Latest / China Economy with Fexingo: Beijing's Policies, Trade, and Asia-Pacific Markets / How Chinas Cash-Strapped Cities Are Selling Land to Tech Companies
Transcript
- Lucas: Hangzhou just sold a two-hundred-acre plot to Alibaba's cloud computing division for roughly thirty percent of its assessed value. That's not a distressed fire sale in the traditional sense — it's a deliberate strategy, and it's happening in cities across China right now. Luna: Thirty percent of assessed value — that's a massive discount. Why would a city government willingly take that hit? Lucas: Because the old model of selling land to property developers for residential towers is dead. Local governments in China relied on land-transfer fees for about forty percent of their fiscal revenue for years. With the property market in a prolonged slump, that tap has run dry. So they're pivoting — selling land not to homebuilders, but to big tech companies who bring jobs, corporate tax, and a friendlier headline. Lucas: It's a trade-off. Less upfront cash, but a longer-term revenue stream. And right now, with local government debt being a major concern for Beijing, this swap from property-driven to tech-driven land sales is being quietly encouraged. Luna: So it's like a fiscal lifeline. And if walking through the economy with us has made something click, a couple of dollars a month is genuinely what keeps these shows ad-free and independent — buy me a coffee dot com slash fexingo, if you've gotten something out of them. Lucas: Yeah, that really does help. And it's listeners like that who let us keep digging into these specific, unsexy but important mechanics. So back to Hangzhou — Alibaba's cloud unit gets a campus at a bargain, the city gets a commitment to hire thousands of engineers locally, and the central government gets to show a 'high-tech' land use rather than another unsold residential plot. Luna: Is there a precedent for this? I mean, have other cities done similar deals? Lucas: Absolutely. Shenzhen sold land to Tencent for its 'Internet Plus' headquarters at a discount a couple years ago. Beijing has allocated plots to ByteDance and Baidu in the Zhongguancun area with favorable terms. What's new is the scale and the urgency. In the first half of this year, land-transfer revenue across China's top fifty cities dropped over thirty percent year-over-year. Mayors are getting creative. Luna: But isn't there a risk? If the tech companies themselves hit headwinds — and we've seen regulatory crackdowns and slowing growth — then the cities are trading one volatile revenue source for another. Lucas: That's exactly the counter-argument. And it's one that some economists at the People's Bank of China have raised privately. A tech campus doesn't generate the same immediate multiplier effect as a housing development — no new furniture purchases, no appliance sales, no property management fees. Plus, corporate tax revenue from these companies depends on profitability, and we've seen margins compress across Chinese tech. Luna: Yet the market seems to be rewarding this pivot. FXI — the big China ETF — is up four and a half percent in the last five days. Alibaba alone jumped sixteen percent in that window. Are investors betting that this land for jobs swap will revive local economies? Lucas: Partly. But I think the rally is more about a broader re-rating of Chinese tech on valuation and a hope that the worst of the regulatory storm is behind us. Still, the land-sale dynamic feeds into a more optimistic narrative: that Beijing is allowing market-based solutions to local fiscal stress rather than a blanket bailout. That's a big deal. Luna: Let's talk about the numbers. You said thirty percent of assessed value. How is that assessed value determined? Is it based on the old residential peak, or a more recent commercial benchmark? Lucas: Good question. The assessed value in these transactions is typically based on the land's 'reference price' published by the municipal land bureau, which itself is a lagging indicator — often reflecting valuations from two or three years ago when the market was stronger. So the discount from that reference can be misleadingly large. A more realistic comparison would be to recent auction prices for similar commercial plots. Even then, the discount is still significant — maybe fifteen to twenty percent below market. Lucas: But the key is that these are negotiated sales, not open auctions. The city picks the buyer and the price. That gives them control over who comes in and what they build. In Hangzhou's case, Alibaba is required to build a research center focused on cloud infrastructure and AI — exactly the kind of industry the central government wants to promote. Luna: So it's industrial policy via land allocation. That has a long history in China — special economic zones, development zones, all built on cheap land. But those were greenfield sites. Now cities are selling prime, often centrally located parcels. Are they cannibalizing their own future tax base by selling too cheap? Lucas: Potentially. But land in China is ultimately owned by the state. The city is leasing usage rights, typically for forty to seventy years. So they're not selling the asset permanently. And if the tech company succeeds, the city can capture value through corporate income tax, personal income tax from high-paid employees, and ancillary business taxes. It's a bet on the company's future rather than a one-time cash infusion. Luna: And if the company fails? Or leaves? Then the city is left with a specialized campus that's hard to repurpose. That's a risk. Lucas: It's a real risk. There are already examples — some of the early 'tech parks' from the 2000s in cities like Chengdu and Xi'an have high vacancy rates because the anchor companies downsized. But those were mostly hardware manufacturing. Today's investments are in software, AI, and cloud — which are more scalable and less capital-intensive. Still, no guarantee. Luna: What does this mean for the broader China economy? We're seeing the yuan stable at 6.79 per dollar, exports still strong — up about four percent year-over-year — but imports growing faster. The trade surplus is narrowing. Are these land deals a sign that domestic demand is so weak that local governments have to resort to bartering land for jobs? Lucas: That's a sharp way to put it. And there's some truth there. Domestic consumption, especially in tier-two and tier-three cities, hasn't rebounded as fast as policymakers hoped. Retail sales growth is still in the low single digits. So local governments need to stimulate investment somehow. Offering cheap land to deep-pocketed tech firms is one of the few levers they still have, given that they can't run large fiscal deficits without central approval. Luna: And the central government is okay with this? I mean, it's a form of subsidy, potentially distorting land markets. Lucas: The central government hasn't openly endorsed it, but it hasn't stopped it either. There are signs of tacit approval. The Ministry of Natural Resources issued guidelines last year encouraging 'efficient use of land for new industries' — that's code for tech. And the Ministry of Finance seems willing to tolerate lower land-transfer revenue as long as local governments don't default on their bonds. The alternative — a wave of local government financing platform defaults — would be far more destabilizing. Luna: So we're in a period of managed adjustment. Land sales shift from property to tech, tech gets cheap expansion, cities get jobs, and the central bank keeps the yuan stable. It's a balancing act. Lucas: Exactly. And the market is watching closely. If this model succeeds, it could become a template for other struggling cities. If it fails — if tech companies don't deliver the jobs or if the land sits idle — then local government debt problems could resurface in a bigger way. Luna: Are there any early signs of success? Has any city seen a meaningful fiscal turnaround from these deals? Lucas: Shenzhen is probably the best case study. Its land sales to Tencent and Huawei over the past decade have helped transform it into a tech hub. Corporate tax revenue from those companies now makes up a significant share of the city's budget. But Shenzhen had advantages — proximity to Hong Kong, a highly educated workforce, strong central support. The question is whether inland cities like Zhengzhou or Changsha can replicate that. Luna: Which brings us back to the present. We're recording this on July 9, 2026. The KWEB China internet ETF is up nearly six percent in the last five days. That's a strong vote of confidence. But it's based on hope as much as reality. The land deals are long-term moves, not quick fixes. Lucas: Right. And that's the tension. Investors are pricing in a recovery narrative, but the underlying fiscal mechanics are still fragile. If you look at the current account balance — it's negative two hundred twenty-six billion dollars as of January. That's a deficit. China is importing more than it exports in services, and the overall trade surplus is shrinking. That puts pressure on the yuan and on local government finances. Luna: So the land for tech swaps are a micro-level response to a macro-level imbalance. They may help individual cities, but they won't solve the national current account deficit or the property overhang. Lucas: No, they won't. But they might buy time. And in the Chinese policy playbook, buying time is often the first step toward a more structural reform. The question is whether that time is used wisely — to broaden the tax base, improve social services, and boost household consumption. If it's just used to kick the can, we'll be having this same conversation in five years, just with bigger numbers. Luna: It's a fascinating story precisely because it's so specific. Not a grand theory, but a real transaction in Hangzhou that reveals how the system is adapting under pressure. Lucas: And that's what we try to do here — find the concrete example that illuminates the bigger picture. For now, the Hangzhou-Alibaba land deal is a window into China's new fiscal reality. Whether it becomes a blueprint or a cautionary tale, we'll be watching.