Will the median annual rent on new residential apartment leases registered in Dubai in December 2026 be lower than in October 2026?

TL;DR

The forecast is 57.1% Yes and 42.9% No, conditional on non-annulment. Falling rents and additional housing supply support a lower December median, while exact ties and changes in the apartment mix keep the result close. Treat this as a modest downside advantage, not a prediction of a sharp rental decline.

Context

Dubai’s rental market is softening, but the question asks something narrower than whether tenants gain bargaining power. It compares the unadjusted median annual rent of qualifying new apartment leases registered in December 2026 with October 2026. The population excludes renewals, non-residential properties and multi-unit contracts, as well as records with missing or zero required monetary values or area. DLD distinguishes registration date, lease start date, annual amount and unit count; those distinctions matter because nearby rental statistics often measure something else.

The forecast is assessed as of 4 November 2026. December registrations are still ahead, and October’s current figures remain provisional. Resolution uses the data published on 20 January 2027. Equality resolves No, so even a market that becomes cheaper for many tenants can fail to produce a Yes outcome.

Evidence

The historical backbone argues against treating a December decline as automatic. A start-dated proxy covering October–December comparisons from 2012–2025 contains seven declines, four equalities and three increases. Its strict-decline frequency is 50%. The dating and property-population differences prevent treating this as the exact resolution series; the publisher’s methodology explicitly uses contract start dates rather than registration dates.

The full annual comparison available in that historical extraction is below. These are proxy observations, not verified registration-dated apartment-only resolution values. The source series is Dubai Real Estate Data.

Year	October median, AED/year	December median, AED/year
2012	48,000	48,000
2013	60,000	60,900
2014	70,000	72,000
2015	75,000	75,000
2016	74,999	70,000
2017	67,000	65,000
2018	61,526	59,000
2019	54,000	53,000
2020	46,000	44,372
2021	42,000	43,000
2022	50,000	50,000
2023	60,000	57,000
2024	70,000	68,000
2025	75,000	75,000

This history supports two conclusions. Year-end declines are familiar, but they are not a reliable seasonal rule. And unchanged medians recur often enough to matter. Excluding ties would answer the wrong question: every equality here belongs on the No side.

Current evidence gives the forecast its downward tilt. A reported extraction of DLD rental records, applying registration-date and single-unit apartment filters, puts October at AED 70,000 across 31,439 qualifying rows, down from AED 73,000 in September. That is the closest reported population match, but it was not independently reproduced. Its treatment of repeated rows and verification of all required monetary exclusions remain unresolved.

A separate published series from YAFF also places October at AED 70,000, across 30,451 contracts using records through 2 November. YAFF restricts leases to approximately one year, which the resolution does not require. Agreement on the headline median is useful corroboration, not proof that either current figure will become the resolving October value.

The full January–October history available for these two series shows both recent weakness and persistent plateaus. The DLD extraction figures are reported calculations, rather than independently verified official monthly summaries; YAFF’s figures remain a differently filtered proxy.

2026 month	Reported registration-dated median, AED/year	Reported qualifying rows	YAFF median, AED/year
January	70,000	23,270	70,000
February	75,000	20,719	75,000
March	70,000	13,171	70,000
April	68,250	15,105	68,000
May	70,000	15,209	70,000
June	70,000	22,604	70,000
July	70,000	24,641	69,000
August	72,000	27,623	70,000
September	73,000	33,361	72,000
October	70,000	31,439	70,000

October’s decline is therefore not evidence of a steady slide throughout the year. Both series show rebounds and repeated round-number medians. That limits how far the latest monthly movement can be extrapolated.

Broader rental research supplies stronger evidence that the recent weakness is real. Cushman & Wakefield Core’s report, published 29 October, says average residential rents fell 4.5% quarter-on-quarter and 7% year-on-year during July–September, with declines across all its major apartment submarkets. These are consultancy measures, not the question’s raw apartment median. They support downward pressure without establishing the December result.

Savills’ Q3 report gives a more restrained version of the same story. New-lease rental rates fell quarter-on-quarter in seven of eight selected apartment communities, with declines ranging from 0.5% to 3.1%; Majan increased 1.3%. Its comparisons control for property type and bedroom configuration and measure rent per square foot. Savills also describes early stabilisation. I read this as evidence for continued softness, but against assuming that recent broad declines continue at an unchanged pace.

Supply supports the downside case, though timing matters. Core reports approximately 40,400 completions during January–September and another 10,500 expected in the fourth quarter, implying 50,900 for the year. It also identifies contractor and supply-chain constraints pushing handovers into 2027. Expected completions are projections, not homes already available for December tenants.

Demand is a real counterweight. The same Core report records new rental contracts rising 32% quarter-on-quarter and 12% year-on-year. Falling rents alongside rising leasing activity fit a market absorbing more supply. They do not describe a collapse in demand.

Published October levels also disagree. UAEPropertyAPI reports an apartment median of AED 65,000, compared with AED 67,000 in September. Its population and adjustment methods do not establish an exact match to the resolution filters. The disagreement with AED 70,000 is a measurement warning, not a reason to average the two levels.

What's non-obvious

A softer rental market does not guarantee a lower median. In the reported October extraction, 48.60% of qualifying rows sit below AED 70,000, 2.86% sit exactly at it and 48.54% sit above it. If that baseline remains unchanged, December needs roughly another 1.40 percentage points of its distribution below the threshold before the median crosses it. Many individual tenants can obtain reductions while the citywide middle contract remains at the same round number.

New supply can also push the statistic in opposite directions. It expands choice and puts pressure on comparable rents, but newly leased buildings can be more expensive than existing stock. RENT DUBAI identifies 85 towers with first leases in 2026, containing 7,358 leases through 15 October; 65 rented above their communities’ per-square-foot rates. Its start-date and lease-duration rules differ from the target, so this does not prove a rising citywide median. It shows why supply growth and a higher-quality registration mix can coexist.

Uncertainties

The largest unresolved issue is the exact October baseline. The closest reported extraction contains rows that are identical across observable fields, but identifiers did not establish whether these represent duplicated records or genuinely separate contracts. Removing identical rows changes its October median from AED 70,000 to AED 65,000. Neither automatic deletion nor automatic acceptance settles the identity question. Complete verification of the monetary exclusions is also missing.

Historical evidence is thinner than its length suggests. The long seasonal series uses start dates and does not establish the exact apartment-only registration population. The current matching series covers only part of one market cycle. It shows stickiness and composition risk, but cannot supply a robust registration-dated seasonal rule.

The remaining market uncertainty is whether further softness outweighs stabilisation, strong leasing activity and delayed deliveries. A December shift toward larger or more expensive apartments can lift the unadjusted median even when comparable rents fall. Conversely, more lower-rent registrations can move it down without a uniform market-wide reduction.

The data published on 20 January 2027 can also revise both months. A fully checked October distribution, clear record-identity treatment, matching historical registration data and November registrations would close the largest evidence gaps. Until then, the evidence supports the saved modest downside tilt while preserving substantial room for equality or a higher December median.