From Beijing’s Stabilization Rhetoric to Singapore’s Cooling: What ‘Steady’ Actually Means in 2026
By: Logan Pierce – AsiaPresswire – On Tuesday, a government in Beijing announced that it was striving to stabilize the real estate market. On Wednesday, a wealth manager in Singapore sent me a chart showing that the city-state’s public housing resale index had fallen for two consecutive quarters for the first time in nearly seven years. On Thursday, a developer in Taipei told a roomful of reporters that the market was “diverging” — which is real-estate vocabulary for “we have no idea what is happening, but we are paid to sound like we do.”
These three statements were all true simultaneously, and they describe three different markets that share a coastline and almost nothing else. The word doing the most work in all of them is “stabilize,” and it is a word that has been stretched so thin you could thread it through a needle. China’s 2026 government work report obliges with “strive to stabilize the real estate market” which is the kind of sentence that sounds reassuring in a communiqué and almost comical on a spreadsheet. Because “stabilize” is carrying at least four different meanings in Greater China alone, and the report is not required to tell you which one applies to your particular apartment block.
That is the puzzle: four real estate markets, one rhetorical commitment, and roughly zero overlap between them.
Stabilization, According to the Communiqué
Let us start with the number that Beijing’s stabilization claim rests on, because it is the least exciting number in the entire piece and therefore the most honest one. The average new home in one hundred Chinese cities sold for ¥17,114 per square metre in the most recent period, which is 2.52% higher than the same period a year earlier. That is it. That is the stabilization. Prices are up, gently, almost imperceptibly, like a thermostat that has been nudged half a degree and nobody has noticed.
Here is what the communiqué does not say, because it is not required to. The 2.52% is an average of one hundred cities, which means it is a number that has been averaged so many times it has lost most of its opinion. Some cities are up double digits. Some are down. The average is a polite fiction that keeps the headline writers employed and the property stocks from falling another 30%. (Of course, the average is also the only number anyone in a hurry can quote, which is why it gets quoted.)
The real stabilization, if it is happening at all, is not in prices. It is in the absence of collapse. After the Evergrande-era implosion of 2021–2023, when the Chinese property sector was removing roughly two trillion yuan of value from the balance sheets of households and local governments in about eighteen months, “not collapsing” counts as a policy success. It is the real-estate equivalent of a patient leaving the intensive care unit able to walk to the bathroom unassisted. The bar is on the floor, and Beijing has cleared it. That is not a criticism — it is an observation about how low the bar was set.
But the gap between “stabilized” and “worth buying” is where the interesting stuff lives, and nobody in the communiqué addresses it. A buyer in a tier-two city looking at ¥17,114 per square metre is looking at a number that has been smoothed across a hundred markets, and the smoothing is doing about as much useful work as the word “stabilize.”
Hong Kong and Taiwan: Two Markets, One Divergence
If you want to see what “stabilize” looks like when it is translated into two very different local dialects, look east. Hong Kong and Taiwan are both markets where prices are rising, where the official line is one of cautious optimism, and where the underlying dynamics are almost entirely opposite.
Hong Kong is the easier one to describe, because it is a single market with a single set of rules and a single problem: too much building. Knight Frank predicts residential prices will rise 8 to 10% in 2026, which is a confident forecast from a firm that has been predicting things for a very long time and has the luxury of being right about half the time. Office supply is peaking, which is developer-speak for “we built too much and now we are finding out.” Land sale revenue is expected to increase by 1.3 times, which sounds like good news until you remember that land sales are a one-time event — you cannot sell the same plot twice — and a 1.3x increase in revenue often means the government is selling more land at lower prices per square foot. The market reads this as recovery. The arithmetic reads it as volume compensating for price.
Taiwan is the harder one, because it is not one market at all. The average new home in Taipei sold for NT$113.8万 per ping, up 5.4% year on year, which is a number that sounds healthy until you realize that “average” in Taiwan is a battlefield. Seven cities are diverging, which is the politest way anyone has yet found to say “some of this is nonsense.” The phrase doing the rounds locally is “the north is strong, the protein areas are loosening.” Protein areas, for the uninitiated, are the commuter suburbs where people live because they cannot afford the city centre, and the image of a protein area “loosening” is the kind of metaphor that only a local analyst could have invented. It means prices are softening in the places where ordinary people actually live, while the northern core keeps climbing on the back of tech money and financial services. That is not a market. That is two markets arguing with each other, and the average is pretending they are the same thing.
The common thread is that both Hong Kong and Taiwan are experiencing the same basic dynamic under different branding: prices up because of scarcity and concentration, not because anyone is getting richer. In Hong Kong the scarcity is physical — there is only so much land, and most of it is already spoken for. In Taiwan the scarcity is regulatory — there is plenty of land, but the rules about what you can build and where are written by people who do not have to buy the results. Both produce the same outcome: a market that goes up in the good parts and goes nowhere in the parts where normal people live, with an average that makes the whole thing feel like a conspiracy against common sense.
Singapore: The Coldest Market in Greater China
Singapore is the outlier in this group, and not just because it is a city-state rather than a region. It is the only market in Greater China where the word “cooling” is not a euphemism and where the government is actively proud of it. That is worth pausing on, because it is almost unheard of: a government that welcomes a housing slowdown.
The numbers are unambiguous. Private residential prices rose 0.5% in the second quarter of 2026, which is the weakest quarterly increase in seven quarters. The first half of the year is up 1.4%, which is the smallest half-year gain since 2020. Those are not dramatic numbers, but they are the right kind of dramatic — the kind that let a central bank say “we are managing this” and be believed.
The public housing market is doing something it has not done in nearly seven years. The HDB resale index fell to 202.7 points in Q2, down 0.3% from the previous quarter, marking the second consecutive quarterly decline. The last time this happened was the third quarter of 2019. Resale transactions totaled 6,268 in the quarter, down 10.2% from the previous quarter. And the reason is not mysterious: 13,484 public housing units reach their Minimum Occupancy Period this year and are flooding onto the resale market, the government has released 9,320 units under the Confined Land Sales programme — more than 50% above the ten-year average — and another 7,960 units are scheduled for launch in October. There is supply, and then there is supply, and Singapore has chosen the second option.
Here is the Singapore Q2 2026 housing data, because the table is the point:
| Indicator | Value |
| Private residential price, QoQ | +0.5% |
| Private residential price, H1 YTD | +1.4% |
| HDB resale index, Q2 | 202.7 points |
| HDB resale index, QoQ | -0.3% |
| HDB resale transactions, Q2 | 6,268 |
| HDB units reaching MOP, 2026 | 13,484 |
| HDB BTO launch, October 2026 | 7,960 units |
| GLS supply, 2026 | 9,320 units |
| Private homes completion pipeline | about 61,000 units |
| Developer sales, Q2 | 2,141 units |
| Top 3 developers, share of sales | 67.5% |
| Resale transactions, Q2 | 2,634 |
| Total transactions, Q2 | 5,420 |
| New sales takeaway rate | 77.5% |
| Units priced below S$2M, Q2 | 1,016 |
| Units priced below S$2,000 psf, Q2 | 144 |
| Landed transactions, H1 2026 | 869 |
| Landed transaction value, H1 | S$5.4 billion |
| CCR landed price, YoY | +20.9% |
| RCR landed price, YoY | +6.4% |
| OCR landed price, YoY | +9.1% |
| GCB transactions, H1 | 10 |
| GCB transaction value, H1 | S$304 million |
| GCB record land rate | S$4,550 per square foot |
| Prestige landed value, YoY | +19.3% |
| Gross rental yield | 3.1% |
Read that table in order and a single story emerges. Prices are barely moving. Transactions are declining. Supply is surging. And yet the new-home market is still absorbing 77.5% of what developers offer, which means demand has not disappeared — it has simply been rerouted. Buyers are flocking to newly launched projects because the pricing is competitive, while the secondary market, where 2,634 units changed hands, is at its slowest since the second quarter of 2020. That is not a demand collapse. That is a two-speed market, and the speed limit is set by the developers, not by the buyers.
The landed housing segment tells a different story in the same table, and it is the story that gets the least attention because it is the most absurd. Landed transactions fell 13.7% in the first half to 869 deals worth S$5.4 billion, but prices did not fall — they rose everywhere. The Core Central Region saw landed prices jump 20.9% year on year. The Rest of Central Region was up 6.4%. The Outside Central Region, up 9.1%. And in the prestige segment, the value of landed homes rose 19.3%. Ten Good Class Bungalows changed hands in the first half for a total of S$304 million, with a record land rate of S$4,550 per square foot achieved on Nassim Road. That is a number that would be staggering anywhere, and in Singapore it is simply Tuesday.
So: the mass market is cooling, the luxury market is on fire, and the government is releasing more supply to cool the mass market while watching the luxury market set new records. That is not a housing policy. That is a controlled burn, and the people holding the matches are not pretending to be embarrassed about it.
The interest rate backdrop makes the whole thing possible. The Singapore overnight rate average is around 1.0 to 1.1%, down from over 3% in 2023. That is the buffer that lets the Housing and Development Board manage an orderly slowdown instead of a panic. When money costs almost nothing, a market can absorb a lot of new supply without breaking. When money costs 3% or more, the same supply becomes a problem. Singapore’s cooling is real, but it is also heavily subsidized by the global interest rate environment, and nobody should mistake that for a structural achievement.
The Data Center Subplot: Where the Money Actually Goes
There is a subplot running through this whole story that most housing reporters miss, and it is the reason the word “real estate” in 2026 means something different than it did in 2015. While the residential markets of Greater China are arguing about whether “stabilize” means flat or gently up or gently down, a completely different kind of real estate is being built at a pace that makes housing look like a hobby. Data centers.
Singapore approved 200 megawatts of new data center capacity this year under the Data Centre Financial Assistance Framework. That is the provisional number, and it is already described as far below the amount applicants wanted. The existing data centers are running at 95% occupancy with a colocation vacancy rate of just 4.9%. That is not a market with excess supply. That is a market that has run out of electricity.
Meanwhile, across the causeway in Johor, the pipeline stands at 8,542 megawatts — the largest in Asia-Pacific — with a colocation vacancy rate of 0.7%. Johor is the place where Singapore’s unmet demand goes to get built, and the numbers are so lopsided that they almost stop being interesting. Singapore has 200 megawatts approved and 1,118 megawatts in operation. Johor has 8,542 megawatts in the pipeline and 1,110 megawatts in operation. The two places are roughly the same size, and one of them is building data centers as if the other one had already been eaten.
The mechanism is straightforward and mildly embarrassing for anyone who still thinks of housing as the primary expression of real estate capital. A 50-megawatt data center in Johor costs about RM1.2 million a year in electricity. The construction cost is about US$9.6 million per megawatt, which is lower than in Singapore, Japan, or South Korea. The Johor-Singapore Special Economic Zone has a green lane that cuts the grid connection time from 36–48 months down to about 12 months. The corporate tax rate is 5% for up to 15 years. And the first full year of the zone approved RM76.98 billion in investment.
Compare that with the residential numbers above. Singapore’s entire housing market — every private condo, every HDB flat, every landed property — moved 5,420 transactions in a quarter. The data center story is about a different asset class entirely, priced by different variables entirely, and growing at a speed that housing cannot match because housing is constrained by land, immigration, and politics, while data centers are constrained by electricity, fiber optics, and the willingness of cloud providers to sign 10-year leases. They share the word “property” and almost nothing else.
The Malaysian utility TNB has secured 59 data center projects with a maximum demand of 8.3 gigawatts. Of that, 36 projects totaling 4.5 gigawatts are already connected to the grid. But the actual load is 1,054 megawatts, which means the utilization rate is about 23%. That is the most honest number in the entire data center story: there is an enormous gap between what has been promised, what has been built, and what is actually being used. The pipeline is 8.3 gigawatts. The reality is 1.054 gigawatts. The ratio is roughly 8 to 1, and it is the same ratio that appears in every other part of this article — between stabilization and prices, between supply and demand, between what governments say and what markets do.
The Global Comparison: What “Steady” Looks Like Everywhere Else
If you zoom out far enough, Greater China’s housing looks less like a regional story and more like a local accent on a global conversation that everyone is having separately. Savills reports that global real estate investment crossed US$1 trillion for the first time in 2026, which is a number so large it stops being a number and becomes a weather system. Europe contributed €52 billion in the first quarter, up 6% year on year, and the Europe-Middle East-Africa region accounted for 50% of global fundraising and 22% of growth. That is a region that has discovered how to make money from buildings and is not looking back.
Japan is the most interesting case because it is doing the opposite of almost everyone else. Land prices in Tokyo’s 23 wards rose 9.0%, the largest increase since 1992, and apartment prices have risen for 70 consecutive months. Seventy months. That is nearly six years of uninterrupted price increases in a country that has spent three decades being told it is finished. The explanation is partly foreign money, partly domestic scarcity, and partly the fact that Japan’s interest rates have been negative for so long that owning a property in Tokyo started to feel like saving money rather than spending it. No one quite knows what happens when that changes, and nobody is in a hurry to find out.
Germany is doing the grown-up thing: national housing prices are up 3 to 4%, and the country has an annual shortage of 320,000 units. That is not a market. That is a country admitting it has a problem and doing something about it slowly and correctly, which is the German way.
London is a buyer’s market, and it is the kind of buyer’s market that makes British estate agents write long, thoughtful paragraphs about “challenging conditions.” The average asking price is £676,248, down 1.2% year on year, and supply is at a 12-year high. That is a market where sellers are discovering that the person standing in front of them has options, and the options are not all in London.
The United States is in what is being called a “Great Reset” year, which is the kind of phrase that sounds like a cryptocurrency and is actually about the fact that median home prices are up 1% while the average 30-year mortgage sits at 6.3%. The gap between those two numbers is the entire American housing market in 2026: prices barely moving, financing expensive, and everyone pretending this is normal.
Canada is splitting down the middle, which is the Canadian way. Nationally, prices are up 1.0%. Toronto is down 4.5%. Quebec City is up 12%. That is not a country. That is three countries sharing a currency and a politeness requirement.
Australia is down 1.1% nationally in 2026, with a forecast rebound of 3.4% in 2027, which is the kind of forecast that is exactly as reliable as every other forecast ever made about anything. Dubai recorded 204,950 transactions, a record, with ready villas averaging AED1,551 per square foot. Saudi Arabia opened residential ownership to foreigners in January 2016 and has seen Riyadh prices rise 8 to 15% with rents up 19.6%, which is what happens when you let people buy something they were previously not allowed to own and there is a limited supply of it.
Mexico delivered a 28.9% median return in the twelve months after the cycle trough, which makes it the best-performing housing market in the world according to Bank of America Securities, and the fact that this is being reported without anyone asking why tells you how thoroughly we have accepted that emerging-market real estate is a casino where the house sometimes loses. Brazil’s São Paulo averages R$12,055 per square metre. South Africa offers gross yields of 7 to 8%, rising to 9% in the Western Cape, with Johannesburg averaging R$1.4 million per property. Nigeria’s Lagos has luxury properties yielding 3%, which is a rental yield so low it is effectively a storage fee, while value areas deliver 8 to 9%. Kenya’s Nairobi is split — detached houses up 4.2%, apartments correcting through oversupply. India’s residential market is up 6 to 7%, with luxury the growth engine and Mumbai up 7.5%. The Philippines is a buyer’s market with over 30,000 unsold units in Metro Manila and Bonifacio Global City averaging ₱224,000 per square metre.
Here is that global picture, compressed into the smallest possible table:
| Market | 2026 figure |
| China, 100-city new home average | +2.52% YoY |
| Hong Kong, Knight Frank forecast | +8 to 10% |
| Taipei, new home average | +5.4% YoY |
| Singapore, private residential, H1 | +1.4% YTD |
| Tokyo 23 wards, land price | +9.0% YoY |
| Germany, national housing | +3 to 4% |
| London, average asking price | -1.2% YoY |
| United States, median home price | +1% |
| Canada, national | +1.0% |
| Toronto | -4.5% |
| Quebec City | +12% |
| Australia, national 2026 | -1.1% |
| Australia, forecast 2027 | +3.4% |
| Dubai, ready villa | AED1,551 per sqft |
| Riyadh, residential | +8 to 15% |
| Mexico, median return post-trough | +28.9% |
| São Paulo | R$12,055 per sqm |
| Johannesburg | R$1.4M average |
| Lagos, luxury yield | 3% |
| Lagos, value area yield | 8 to 9% |
| Nairobi, detached | +4.2% |
| India, residential | +6 to 7% |
| Mumbai | +7.5% |
| Metro Manila, BGC | 224,000 PHP per sqm |
Read that table from top to bottom and the first thing you notice is that there is no such thing as a global housing market. There are twenty-two entries in that table and at least fifteen different stories. The second thing you notice is that almost every number is being explained by something local — interest rates, immigration, regulation, land supply, political cycles — and none of those explanations travel. The third thing you notice is that the word “stabilize” appears in exactly one of those rows, and it is the row belonging to the government that has the most to lose if the story goes badly.
What “Stabilize” Actually Means
So what does Beijing’s actually mean in practice, if you set the communiqué aside and look at the numbers? It means at least four things, and they are not all compatible:
- It means the government has decided that further price declines are more dangerous than further price increases, and it is willing to use every tool at its disposal — mortgage rate floors, purchase restriction tweaks, developer financing windows, local government land quotas — to enforce that preference. That is a rational choice, and it is also a choice that protects the people who bought at the top of the cycle at the expense of the people who were about to buy at a better price.
- It means the government is no longer trying to reform the property sector’s business model, which was always the harder and more necessary project. The old model — developers borrow short to lend long, land sales fund local governments, households store wealth in apartments — was unsustainable, and everyone knew it. The new model has not been announced, because announcing it would require admitting that the old one is being retired. Stabilization, in this reading, is the pause between models, and it is the most honest thing the government has done in years.
- It means the gap between the average and the reality is being managed rather than closed, because closing it would require either prices to fall or supply to shrink, and neither is politically acceptable. The 2.52% average is a target, not a description, and the fact that it is being met is the entire policy achievement.
- And it means, finally, that the most important housing policy in Greater China is not being made in Beijing at all. It is being made in Singapore, by a government that has the honesty to call a cooling a cooling and the tools to make it orderly. Singapore is the only market in this article where “stabilize” means “release more supply and let prices find their level,” and it is the only market where that approach is actually working, because it is the only market that still has the capacity to build.
The forty-year mortgage that Indonesia introduced in August — 1% down payment, 2.75% teaser rate, monthly payments as low as Rp70,000 — is the logical endpoint of the stabilization strategy, and it is worth sitting with for a moment because it is the most revealing thing any government has done this year. It is not a housing policy. It is an intergenerational wealth transfer dressed up as affordability, and it is the perfect expression of what “stabilize” means when you have run out of other options: you do not make the house cheaper, you make the buyer older, and you call the difference a subsidy. That is not a criticism of Indonesia, and it is not a model for anyone else. It is simply the clearest available definition of what the region has chosen to mean by housing stability in 2026.
Greater China’s four markets will continue to diverge. Beijing will keep issuing communiqués. Singapore will keep releasing land. Hong Kong will keep building offices nobody wants. Taipei will keep pretending its averages describe a single market. And the data centers in Johor will keep consuming electricity at a rate that makes the residential numbers look like a rounding error. That is not a forecast. It is the current arrangement, and it is working for the people who built it, which is the only endorsement that matters in real estate.