pollution, the local government took action. Steel mills received stricter emissions standards, unregulated private mines were shut down, coal washing within the city limits was forbidden, and expansion of new industrial capacity was strictly controlled. In the following years, the air gradually improved, and today I can say it’s not bad…for China anyway. Not award-winning or anything, but certainly no longer ranking among the worst in the world either. But the cost of this improvement was the substantial damage to GDP and employment. The biggest nail in the coffin came a few years later, when local champion Linfen Iron and Steel (Lingang) shuttered most of its production amidst industry consolidation in 2015-16. Linfen’s economy was shattered. Over 2015-2018, GDP basically flatlined.

From Heilongjiang to Shanxi, this is the story of many cities in China’s “heavy industry rust belt.” Following the downsizing of the traditional heavy industries, Linfen has sought a new economic identity, pursuing “modern metals”, solar energy, natural gas extraction, petrochemical refining, and tourism. According to its statistics bureau, it appears to be working. Linfen’s annual GDP growth over the last 5 years has averaged 6.3% - more than healthy - but GDP per capita was still fairly low in 2024, just 62,994 CNY per person, or about 67% of the national average.3

That’s the 2025 Linfen economy on paper - but what does does the city actually look like on the ground?

High Costs, Low Wages

“I don’t think Linfen has a good future” my airport taxi driver told me as we headed into the city. “That’s why I only drive part time here, when I have free time. Actually I work full-time in another city now. After Lingang downsized, thousands of jobs left too and there’s little heavy industry left now. There are few good jobs now - only low-wage jobs like retail or delivery.”

I asked if Linfen has any specialties it can use to promote itself. He laughed bitterly in response.“Hah! Linfen’s speciality is low wages and a high cost of living. Even though the steel jobs left, our prices are still high, based on the old days - and they haven’t come back down.”

Complaints about the high cost of living were common to all the locals with whom I chatted during my visit. Online, I found real estate in downtown Yaodu District listed for 9-10k CNY per square meter, an unremarkable number for a large Chinese city, but notably steep versus other 4th-tier cities, where something like 4-7k would be more typical. In Linfen, I could only find those “normal” real estate prices among older, secondhand properties, or by setting my location to one of the suburban districts. “Everything in Linfen is too expensive…even our taxis!” the front desk attendant at my hotel complained. “The flag fare here is 6 CNY…in Xi’an it’s 7 CNY! Why is it just 1 CNY cheaper than Xi’an? We’re much smaller than Xi’an.”

So…is Linfen’s problem more that things cost too much…or that wages are too low? Driver Huang taking me back to the airport on my last day believed the latter. “Taxi rates are just way too low in Linfen.” Huang told me. “I drive 9 hours in a day to make just 300 CNY (about 42 USD). A fare in the city can earn only the flag fare of 6 CNY. A trip out to the airport is one of the only rides worth anything.”4 A Meituan delivery driver I bumped into while ordering at Luckin agreed. “We only get 3.5 CNY per order. I need to deliver over 50 orders a day just to make 200 CNY. It’s tiring. But that’s the only kind of job for people like me in Linfen, since there’s little manufacturing here.”

However, barista Zheng put more emphasis on Linfen’s high cost of living. “I currently make 5000 CNY per month. That’s not a lot by the standard of large cities, but it’s okay for a small city. Together with my wife, our income is almost 10,000 per month. But my daughter is five years old now, and the little one is just one year old. There are so many things to spend money on now…the house loan, my daughter’s art classes, school supplies. By the time I pay for everything, the money is all gone. How could I save anything?”

Getting Out While You Can

Actually, there’s nothing wrong with the quality of life in Linfen, barista Zheng continued to tell me. It’s comfortable enough for a small city, for someone who just wants to “pass the days”, (过过小日子) especially if you don’t have many expenses. Its biggest weakness is that it lacks excitement, and opportunities to do bigger things. Zheng regrets having children so young (he’s 30 now). “If I had waited a few more years, I could have saved up some cash, gotten a little more comfortable, and I’d have less stress now. I got married and started a family at an age that’s considered normal for here…for a small city like Linfen. My parents were already starting to get anxious. But I wish I had waited longer.”

Though he hopes to find a opportunity to leave Linfen someday, he knows his dissatisfaction isn’t shared by everyone. “Our parents’ generation doesn’t think about these things so much. They only compare to the conditions when they were younger, so of course they say it’s better now. They live in comfortable houses, eat good food. It was much harder when they were young, so they don’t have many demands for the future now. They just hope I can have a stable job and a comfortable life close to home. But I think my daughter has better chances in a bigger city. I’m not the kind of person who hopes his daughter grows up to be a phoenix.5 That’s the kind of pressure that the kids in the coastal cities have, not here. But I hope my kids can go to a larger city to see more of the world.”

The front desk attendent at the Atour hotel shared similar comments about Linfen. “The city is doing okay. Definitely it’s better now than when we were young. But its prospects are still limited. I was in Beijing before. I signed up for college, but I didn’t finish. I had to come back and work. I’m the oldest one, so I take responsibility for my younger brother and sister. They’re going to college, so I need to make money here, help out my parents with their education fees. It’s 30,000 CNY per year for tuition…they could never afford that by themselves. But I wish I could have finished college. To tell you the truth, I don’t want to spend my life in Linfen. It’s okay here, but…there’s no future…” She couldn’t hide the resignation and disappointment in her voice as she imagined what her life might have been like if she’d stayed away from Linfen and completed school.

“We used to have one good university here…Shanxi Normal University. But in 2021, it was moved away to Taiyuan. Now there is a new college, the Shanxi University of Electronic Science and Technology, but it’s just a second batch (二本) school.6 I went over there and looked at it, but it seemed not impressive. I even thought it was a scam school at first. I guess it’s real though. This year, we had parents arriving in the fall to stay at our hotel and said they were here to register their kids to start college. They were from everywhere…Hebei…Tianjin…Chengdu. I guess the only people that wouldn’t go there are people from Linfen. Because if you had the chance to go to college, why would you stay here? Why wouldn’t you take the chance to get out?”

No Advantages for Industry

In the earliest years of China’s economic reform, the first to develop were the coastal provinces like Zhejiang and Jiangsu, able to leverage their ready access to ports and personal connections to the overseas Chinese disapora to attract investment spending. The inland provinces provided raw materials and labor to drive this boom. Only later in the 2000s and 2010s did the first ring of inland provinces begin to see a similar surge in manufacturing investment, as coastal producers moved inland in search of cheaper land and labor, setting up shop in nearby provinces like Henan, Anhui, and Hubei. But this wave of factory relocation has only occasionally swept so far inland as Shanxi, and when it does, it certainly lands in Taiyuan, the capital.

“Linfen’s biggest problem is our poor transportation development” barista Zheng told me. “We don’t have very big mountains, but they’re still mountains, and they block transportation. Compared to the coasts, or a province with good inland rivers, why should someone come here to open a factory? We can’t easily connect to the coasts, or the big cities.7 All we have is the capital, but Taiyuan is just a Tier 2 city, not so strong itself. Our only advantage is our coal. But now that we can’t use it for heavy industry, and we never had a chance to develop light industry either, so what are we supposed to do? There’s not much opportunity in Linfen. This is why we have so much brain drain to big cities.”8

What about the Fen River? I wondered. Can you transport goods to the coasts via the city’s namesake, this famous tributary of the Yellow River? “Hah, it’s almost dry now” Zheng chuckled. “The droughts in northern China are so severe. When I was young, it was navigable by boats, but now? Forget it! This year we had very weird weather in Shanxi and it rained for two months straight in the summer, so it was possible to take a boat on it, but it’s not normally like that. Our rivers are not like those in southern China. You can’t use them to transport goods.”

If manufacturing isn’t an option, then what about services? The local government is putting a lot of effort into promoting tourism. A lot of money has been sunk into improving local infrastructure, and clearly the hotel options are improving too (after all, there was a new Atour hotel for me to stay in). But Huang Shifu, my Didi driver back to the airport, was not so sanguine about the tourism prospects. “The city leaders plan to promote tourism…hah. I don’t believe in that. Sure, there’s the Hukou Waterfall, but that’s too far away from the city. And Xiaoxitian Temple became popular over the last year, because of the Black Myth: Wukong game, but that’s also far away. What does that kind of tourism have to do with Linfen city? That will only benefit those far-away counties, not Linfen.”9

He clearly believed that tourism in Linfen’s administrative hinterlands would have little impact on his life in Yaodu District. I understand his pessimism, though I am more inclined to take a macro view of the broader salutatory effects of tourism on the city’s economy.

“Our infrastructure is much nicer now, that’s true” he allowed, after I asked him about the city proper. “In the last two years, we’ve had a new leader in Linfen, and he wants to make the city more beautiful for tourism, so they are spending a lot of money in this way. Like those leaves” - he pointed to the bright yellow ginkgo leaves on the wet sidewalk - “those will be swept up before the end of the day. The improvement of the streets, the city wall, the renovation of the old God of Wealth Temple, this was all in the last few years. People always say Linfen’s streets are very clean now - even cleaner than Xi’an’s, and I think it’s true. But that won’t matter if there’s no tourists here.”

Was Linfen the Hypothesized Median City of my Dreams?

What did I learn in the end from my trip to Linfen? Well, I’ve been obsessed with this idea for a few years now - that understanding the mid-tier cities of China is key to getting an accurate read on modern Chinese politics, economics and society. After all, China’s Big 4 cities (Beijing, Shanghai, Guangzhou, Shenzhen) have a total population of just 84 million. After this, the total population of the next 15 cities (the “New 1st Tiers” like Hangzhou, Chengdu etc.) is roughly 200 million. Another 20% of the population lives in 2nd-tier cities, and then a full 60% of the population (840m+) lives in a 3rd, 4th, or 5th-tier city.

While I’ve concluded 4th Tier Linfen isn’t really a suitable proxy for median China, I think it’s a great representative of the “semi post-industrial north/central city trying to reinvent itself” archetype. In that regard, Linfen likely is an excellent stand-in for many small cities in Shanxi, Shaanxi, Hebei, Liaoning, etc. The recent boom in infrastructure spending, the effort to transition to new industries after severing ties of dependency with a polluting heavy industrial past, the improved air, the slower pace of life, these are all typical narratives that could belong to many parts of China.

But…it has some features making it less than ideal as national representative. The high cost of real estate is bizarre for a city of its size. Wage levels and goods prices seemed normal for a small city, but past years of uneven growth mean Linfen's GDP per capita still undershoots the national average. Most cities offer their highest living standards and wages in the urban center, but in Linfen, I found residents in the urban center to be most dissatisfied with Linfen’s prospects. Meanwhile, citizens in rural Ji County (which I’ll cover in a different post) were far more optimistic. Reviewing the regionality of the drivers of growth for the last few years helps to explain why.10

Can Yicai’s Rankings REALLY Lead Us to the Median City?

I came to Linfen because it was the median-ranked city in Yicai’s city tier rankings in 2024. But in 2023, Linfen was ranked well below the median - 23 spots lower actually, at #192. Yicai’s methodology puts a lot of emphasis on the city’s exposure to “new industry” segments like clean energy, and the city’s recent efforts to develop those sectors probably contributed heavily to Linfen’s rising rank this year. Everyone I talked to seemed to share the vibe that the city is doing okay but not great, with clear eyes on their limitations to growth versus more well-endowed regions of the country. It’s possible the transition to a stronger economic structure is already sufficently advanced to be captured by Yicai’s bean counters, but not mature enough yet to be broadly felt in all quarters on the ground. Maybe.

In 2022, Linfen’s rank was #146, so we’ve also got to consider there’s apparently quite a lot of methodological volatility going on here (although at least Linfen has always consistently been in the 4th Tier). Yicai’s ranking system has always been more focused on identifying up-and-coming top cities, and less-focused on implications for lower-tier cities.11

I thought it was notable that although Linfen has an airport, a high-speed rail station, and is served by at least three national expressways, the locals still consider the city to have poor transportation, limiting their options for growth. I’m not sure what else can be done in that regard. Frankly, I think they have pretty good hard transportation infrastructure now, but at this point are hemmed in by geographic realities. There’s little incentive for a manufacturer to site their production so far inland as long as closer options exist.

For Linfen and cities like it to survive the coming decades, a broad and proactive economic re-orientation is truly the only way. If the city can succesfully realign towards its new primary industries - ones more future-proofed against China’s energy transition - and complement it with tourism, it may be able to stem the brain drain and hold onto its people. While the taxi driver was pessimistic about the prospects for tourism, I think he underrates the economic draw of a 5A-rated scenic site (which Hukou Waterfall is). In any event, Linfen probably has a better shot at succeeding in this endeavor than other cities up in the northeastern provinces. But only time will tell.

The End…for Now

“Thanks for the talk. I’m heading to the airport now” I told barista Zheng, who had been such a champion, chatting with me all afternoon (the coffee shop was empty except for me, and he was bored).

“Oh…you’ve leaving! I hoped you were still staying another day in Linfen. I was going to invite you to dinner tonight. I would like to treat you to drink some of our local fenjiu 汾酒.”

“Unfortunately not this time…but maybe someday. If I come back to Linfen, I’ll definitely accept that offer. But you can also come visit me in Shanghai - I’d be happy to receive you.”

“Okay, let’s do that then. Let me add your WeChat. I hope to take my daughter to Disney in Shanghai someday.”

“Okay, it’s a deal. See you next time in Shanghai!”

-END