Prepared for Amwal Capital. Enter your email to continue.
Confidential. By continuing you agree we may record which sections you view.
We want to share where we think the Dubai market is heading, the work we did to get to that view, the underwriting decisions we took off the back of it, and the data we are leaning on.
When the conflict hit, it was impossible to predict where the Dubai market was heading in the near term. Rather than deploy capital on assumptions that no longer held, we paused and did the work to form a clear view.
We feel we now have that view, and we are acting on it. In this memo we cover where we think the market is going, the research behind it, the underwriting decisions we took as a result, the public and internal data we are relying on, and why we think the book this produces holds up, including what it costs us to build it this way.
The correction so far sits in the ready, secondary market, and within it the areas that still have a lot of off-plan supply to be delivered. We read it as an off-plan repricing that is showing up early through the ready market.
Solid is what has happened, dashed is our expectation. We expect volume to recover after the summer. Value is the more uncertain part: it depends on the memorandum, and could bottom later in the year.
The memorandum signed in mid-June is the biggest single factor for where the market goes from here. We think it is a real step down from the conflict and it takes the worst-case risk off the table for now. We also think it is provisional rather than a settled peace, and that is the part that shapes how we underwrite.
The bigger signal for us is what the memorandum says about appetite for more conflict. The only belligerent force right now is Israel, and Israel cannot run a full-scale war on its own without US backing. Given where US domestic politics sit, and now the memorandum on top of that, we do not think that backing continues, so we expect any further escalation to stay limited. We also note that in the latest round the UAE was not targeted, having been heavily targeted earlier in the conflict, which we read as a sign of serious back-channelling between the UAE and Iran. The situation could still turn structurally unsustainable, and we believe we will get clarity on that over the next six months, but our base case is peace with occasional and limited escalations.
The memorandum was agreed in principle in mid-June 2026, with formal signing in Switzerland expected on 19 June. The terms here are from public reporting, and the follow-on talks on nuclear and sanctions issues are still open.
We look at two angles that are correlated and feed into each other. The conflict outlook drives the market path, and underneath both sits the trend we watch most closely: whether new supply outruns population growth.
Dubai's oversupply risk is real but conditional, and the conditional part is what matters. About 366,000 units are scheduled to deliver across 2026 to 2028 (handover forecasts), but on past form only about half actually land, so the realistic number is closer to 180,000. At an average of three people per household, absorbing that fully takes roughly 180,000 new residents a year. Dubai was running well ahead of that before the conflict: the population grew about 6% in 2025, roughly 231,000 people, to just over 4 million. The risk is the post-conflict downgrade. In its March assessment, Citi cut the forecast to about 1% growth in 2026 and 2.5% a year from 2027 to 2031 (Reuters), which adds closer to 85,000 residents a year and leaves the realistic supply running ahead of demand by roughly 32,000 units per year over the three years. That gap is the oversupply, and it closes as soon as growth recovers: back toward 4% to 5% a year, still below 2025's 6%, absorbs the realistic supply in full. So the oversupply is driven by the post-conflict slowdown, not by the supply itself, which is the link from the conflict outlook straight to the market and part of the softening we are already seeing.
| Scenario | Likelihood | How we respond |
|---|---|---|
| MoU holds, occasional sparring Base case: flare-ups, no sustained war | Around 55% | We expect temporary market adjustments and keep deploying consistently, with the discipline on employment profile and location. Demand picks up through the wobbles, as buyers treat the temporary corrections as buying windows. |
| MoU holds, lasting peace A durable settlement is reached | Around 25% | Confidence returns and the market re-rates up steadily over 6 to 12 months. We lean in and speed up capital deployment. |
| MoU breaks, conflict restarts Hard to sustain without US backing | Around 20% | We do not pause. We get more conservative, focus on the most resilient stock, and protect the book. |
| Scenario | Likelihood | How we fare | Our plan |
|---|---|---|---|
| Choppy for 12 to 18 months Prices range about +5% to -15%, like the 2014 to 2020 soft cycle | Around 60% | The book holds, because we deploy into resilient, end-user stock that is less exposed to upcoming off-plan supply and softer rental demand. | Deploy steadily and selectively, at a more conservative pace than usual. |
| V-shaped recovery Like COVID 2020: about nine months from bottom to recovery (Gulf News, 2020) | Around 20% | The book is well placed and collateral re-rates up quickly. | Scale lending faster and ride the recovery alongside our buyers. |
| Managed correction, up to about 30% Still milder than the 2008 crash, which fell about 50% | Around 20% | The hit lands mostly in oversupplied, investor-heavy stock we avoid. Even a 30% area drop rarely drives defaults in an end-user book (see below). | Keep the funnel tight, lean into prime and owner-occupier, and use the distressed openings it creates. |
The likelihoods are our own rough weightings and will move as the picture develops.
We looked at the last few Dubai and US real-estate crises to understand the current situation and draw parallels. We were after two things: how far prices actually fell, and what made some borrowers and some areas hold up while others did not. The same two drivers came up every time, and they are the two our updated underwriting is built on: which type of borrower kept their job, and how much new supply was coming to the area they bought in.
| Crisis | What happened, in numbers | What we took from it |
|---|---|---|
| US GFC 2008–2012 |
US national house prices fell about 27% from the 2006 peak to the 2012 trough. It was also an employment shock: roughly 8.7 million jobs went, concentrated in construction (down about 20%), finance and manufacturing. The worst price damage was in over-supplied, speculative Sun Belt markets, while supply-tight cities held up far better. | Even the worst US crisis in memory was about 27% nationally, not the wipeout people picture, and the damage tracked supply and leverage, not the average. The defaults came from lost jobs, not from price falls on their own. |
| Dubai crash 2008–2010 |
Dubai prices fell roughly 50% from the 2008 peak, with the worst, 50% to 60% in places, in off-plan and speculative stock. Completed, end-user stock in core areas fell less and recovered first: Downtown and Marina were back to new highs by 2013 to 2014, with villas up 23% and apartments up 14% year on year by 2012. | Completed, end-user, freehold property is structurally safer than off-plan and recovers first. This is the single biggest reason we lend on ready stock only. |
| Dubai soft cycle 2014–2020 |
A long, slow drift of roughly 25% to 35% over six years on lower oil and steady oversupply. No crash, just years of gradual decline, with communities moving very differently from one another. | A slow drift is fine for our book. End-user borrowers hold through it, they do not realise the loss, and they keep paying as long as they are employed. Even a flat market works for us. |
| COVID, Dubai 2020–2021 |
A confidence shock and a brief dip of around 5% to 10%, then a fast recovery: Dubai transactions were back to February levels by June 2020, within about four months, and prime prices were up sharply by late 2021. The job losses landed in hospitality, tourism, aviation and retail, where leisure and hospitality alone shed close to half its workforce at the trough. | Shocks can reverse fast, and the sector you work in decides who keeps paying. The hit fell on hospitality, tourism and aviation, the same cyclical sectors we now tier down, while government, finance and tech held. |
| US rate shock 2022 |
Mortgage rates roughly doubled, from about 3% to about 7%, and existing-home sales fell by roughly a third from their 2021 peak. Prices stayed fairly firm nationally but split hard by supply: Austin, which permitted close to 1,000 apartments per 100,000 residents, saw values fall about 24% from their highs, while supply-starved Northeast metros like New York, where permits fell about 85%, rose around 11%. Markets that built in line with demand stayed roughly flat. | What separated the fallers from the holders was upcoming supply. That is the core of our updated underwriting: avoid areas with a wall of off-plan to come, favour supply-constrained, end-user areas. |
Figures are drawn from the public sources listed at the end, and are there to show the patterns we tested against.
We are lending into a market that is losing value on average, so the way we add value is by being selective about what we take on. We look at two things that history shows actually matter in a downturn: the borrower's employment, and the property itself.
We rank every employer from tier one, the strongest, down to tier five, which we exclude by default. We also differentiate by size, since SMEs are more prone to disruption than large, established employers. Seniority matters too, so a senior manager or partner at a global firm sits higher than a junior at the same firm. We also set a minimum time-in-job that gets stricter as the tier gets weaker, and our system can re-rank whole sectors under a stress scenario so the selection tightens when the environment does.
| Sector | Normal market | Under regional stress |
|---|---|---|
| Government & public sector | Top tier | Gets stronger, held first |
| Defence & state-backed telecom | Top tier | Gets stronger |
| Banking, healthcare, technology | Strong | Mostly holds |
| Consumer goods (CPG/FMCG) | Strong | Mostly holds, defensive |
| Hospitality & tourism | Mid | Drops one to two tiers |
| Commercial logistics & shipping | Mid | Drops, trade-route exposed |
| Low-cost aviation, private media | Lower | Drops to the floor |
The scenario re-ranking is something we switch on to match the environment. It is not a live monitor of each borrower, and we test affordability when we first write the loan.
The property decides what we recover, and recovery comes from re-leasing. So before anything else, for any unit we ask two things: how strong is rental demand right there, and how much new supply is coming to compete with it.
Our baseline does not move: we lend only on ready or secondary units, either handed over years ago or at handover, and we never deploy into off-plan. We study off-plan closely, because its pipeline is what threatens the areas we lend in, but we do not hold it.
The value is in the nuance, not the headline community number. Three live examples:
Pulled from DLD-registered transactions (recent window, June 2026), and ranked by how well each is holding rather than by price. The clearest signal is ownership: stock that is mostly owner-occupied holds, because owners do not force-sell in a correction the way leveraged investors do. So we lean in on the supply-tight, owner-occupied stock, the villa communities and the land-scarce, where values have held best and in the strongest cases risen. We are selective in the investor-led prime apartment zones, deep and liquid but corrected from an overextended peak, where we take only the best end-user stock. And we cap the cheapest, highest-supply apartment zones.
| Community | Ownership | Stance |
|---|---|---|
| Dubai Hills, villas | Owner-occupied | Lean in |
| Arabian Ranches, villas | Owner-occupied | Lean in |
| Palm Jumeirah | Mixed | Lean in |
| Dubai Marina | Investor-led | Selective |
| Downtown | Investor-led | Selective |
| Business Bay | Investor-led | Selective |
| Arjan | Investor-led | Cap, supply |
| Dubai South | Investor-led | Cap, supply |
| JVC | Investor-led | Cap, high supply |
Source: Sooner stance from DLD-registered transactions and ValuStrat, June 2026. Ownership reflects the dominant buyer type in each community. Upcoming-supply pipeline counts are tracked separately.
On concentration. We also cap how concentrated the book can get, as a backstop. We cap any single developer's completed stock, which speaks to the off-plan and developer-stress narrative directly. We cap by employer and by employment sector, so the book is never overexposed to a single sector shock. We cap per building, which protects against a one-off building event like a fire. And we cap per community, to limit the impact of an area-level shock. Because we lend ready-only, none of this is the main event, but it is a clean backstop.
We take our market data from the Dubai Land Department's registered transactions, and we lean on ValuStrat for forward-looking valuation. On top of that sits our own analysis. The picture is consistent: volume led prices down into a low point in May, the correction is concentrated in the ready and mid-tier part of the market we lend into, early June is trending back up as deferred demand returns, and the supply risk is local rather than market-wide.
| Signal | What the data shows | Our read of where it goes |
|---|---|---|
| Prices | The ValuStrat index had its first quarterly fall since the post-pandemic recovery, but the monthly drop is slowing fast: 5.9% in March, 1.9% in April and just 1.2% in May, to 222.1 points, which ValuStrat reads as pointing toward stabilisation. The split is wide: villas are up 5% on the year, apartments down 1.4% (their first annual decline in six years). It also splits by community: supply-tight, established villa areas lead (Emirates Hills +12.6%, The Meadows +12.9%), while newer, higher-supply ones like Arabian Ranches Phase 2 are down 2.4%. ValuStrat, May 2026. | The slowdown points to stabilisation. The damage stays in high-supply, apartment-heavy stock; supply-tight villa communities hold and rise. |
| Volume | Roughly 12,900 transactions in May against about 17,800 in April, with value down close to 49% on the year, the low point of the correction so far. June is part-way through and early activity is trending back up. DLD. | May was the trough. Volume recovers as confidence returns, faster if the memorandum holds. |
| Two markets | The strong headline is off-plan, up about 9% on the year but held in place by developer payment plans rather than real demand. The ready, secondary market we lend into is moving first, with ready transactions down about 9% on the year in Q1, even as the average ready price held, up 5.6%. DLD; our analysis. | Off-plan reprices later. Our ready, supply-tight segment is where we want to be, and it re-leases at value. |
| Supply | Around 77,500 units are scheduled for 2026, but on past form only about half get delivered on time, concentrated in studios and one-beds in a few communities. Oversupply is local (JVC, Dubai South, Arjan), while Palm and the villa communities, Dubai Hills, Arabian Ranches and Emirates Hills, stay supply-tight. DLD pipeline. | Weakness stays concentrated in the high-pipeline apartment zones. We lean into the low-supply communities, stay selective in the investor-led prime apartments, and cap the cheapest high-supply stock. |
| Rents | Rent levels and volumes are the leading signal by area. Where new supply is heavy, rents soften first, and values follow to restore the equilibrium between rents and prices. We lend where rental demand is deep and supply is thin, so both hold. DLD. | Rents lead, values follow. Watching rent volumes shows us where values are heading before they move. |
Outside the conflict, most of the recent narrative around Dubai real estate has been about stress on some developers and their financial position. We track it closely. It mostly threatens off-plan, and we do not have exposure to any single developer's delivery or balance sheet.
We do think developer stress will be a real driver of off-plan stress, and we are already seeing it feed through into the ready market, as part of the repricing we described earlier. If it plays out, we expect two things. First, a managed correction rather than a collapse, supported by government intervention, as in 2009 when Abu Dhabi put up a $10 billion backstop for Dubai and Nakheel's bonds were met. The same government-related entities carry roughly $11.5 billion of debt maturing in 2026, and banks are expected to step in again. Second, a fall in upcoming supply if some developers pull back or fail to deliver.
It can move sentiment, mostly among foreign investors, who are not our target borrower. We expect some of that, and we are pricing it in.
Less supply means the market ends up less oversupplied, especially in the supply-tight, end-user areas we lend into. We carry no developer-balance-sheet or off-plan risk, our collateral is completed homes with resident borrowers, and the areas we focus on are the least exposed to a developer-led supply shock.
We know this is coming, we can see it, and we are pricing it in. We expect some impact on sentiment, and we are well protected from the substance of it.
These are deliberate trade-offs, and we want to be straight about both sides.
It is a selective market and we are taking a disciplined approach to it. As the book builds, we will share how it breaks down by tier, sector and area, so you can see the resilience in the actual numbers.