Will the number of Dubai residential sales worth at least AED 20 million in December 2026 be at least 25% lower than in October 2026?

TL;DR

The forecast is 38.7% Yes and 61.3% No, conditional on non-annulment. Treat No as the leading outcome: the evidence establishes luxury-market weakness, not another decline of at least 25% from an already subdued October. The decisive facts still missing are October’s exact eligible count and the concentration of December registrations.

Context

This question measures a narrow change in activity, not whether Dubai’s property boom has ended. Yes requires December’s count of eligible residential sale records worth at least AED 20,000,000 to be at most 75% of October’s count. The population includes qualifying single apartments and villas in the Sales group, but excludes land, whole buildings, mortgages, gifts and multi-property records. The resolving source is DLD’s Real Estate Data, using its publication on 20 January 2027. An October count of zero annuls the question.

As of the November 4, 2026 forecast date, the evidence shows a weakened market with pockets of resilience. Cavendish Maxwell reported approximately 33,900 residential sales in Q3, down 38.6% year on year but only 2% quarter on quarter. That difference matters: a large annual contraction can coexist with a much slower current rate of decline. (Cavendish Maxwell)

Evidence

The historical starting point is that December does not reliably fall far enough below October to satisfy this question. An adjacent luxury-sales series from Projectory provides six October–December comparisons. Its premium category corresponds to AED 10 million-plus sales, not the question’s AED 20 million threshold. The complete supplied seasonal history is below. (Projectory aggregate data)

Year	October premium sales	December premium sales	December/October ratio
2020	53	33	0.623
2021	97	127	1.309
2022	233	180	0.773
2023	398	329	0.827
2024	606	313	0.517
2025	534	460	0.861

Only two of these six comparisons crossed the required decline threshold. This is a small reference class, with a different price cutoff and changing market conditions. It does not determine the forecast. It does establish that a sharp December decline is neither automatic nor absent from the historical record. Across broader overlapping two-month windows in the same proxy, declines of at least 25% occurred in about 19% of comparisons since 2021 and 35% since 2025. Those windows overlap, so they are not independent observations. (Projectory aggregate data)

Registration timing is part of that history. Property Monitor’s October 2023 report described apparently weak off-plan activity as incomplete reporting and discussed developers’ registration lag. The lesson is specific: a registration count is not a clean, immediate measure of the month in which buyers committed. A batch of registrations can change a monthly comparison without a matching change in underlying demand. (Property Monitor)

The current regime is weaker than that broad historical reference class. Projectory’s supplied monthly figures show a marked contraction during 2026. The full January–October history provided for its premium and all-home series is shown here. These remain provider-defined populations rather than certified resolution counts. (Projectory aggregate data)

2026 month	Premium sales	All-home sales
January	500	14,928
February	434	14,956
March	320	12,511
April	413	12,857
May	223	9,568
June	189	12,196
July	236	12,737
August	199	11,163
September	184	10,100
October	160	9,999

This supports two conclusions that pull in opposite directions. Luxury activity has deteriorated enough to make another substantial decline credible. But October is already the lowest premium-sales month in this supplied sequence. The question does not compare December with the stronger conditions at the start of the year; it compares December with that depressed October base.

Evidence nearer the target threshold strengthens the bearish case. Cavendish Maxwell reported 257 residential transactions worth at least AED 20 million in Q3, down 28.4% quarter on quarter and 40.5% year on year. The total comprised 148 off-plan and 109 ready-property transactions. These are quarterly figures, and their record-level alignment with the question’s exclusions has not been established. Even so, they are direct evidence that weakness extends to the relevant price band. (Cavendish Maxwell)

The higher-priced market tells a similar story. Knight Frank reported 131 US$10-million-plus residential sales in April–June and 86 in July–September, compared with 103 in July–September 2025. Within Q3 2026, the monthly counts were 36 in July, 32 in August and 18 in September. The threshold is substantially above the question’s, so this is directional evidence rather than a substitute denominator. Its cumulative headline also does not reconcile with the earlier first-half figure, which limits the usefulness of that headline. (Knight Frank)

The main counterweight is the split between off-plan weakness and ready-home resilience. Cavendish Maxwell reported ready luxury sales rising 6.9% quarter on quarter. Projectory’s October report also described recovering ready-home activity. Springfield reported October residential sales value increasing 12.4% month on month and secondary-market volume increasing 9.28%. These facts do not prove that qualifying luxury counts will hold up. They do show that the market is not declining uniformly. (Cavendish Maxwell; Projectory; Springfield release)

Prices provide a weaker stabilization signal. ValuStrat reported a September residential-value decline of 0.1% month on month, with values unchanged in approximately 80% of tracked villa communities and 77% of apartment communities. Stable prices do not ensure stable transaction counts, but this evidence weighs against a simple story of accelerating deterioration across every measure. (ValuStrat)

There is also a potential registration pipeline. A Solaya Building 8 project brief specified an expression-of-interest deadline of 28 October 2026 at 17:00 Dubai time. That establishes a marketing milestone, not completed sales or December registrations. No verified reservation count, qualifying price distribution or registration schedule is available. It is therefore a limited upside signal for activity, not an assumed December sales batch. (Project brief)

What's non-obvious

A weak market is not the same as a qualifying two-month collapse. The obvious bearish reading focuses on annual declines and the retreat from earlier luxury peaks. The resolving comparison starts later, after much of that retreat. October’s already-low activity makes another large proportional decline a separate proposition. This is why the evidence supports meaningful downside risk while leaving No as the leading outcome.

The other easily missed issue is population definition. Projectory reported 160 October home sales above AED 10 million, while Continental Club reported 104 at that threshold. Projectory includes some villas recorded under land categories after matching them to villa inventory; the question excludes land. Continental Club applies an additional price-per-square-metre validity filter. Springfield separately reported 66 October residential sales above AED 25 million, including 19 above AED 50 million. These figures are not interchangeable, and the differences have not been fully reconciled. A plausible-looking published luxury total is not necessarily the count this question resolves on. (Projectory; Continental Club; methodology; Springfield release)

Uncertainties

The largest gap is the absence of a certified October denominator satisfying every rule. A provisional count of 20 has not been established as complete and eligible. It must not be treated as the final baseline. The missing evidence is a reconciled October record list with valid values, correct Dubai dates, eligible procedures and verified single-property status.

December’s project-level concentration is also unknown. An unusually large October batch would make the required decline easier to reach. A December luxury registration batch would work the other way. The available pipeline evidence does not establish which month will receive qualifying records.

Finally, the resolution uses the 20 January 2027 publication vintage. October has more time than December for late records to appear, which can depress the measured December/October ratio. Corrections to prices, dates and classifications can change either count. Historical seasonality for the exact eligible population remains unavailable. These gaps leave substantial room for Yes, but they do not displace the forecast’s central distinction: established market weakness does not establish another decline of the required size.