Bangkok Sold More Condos. It Took In Less per Sale.

Thailand’s 2026 recovery runs through a 0.01% transfer fee, a relaxed loan rule and 65,300 unsold units nobody has repriced

AsiaPresswire – Saranya runs a brokerage out of a shophouse in Nonthaburi, on the northern edge of Bangkok’s rail network. In the first week of October 2026 she closed four units. All four had been completed two or three years earlier. All four were sold by developers who had stopped defending the original asking price. Her commission per deal was smaller than it was in 2024, and she closed more deals anyway.

That is the Thai housing market this year compressed into one broker’s month. Units are moving. Money is not moving with them. Almost nobody in the sales material wants to talk about the difference.

The headline figure is real. Transfers in Bangkok and its five surrounding provinces rose 17.6% by unit count in the first half of 2026, according to Real Estate Information Center data cited by the Thai Real Estate Association in October 2026. Transfer value over the same six months rose 9.8%. Divide one by the other and the average transaction got smaller, by roughly 6.6%.

A market where volume climbs twice as fast as value is not a market where buyers have returned. It is a market where sellers have capitulated on the low end and buyers have traded down to meet them. The policy that made that possible has an expiry date. The inventory that made it necessary does not.

The Gap Between 17.6% and 9.8% Is the Story

Start with what the two numbers measure. Unit count records how many titles changed hands at a Land Office. Value records the appraised price written on those titles. When the first outruns the second by seven percentage points, the mix of what sold has shifted toward cheaper property. That is arithmetic, not interpretation.

REIC’s own breakdown points the same way. In the first half of 2026, transfers in Bangkok and its vicinity rose 20.8% year on year, according to REIC figures reported by Money and Banking magazine in September 2026. The nationwide figure was softer. The capital’s outer ring did the heavy lifting.

Now look at who was buying. Foreign transfers fell sharply in the same six months, per the same REIC dataset. Chinese demand, which carried the Bangkok condo market from 2017 through 2023, has not come back. Arrivals from China fell 33.6% in 2025, and condo transfers to Chinese buyers fell 13% in the same year. The first half of 2026 did not reverse either trend.

So the 17.6% is domestic. It is also policy-fed. In 2026 the government kept transfer fees and mortgage registration fees at 0.01% for homes priced at or below THB7 million, having extended the scheme to 30 June 2027. The normal rates are 2% of appraised value and 1% of the loan. On a THB5 million unit, the saving runs to roughly THB90,000 at the Land Office window, and the discount applies only to Thai citizens. Foreign buyers still pay the standard 2%.

Pair that with a relaxed loan-to-value rule and a mortgage market that grew. New housing loans issued in the first half of 2026 reached THB289.3 billion, up 12.2% year on year, with the second quarter alone up 18.4%, according to Bank of Thailand data cited by the Thai Real Estate Association in October 2026.

Three things pushed the same direction at once: a fee that costs almost nothing, a down payment that got smaller, and a developer who will pay the transfer cost on your behalf. That is not demand recovering. That is the cost of transacting collapsing, which pulls forward purchases that would otherwise have happened in 2027 or 2028.

Two Variables, and Neither One Is Demand

The market has two structural variables right now. They are independent of each other, which is why the recovery story sounds convincing and why it will not hold.

The first is the transaction-cost wedge. It is a fiscal instrument with a legislated end date. Every baht of it moves a purchase forward in time; none of it creates a household that did not exist. When the scheme lapses on 30 June 2027, the wedge closes and the pull-forward reverses. A market that borrowed demand from 2028 will spend 2028 paying it back.

The second is the shape of the unsold stock. It is physical, it was built under permits issued years ago, and it cannot be legislated away. Bangkok holds 65,300 unsold condo units worth about THB370 billion as of the third quarter of 2026, according to Cushman & Wakefield Thailand data reported by Nation Thailand in October 2026. Roughly 25,700 of them are finished and ready for transfer, about 40% of the total.

These two variables interact, but neither explains the other. A fee cut does not create 25,700 completed units. A completed unit does not extend a fee cut. You can model them separately, which is exactly what the market is not doing.

Table 1: What actually moved in the Thai housing market between 2025 and the first nine months of 2026

Indicator 2025 2026 to date Reading
Transfers, Bangkok and vicinity (units) Baseline +17.6% in H1 Volume recovering
Transfers, same area (value) Baseline +9.8% in H1 Price per deal falling
New condo launches, nationwide 16,408 units 14,377 units in 9M, +6% YoY Supply still being added
New housing loans issued Baseline THB289.3 billion in H1, +12.2% Credit easing, not income growth
Unsold condo stock, Bangkok Rising 65,300 units, THB370 billion Unchanged in substance
New condo price index, Bangkok +3.5% in Q1, then falling +2.2% YoY in Q3 Cost-push, not demand-pull

Source: REIC via Thai Real Estate Association 🔗, October 2026, and Cushman & Wakefield Thailand via Nation Thailand 🔗, October 2026

The launch column deserves its own paragraph, because it is where the recovery story breaks. Developers put 4,876 new units on the market in the third quarter of 2026, more than double the second quarter, bringing the nine-month total to 14,377, up 6% year on year. Full-year launches are tracking toward 16,000 to 17,000 units, according to Cushman & Wakefield’s Thailand research head Surachet Kongcheep, speaking to Nation Thailand in October 2026.

Read that against the 2 to 3 years it would take to clear existing stock with zero new supply. The market is adding roughly one year’s worth of absorption back into the pipeline every twelve months, while sitting on a two-to-three-year backlog. The arithmetic does not close on its own.

Where the 65,300 Units Actually Sit

Aggregate inventory numbers hide the only thing that matters, which is composition. REIC and Cushman & Wakefield both break the unsold stock by price band, and the concentration is brutal.

About 55% of Bangkok’s unsold units sit below THB3 million, or at a per-square-metre price no higher than THB100,000. Add the band between THB100,001 and THB150,000 per square metre and the share rises to about 76%, per the same October 2026 data.

Table 2: Bangkok’s unsold condo stock by price band, third quarter of 2026

Price band Share of unsold stock Mortgage needed to buy Why it clears slowly
Below THB100,000 per sq m 55% combined Yes, near fully leveraged Buyers fail loan approval, not price
THB100,001 to THB150,000 per sq m Lifts total to 76% Yes, high loan-to-value Competing with new launches nearby
Above THB150,000 per sq m 24% Often cash or offshore Thin buyer pool, slow turnover

Source: Cushman & Wakefield Thailand via Nation Thailand 🔗, October 2026

That distribution explains why the fee cut worked and why it will stop working. The 0.01% rate applies to homes at or below THB7 million, which covers most of the 76% block. The policy and the glut are aimed at the same shelf. For one year, the shelf moved.

It also explains the rent-to-own arrangements now appearing in the sub-THB100,000-per-square-metre segment. Some developers let a buyer move in as a tenant and apply for financing later, once their credit file can carry the loan. That is not a marketing innovation. It is a workaround for a mortgage approval process that has become the binding constraint on the entire low end of the market.

Surachet put the same point more plainly in October 2026: the gap between a booking and a completed transfer is where the market is leaking. A buyer can pay a reservation deposit and still fail to complete because the bank will not lend. Reservation volume is therefore a poor indicator, and transfer volume is a lagging one.

A Price Index Built Out of Cement

Here is where the official narrative gets interesting. REIC’s new condo price index for Bangkok and its vicinity rose 2.2% year on year in the third quarter of 2026. On its face, that contradicts everything above: how can prices rise in a market with 65,300 unsold units?

Look at the quarterly path first. The index ran +3.5% in the first quarter of 2025, +1.4% in the second, -0.6% in the third, -0.4% in the fourth, then +0.2%, +1.6% and +2.2% across the three quarters of 2026. That is not a price cycle. That is a series that went negative for two quarters and has been climbing back toward zero ever since. A 2.2% annual gain in a country with 2% to 3% construction cost inflation is a flat real price with a rounding error on top.

Now look at what is pushing it. REIC attributes the increase to construction costs and to land prices in prime locations, and its own materials data supports that: concrete products rose 13.2% quarter on quarter in the third quarter of 2026. Developers are not raising prices because buyers are bidding. They are raising asking prices because cement got more expensive and because the only land left near a station is expensive.

The gap between asking and clearing shows up in the same dataset. Average launch price in the second quarter of 2026 was THB150,420 per square metre, while average sale price in the first half of 2026 was THB120,360. Asking prices sit about 25% above what buyers actually signed for. In a market with two years of stock, that gap does not close upward.

There is a second, quieter signal in the same REIC release. The share of projects offering gifts rather than cash discounts rose to 38.1% in the third quarter of 2026, from 30.1% in the second. Cash discounts fell to 30.5% from 35.3%. The share of developers paying the buyer’s transfer fee fell to 31.4% from 34.6%.

Read that sequence carefully. Gifts up, cash down. A gift preserves the headline price; a cash discount breaks it. Developers have moved from cutting price to protecting the number, which is what happens when the number is being watched by banks, by bondholders and by the land office that assesses the next project.

The comparable sale in that comparison is usually two stations away and already finished. A buyer choosing between a new launch at THB150,420 per square metre and a completed unit at THB120,360 does not need a discount rate to reach a conclusion. The gift is what the sales office offers instead of answering that comparison.

The Part of Thai Property That Is Not in Trouble

It would be easy to finish with a general verdict that Thai real estate is oversupplied. That verdict would be wrong for about half the asset class, and the exception is informative.

Industrial property is running on a different cycle. Ready-built factory occupancy stood at 89.55% in 2026, and ready-built warehouse occupancy at 85.28%, according to HataMatata’s 2026 industrial survey. Industrial land in the eastern corridors was quoted at around THB8.4 million per rai in the same survey. Those are not numbers from a market with a two-year overhang.

The reason is mechanical. Factory and warehouse demand tracks goods movement and supply-chain relocation, which is a flow that renews every quarter. Condominium demand tracks household income and credit access, which is a stock that has to be rebuilt slowly after a debt cycle. One is a lease signed by a company with a balance sheet. The other is a thirty-year loan signed by a household with a debt-service ratio.

This matters for the recovery argument because it removes the excuse. If the problem were national demand weakness, industrial occupancy would be falling too. It is not. The weakness is concentrated in one product type sold to one kind of buyer through one kind of credit, and the policy response has been aimed at the transaction cost rather than at the credit constraint.

There is a second contrast worth holding next to it. Bangkok’s CBD grade A office vacancy was 21.9% in 2026 and office-linked condominium vacancy was 3.1%, per separate broker datasets. Two property types, one geography, opposite results. Merging them into a single sentence about Bangkok property destroys the only useful information in either number.

Who Feels It First

The first group to feel this is not buyers. It is developers holding the 25,700 completed units, because a completed unit starts costing money the day it is finished. Common-area fees, security, maintenance, and the interest on construction finance all run whether or not a title transfers. Revenue recognition waits. Costs do not.

The second group is the sub-THB3 million buyer, who is not really a buyer until a bank says so. The fee cut helps at the Land Office, but the binding constraint sits earlier in the process. Thailand’s household debt is among the highest in the region, and lenders have tightened documentation checks on foreign-linked structures and source-of-funds reviews through 2026. A 0.01% fee does nothing for a household that cannot pass the debt-service ratio.

The third group is the developer balance sheet, and this is where the price index becomes misleading rather than merely unhelpful. If input costs rise 13.2% in a quarter while the achievable price rises 2.2% a year, the margin absorbs the difference. That shows up as slower land purchases, delayed launches, and the shift toward smaller projects that Cushman & Wakefield documented in its third-quarter 2026 survey. It also shows up in the permit data: building permits fell 28.25% year on year, according to Bank of Thailand figures reported by Good Yield in 2026.

A 28% drop in permits alongside a 6% rise in launches sounds contradictory until you remember the lag. Permits issued in 2026 become launches in 2027 and 2028. The current launch number is the ghost of a decision made two years ago, when the market still looked like it could absorb 16,000 units a year without discounting.

The fourth group, and the one nobody writes about, is the agent. Saranya’s four closings in a week came with smaller commissions each, because the units were cheaper and the developer’s margin was thinner. Volume up, income flat. That is the same trade the whole market is making.

What Would Prove This Wrong

A thesis that cannot be falsified is a mood. Three specific observations would break this one, and each has a date attached.

The first is average value per transfer. If the mix shift were temporary, the ratio would recover as the cheap stock clears and buyers move up the price ladder. Watch the second half of 2026: if transfer value grows faster than transfer volume for two consecutive quarters while volume holds near the first-half pace, the down-trading argument fails and genuine demand has returned.

The second is the sub-THB100,000-per-square-metre launch pipeline. If developers restart launches in that band at 2023 volumes and clear them within twelve months, the inventory overhang was a pricing problem rather than a structural one. The current evidence points the other way: 90% of second-quarter 2026 launches were within walking distance of an operating BTS station, which is where land is scarcest and prices are highest. Developers are not testing the cheap end.

The third is the expiry test. The fee cut ends on 30 June 2027. If transfers in the second half of 2027 hold at or above the 2026 pace, the policy was never the driver and this analysis is wrong about the mechanism. If they fall by anything like the amount they rose, the recovery was a calendar effect wearing a policy costume.

There is a fourth test, and it is the least comfortable. The 2 to 3 year clearing estimate assumes no new launches. If launches continue at 16,000 to 17,000 units a year against absorption in the low teens, the clearing period does not stay at 2 to 3 years. It extends, and the estimate becomes a moving target that never arrives.

One number is worth watching more than the rest, because it is the only one that cannot be discounted, gifted or subsidised away. It is the count of completed units still held by developers at the end of each quarter. If that number falls through 2027, the market is clearing. If it holds above 25,000 while launches continue, the industry has decided to store its own product and call the storage a pipeline.

Data basis

Data as of October 2026. The transfer figures come from REIC, the Real Estate Information Center under the Government Housing Bank, as reported by the Thai Real Estate Association in October 2026 and by Money and Banking magazine in September 2026. Unit counts and values are Land Office registrations, which means they record completed transfers rather than reservations, signed contracts or sales offices claims. The 17.6% and 9.8% are year-on-year comparisons for January to June 2026 against the same months of 2025.

Inventory figures come from Cushman & Wakefield Thailand, collected at the end of the third quarter of 2026 and reported by Nation Thailand on 14 October 2026. The denominator for the price-band shares is the full unsold stock of 65,300 units, not the completed subset. The 40% completed share is calculated against that same base.

Price index readings are REIC’s new condo index for Bangkok and vicinity, released 12 October 2026, with the quarterly series reconstructed from the published year-on-year changes. The gap between launch price and sale price uses HataMatata’s second-quarter 2026 launch survey and first-half 2026 transaction survey; the two are different samples, so the 25% figure is an indication of direction, not a measured spread.

Two limits on this reading. First, the transfer value series reflects appraised values used for fee assessment, which can lag market prices in a falling market. Second, the clearing estimate of 2 to 3 years is Cushman & Wakefield’s, computed at the current absorption pace with zero new supply, which is a hypothetical rather than a forecast.

Sources

This article was prepared by the expert team at Global Infinity Labs, drawing on insights and data provided through its MCP data services.

About Global Infinity Labs

Global Infinity Labs is an AI innovation company specializing in MCP data services.

Email: support@globalinfinitylabs.com