The bridge is 90% likely, the eco-city under 25% — and Malé needs 1,900 homes a year, not 65,000
India-financed corridor works will finish late but finish; RasMalé's promised 65,000-unit city is reclaimed sand with no financing for anything on top. The measured housing problem, meanwhile, is a bounded Malé deficit inside a much larger affordability gap — and pledged paper supply already exceeds every demand scenario Nyra can construct to 2040.
Three tiers of credibility
Every infrastructure conversation in the Maldives mixes three different kinds of object as if they were one: projects that are finished, projects that are financed, and projects that are announced. As of July 2026 the sorting is unambiguous. Finished: the new Velana International Airport Terminal 1, opened 26 July 2025 with 7.5m passengers/year of design capacity [OFFICIAL, MACL via trade press] at a terminal cost of ~USD 585m [MEDIA, Maldives Independent reconstruction — no audited total exists] inside a ~USD 1bn airport programme; and Hanimaadhoo International Airport, opened 9 November 2025 — 1.3m passengers/year, USD 136.6m on an India EXIM line [OFFICIAL]. Financed and committed but late: the Thilamalé bridge (Greater Malé Connectivity Project), a 6.74 km Malé–Villimalé–Gulhifalhu–Thilafushi corridor on a ~USD 500m Indian grant-plus-credit package [OFFICIAL], reported ~52% complete in March 2025 [MEDIA relaying official statements] and officially rescheduled to 2027 after three missed interim deadlines [OFFICIAL via Atoll Times, Mar 2026]. Announced and unfinanced above the sand: RasMalé, the 1,009-hectare Fushidhiggaru reclamation billed as a 65,000-unit eco-city, roughly 60% reclaimed at June 2026 [MEDIA, Dredging Today] with no identified financing for trunk utilities, connection or housing superstructure [OWN-CALC assessment on OFFICIAL record].
Data table
| Period | Thilamalé bridge (GMCP) | RasMalé reclamation |
|---|---|---|
| Sep 2023 | 33 | — |
| Jul 2024 | — | 10 |
| Dec 2024 | 43.6 | — |
| Mar 2025 | 52 | — |
| Nov 2025 | — | 30 |
| Mar 2026 | — | 50 |
| Jun 2026 | — | 60 |
The chart’s real content is the gap between the lines and the calendar around them. GMCP has missed every published interim milestone (Malé–Villimalé by December 2025, Malé–Gulhifalhu by March 2026, full corridor by September 2026) yet keeps advancing — the profile of a project that finishes late. RasMalé’s line climbs too, but every completion statement made about it since December 2023 — eight months to finish; majority by end-2025; done “within 2026” — has been superseded [MEDIA record]. Neither project has ever had an independent engineering audit; all percentages are government or contractor claims relayed by media, and the RasMalé points do not fully reconcile across sources. We plot them as reported and weight them accordingly.
Project by project: status, financing, and who gains
| Project | Scope / cost | Financing | Status Jul 2026 | Nyra odds (OWN-CALC) | Property winners / losers |
|---|---|---|---|---|---|
| VIA Terminal 1 | 7.5m pax/yr; terminal ~USD 585m, programme ~USD 1bn | Saudi ~USD 217m, Abu Dhabi ~90m, Kuwait ~50m, OPEC Fund ~50m + GoM [OFFICIAL/MEDIA] | Open since Jul 2025 | Realized | Resort and guesthouse land economy-wide; Hulhumalé; Hulhulé zone |
| Hanimaadhoo airport | 1.3m pax/yr, 2.46 km runway; USD 136.6m | India EXIM LoC (2019) [OFFICIAL] | Open since Nov 2025 | Realized | HA/HDh/Sh northern tourism belt; Kulhudhuffushi |
| Thilamalé bridge (GMCP) | 6.74 km corridor, ~3.6 km over water; ~USD 500m | India: USD 100m grant + 400m EXIM credit [OFFICIAL] | ~52% (Mar 2025); official date 2027 | 90–95% complete by H1 2028 | Villimalé (largest relative repricing); Gulhifalhu; Thilafushi industrial; Malé warehouses lose scarcity rent |
| Gulhifalhu | ~150 ha reclaimed 2020–24 (Boskalis) [INDUSTRY-EST] | Reclamation done; port relocated to Thilafushi Jan 2024 | Land banked; ~3,640 Binveriya plots designated [MEDIA] | Utilities are the gate | Housing value up on port removal + bridge; realization waits on trunk power/water/sewerage |
| Thilafushi masterplan | ~152 ha reclaimed [OFFICIAL, HDC]; MPL logistics terminal; ~500 t/day WTE plant | HDC plot sales; ADB/AIIB (WTE) | Plot sales live; WTE last verified “on schedule for late 2026” in Mar 2025 | ~65% for WTE + terminal by end-2027 | Industrial land up with price discovery; Malé godown premium down; air quality up if WTE runs |
| RTL ferries | 6-zone national network; 11 atolls served Jul 2025 [OFFICIAL, MTCC] | MTCC, fare-subsidised | Near-national; exact mid-2026 coverage unverified | Operating; service cuts possible under fiscal stress | Outer-atoll utility broadly: Kulhudhuffushi, Addu, Fuvahmulah, atoll capitals |
| RasMalé | 1,009 ha (~3x Hulhumalé); claims of 30,000–65,000 units [OFFICIAL claims, unreconciled] | Reclamation on state/MACL account; superstructure unfinanced | ~60% reclaimed; fuel-cost slowdown Jun 2026; zero vertical construction | ~80% reclamation finishes; under 25% for at-scale city within 10 years | Near term: sentiment overhang on Hulhumalé Phase II; if ever delivered: dilutes Malé/Hulhumalé scarcity premium |
| Regional hospitals | 5 tertiary sites; CNEEC ~USD 38m for Kulhudhuffushi + Thinadhoo [MEDIA] | China contracts; AIIB add-ons (MVR 53.56m KRH, Jan 2026 [OFFICIAL]) | Staged construction | Buildings likely; service quality the risk | Kulhudhuffushi clearest gainer — airport + hospital + RTL node compound |
Three causal readings matter more than any row. First, the completed airports are realized value events, already partly capitalized into land: VIA’s 7.5m-passenger capacity against ~2.2–2.4m arrivals in 2025 removes the physical ceiling on tourism growth for a decade, and Hanimaadhoo re-rates the northern belt where new resort development has clustered [OWN-CALC on OFFICIAL capacity figures]. Second, the bridge is the single most property-relevant live project in the country because it converts three different land stories at once: Villimalé from ferry-only dormitory to connected Greater Malé node — the largest relative repricing available anywhere in the market; Gulhifalhu from stranded reclamation to urban extension; Thilafushi from landing-craft logistics to direct trucking. Third, reclamation percentages are the wrong indicator to watch. Land becomes valuable when trunk utilities arrive, not when sand does — Hulhumalé Phase II plots waited years for services, and Gulhifalhu’s ~3,640 plot recipients are waiting now. Utility financing, not dredging progress, is the leading indicator for value realization [OWN-CALC].
These probabilities are judgements, not measurements, and we state the method: financing committed versus identified versus absent; physical progress; and each project’s own deadline record. The gradient is steep because the financing tiers are categorical. The bridge is ring-fenced bilateral Indian money with prestige value to both governments — it finishes. RasMalé’s reclamation is cheap relative to what comes next and runs on state dredgers — it probably finishes too, on a date no official statement has yet predicted correctly. But the eco-city on top requires utilities, a ~7 km connection to the Malé region, and vertical construction at roughly six times the scale of Hiyaa — against a delivered state track record of ~530 units a year [OWN-CALC, below] — with no financing package announced as of the vantage date.
The fiscal wall behind every announcement
The reason announcement-tier projects deserve heavy discounting is not cynicism; it is arithmetic the IMF has now stated twice. The February 2025 Article IV mission called reprioritization of the public investment programme “critically necessary”; the June 2026 mission (PR 26/208) credited the 2025 consolidation and the handling of the USD 500m sukuk maturity with easing immediate solvency concerns, but kept overall and external debt-distress risk at high, projected public debt to climb above 140% of GDP over the medium term, and cut 2026 growth to ~1% [OFFICIAL]. Total state debt stood at MVR 155bn — roughly USD 10.1bn at the 15.42 peg — or ~130% of GDP at end-2025 [MEDIA citing official figures]. Gross reserves fell from USD 1.3bn in March 2026 to USD 718m in April after the sukuk repayment; usable reserves are USD 244m on Fitch’s measure and USD 148m on the World Bank’s — a definitional conflict we report rather than average. Gross reserves equate to about 1.4 months of imports; on either usable measure, cover is well under one month [MEDIA]. The 2026 fiscal-deficit projections diverge just as sharply: 10.9% of GDP (World Bank) against 14.6% (Fitch), reflecting different treatment of arrears and SOE support [MEDIA; conflict carried].
The parallel exchange rate makes the constraint physical. At ~MVR 20.5–20.7 to the dollar [MEDIA, mid-2026 — above 20 for longer than at any time on record] against the official 15.42, every import-dependent project — which is all of them — faces effective input costs roughly a third higher than peg-based budgets assume, squeezing exactly the SOE contractors (MACL, MTCC, HDC) on whose balance sheets the announcement-tier pipeline sits. The state can still complete what others are paying for. It cannot currently start anything large of its own.
What the housing stock actually looks like
Against this infrastructure story stands the demand side, and here the first finding is institutional: nobody has measured the shortage. The Ministry of Housing, Land and Urban Development told a UN statistics webinar in June 2024 that no timely statistics exist on either the supply or the demand side of housing [OFFICIAL admission]. There is no dwelling-stock register, no vacancy count, no completions series. What exists is the census — which counts households, not dwellings.
Census 2022 enumerated 515,132 people and 94,424 households — 42,571 (45%) in the Malé region [OFFICIAL, MBS]. Some 83,965 households lived in housing units (an occupied-stock proxy; vacant units are invisible to the census) and 10,459 in collective quarters, overwhelmingly foreign-worker dormitories averaging ~11.7 persons [OFFICIAL]. Average household size has fallen from 6.5 (1990) to 5.5 nationally and 5.0 in Malé [OFFICIAL]. Severe overcrowding — more than three persons per sleeping room — affects 8% of Malé households on the census definition; the World Bank’s broader 2021 measure put Greater Malé at ~12% [OFFICIAL/IFI — different definition and year, reported side by side]. Exactly 29 households were recorded homeless nationwide [OFFICIAL]. Half of Malé households rent, at an average MVR 14,364/month — USD 931 at the peg, but only ~USD 700 at the parallel rate, which is the honest dollar value of a landlord’s cashflow in mid-2026 [OFFICIAL rent; OWN-CALC conversion].
Data table
| Period | Republic | Malé region | Atolls |
|---|---|---|---|
| 1990 | 29,823 | 5,613 | 24,210 |
| 1995 | 35,241 | 7,298 | 27,943 |
| 2000 | 41,177 | 9,886 | 31,291 |
| 2006 | 46,194 | 14,107 | 32,087 |
| 2014 | 68,249 | 26,739 | 41,510 |
| 2022 | 94,424 | 42,571 | 51,853 |
The chart explains why demand pressure persists even as population growth slows: household formation, not population, is the demand engine. Households grew 38.4% in 2014–22 — about 3,270 a year nationally and ~1,980 a year in Malé — while household size fell from 5.9 to 5.5 [OFFICIAL]. Fission does the work: on a constant population, size falling from 5.5 toward 4.5 alone would add roughly 20,000 households [OWN-CALC]. Every year, the same people need more front doors.
Deficit claims, attributed
The deficit numbers in circulation are claims, and they should be quoted with their owners attached. The ministry’s working figure is a backlog of “around 25,000 people in need of housing in the Malé region” [OFFICIAL claim, Jun 2024] — persons, not dwellings; at Malé’s 5.0 persons per household that is ~5,000 household-equivalents [OWN-CALC, illustrative]. The same deck claims two-thirds of the population resides in the Malé region — a figure the census contradicts (41% of enumerated population, 45% of households) and which we treat as rhetorical [conflict carried; census preferred]. No World Bank or IMF document we reviewed states an independent unit-deficit estimate; the Bank frames the problem as affordability and fiscal risk, not a unit gap [OWN-NOTE].
The affordability arithmetic is the strongest number in the whole domain. The World Bank’s October 2022 housing study found the cheapest available 2BR flat (~MVR 1.8m / USD 117,000) requires roughly MVR 46,000 a month (USD 3,000) of income to finance — attainable only for the top income quintile [OFFICIAL/IFI]. The cheapest 2BR rental was assumed at MVR 18,900/month against bottom-60% incomes that force income-pooling — hence two to four families per unit. Rent absorbed ~35% of the average Malé household’s income in HIES 2019, and over 68% for the poorest [OFFICIAL/IFI]. The queue for state housing is mostly priced out, not roofless.
Demand to 2040, with the arithmetic shown
Our scenarios [OWN-CALC] are deliberately simple: official population projections divided by assumed household size, minus the 2022 base. The basis is the MBS Population Projection 2022–2062 (Feb 2026) [OFFICIAL], whose mid-year 2022 resident base (547,651; Malé 227,613) exceeds the enumerated census count — we therefore compute an implied 2022 household base on the same footing: 547,651 ÷ 5.5 = 99,573 nationally, 227,613 ÷ 5.0 = 45,523 for Greater Malé. Populations are interpolated geometrically between MBS’s decadal anchors. The scenario lever is household size: low freezes it at 2022 levels; base continues the observed decline of ~0.05/year (Malé: 4.7 by 2030, 4.5 by 2035, 4.3 by 2040); high assumes faster fission toward regional urban norms (Malé 4.5/4.2/4.0).
| Incremental households vs 2022 | 2030 | 2035 | 2040 |
|---|---|---|---|
| Republic — low | +13,463 | +19,154 | +22,682 |
| Republic — base | +19,985 | +31,027 | +40,510 |
| Republic — high | +27,305 | +42,384 | +56,799 |
| Greater Malé — low | +9,437 | +14,177 | +17,817 |
| Greater Malé — base | +12,945 | +20,810 | +28,128 |
| Greater Malé — high | +15,544 | +25,548 | +33,652 |
Data table
| Period | High (faster household fission) | Base | Low (household size frozen) |
|---|---|---|---|
| 2022 | 0 | 0 | 0 |
| 2030 | 15,544 | 12,945 | 9,437 |
| 2035 | 25,548 | 20,810 | 14,177 |
| 2040 | 33,652 | 28,128 | 17,817 |
Read the base row causally: Greater Malé needs roughly 1,000–1,900 new dwellings a year to 2040 across the scenario band — 28,128 over 18 years in the base case, or ~1,560 a year. Add the ministry’s backlog claim (~5,000 household-equivalents, if taken at face value) and a nominal replacement allowance of 0.3–0.5% of the ~84,000-unit occupied stock a year (~4,000–8,000 units by 2040 [OWN-CALC assumption — no stock-condition data exist]) and the ceiling of plausible Malé-region need to 2040 is on the order of 40,000 dwellings. Three caveats bind: 2040 values extrapolate beyond MBS’s 2032 anchor and carry ±10%; foreign workers (30% of residents, projected 36% by 2032 [OFFICIAL]) are largely housed in labour quarters, a different product; and these are formation-based needs, not financeable demand — at World Bank affordability parameters, most incremental households cannot buy at market prices, so unsubsidized supply does not automatically clear.
Pledges vs delivery: the pipeline in one frame
The state’s delivered record in Greater Malé over two decades is ~10,656 units: Hulhumalé Phase I 2,592 (2004–17), Hiyaa 6,720 (occupied from 2022), Vinares 1,344 (2023–24) — about 530 units a year, exceeding household formation only during the 2022–24 handover spike [OFFICIAL components; OWN-CALC rate]. Under construction at mid-2026: roughly 7,260 units — FDC’s 4,000 (USD 226.5m India EXIM) plus BML Affordable Home Leasing’s 3,260 (started January 2026, due August 2028) [OFFICIAL announcements; no completions register verifies any of it]. Against that, the pledge stack: the 2026 “Housing for All” policy promises 22,900 homes — 15,000 plots of 1,250 sq ft (116 sqm) plus 7,900 flats [OFFICIAL policy]; the Binveriyaa scheme has already allocated 9,003 plots to ~18,950 recipients, much of it unreclaimed lagoon at allocation and partially redrawn in January 2026 [OFFICIAL/MEDIA]; RasMalé claims 30,000 or 65,000 units depending on the statement [OFFICIAL claims, mutually inconsistent — carried, not resolved].
The frame yields one of two futures, and both are stated in the chart. Either most pledges do not materialize on anything like the stated timeline — the base case, given financing — or, if RasMalé were actually built out, the Malé region would face a localized oversupply of a specific product type while the affordability problem it was meant to solve remains untouched. Paper supply exceeding even high-scenario need is not evidence of a solved shortage; it is evidence that pledging has decoupled from delivering. The historical delivery model is not repeatable either: every completed programme was financed by external debt that now constitutes about a third of external public-and-publicly-guaranteed debt (USD 739m at end-2021, 86% to Chinese lenders) [OFFICIAL/IFI] — precisely the exposure the IMF’s debt-distress rating now prices.
Diagnosis: four shortages wearing one name
Physical shortage — real, bounded, Malé-concentrated. Severe overcrowding of 8–12% of Malé-region households depending on definition, 29 homeless households nationally, formation of ~2,000 households a year in Malé against long-run delivery of ~530. The physical gap is thousands of units — not the tens of thousands implied by reading pledges as need. Confidence: medium-high [census-anchored; no dwelling or vacancy register].
Affordability shortage — the dominant binding constraint. Only the top quintile can finance the cheapest 2BR; the cheapest rental forces income-pooling on the bottom 60%; rent burdens run 35–68% of income. Most of the housing queue is a financing queue. Confidence: high [WB, HIES and census rents concur].
Type mismatch. Hiyaa delivered 6,720 identical 550 sq ft (51 sqm) 2BRs at MVR 1.55m per unit all-in cost (USD ~101,000) [OFFICIAL/IFI] for households averaging 4.5–5.0 persons; plot schemes hand 1,250 sq ft self-build plots to households without construction finance, where self-build averages ~15 years [OFFICIAL]; atoll programmes built ownership units where demand is rental and mobility. Supply and need are mismatched on size, tenure and location simultaneously. Confidence: medium-high.
Allocation inefficiency. Rent-to-own default of 37% (90+ days, Jul 2022) [OFFICIAL/IFI], with HDC carrying MVR 7.2bn (USD ~467m) of guaranteed debt and MVR 628m (USD ~41m) of unpaid rent by early 2026 [MEDIA]; subleasing of subsidized units at up to 10 times the programme rent described as common practice [IFI-reported; magnitude INDUSTRY-EST]; atoll units vacant 2–7 years before lease-up [OFFICIAL/IFI]; plots allocated from unreclaimed lagoon and redrawn with the political cycle. A material share of delivered subsidy leaks to non-target users while allocated assets sit idle. Confidence: high.
The synthesis for an investment or policy reader: the Maldives does not have one housing shortage. It has a moderate physical deficit wrapped inside a large affordability gap, amplified by product mismatch and allocation leakage, financed by debt the sovereign can no longer cheaply raise. Infrastructure changes where value sits — Villimalé, Gulhifalhu, Thilafushi and the Hanimaadhoo belt are the genuine winners of the committed pipeline — but no bridge resolves a gap whose binding constraint is income against price. Policy that measures nothing will keep solving the wrong problem at the wrong scale in the wrong place.
What we don’t know
- True project progress. No independent engineering audit exists for GMCP or RasMalé; all percentages are official claims via media, and RasMalé’s reported area basis itself conflicts (1,009 ha in most statements; 1,153 ha also cited at launch).
- The dwelling stock. No register of dwellings (occupied plus vacant) exists anywhere; ~84,000 occupied units is a census proxy. Vacancy is unmeasured — the atoll social-housing vacancy evidence is a World Bank case narrative, not a statistic.
- The pipeline. No building-permit or completions series exists; every under-construction count in this paper is announcement-based. Private-sector supply is entirely unquantified.
- Utility schedules. Gulhifalhu trunk-utility and plot-handover timing is unpublished; the STELCO Sixth Power Project (100MW, Hulhumalé) is unverifiable beyond announcement; Thilafushi WTE commissioning has not been re-verified since March 2025.
- Incomes since 2019. All affordability arithmetic rests on HIES 2019 and WB 2021–22 parameters; post-COVID, post-Hiyaa burdens have not been re-measured.
- Which exchange rate governs. Conversions here use the MVR 15.42 peg; at the documented parallel rate of ~MVR 20.5–20.7 [MEDIA], dollar-denominated project costs rise and the dollar value of MVR rents falls by roughly a quarter to a third. Which rate a given contractor, landlord or lender actually faces is undisclosed — and materially changes every economics statement in this paper.