Case study APPLIED ANALYSIS 01

Should we build it?

A hypothetical 60-unit residential scheme in Hulhumalé Phase 2, modelled end to end on public evidence. The base case loses MVR 43m. The land would have to cost MVR 316 a square foot — against an auction floor of MVR 4,000 — for the numbers to clear at mid-market prices. That result is not a quirk of the assumptions; it is why private residential supply in the Maldives is as thin as it is.

Published 11 August 2026 6 min read Data 2022–2026

MEDIUM CONFIDENCE the direction of the result is robust across every scenario tested, but three of the four principal inputs are asking-price or Nyra-assessed rather than transacted, because the Maldives publishes no transaction register and no construction-cost survey

NYRA INDEPENDENT ANALYSIS

This is not client work. No party commissioned it, no client is named or implied, and the scenario is hypothetical. It is published to show how Nyra applies its public research to a specific commercial decision — the method is real, the project is not.

The decision

A landowner holds a 10,000 sq ft plot in Hulhumalé Phase 2 and can either develop 60 apartments for sale, or not. Does the development clear its cost of capital on evidence available today?

Service line

Feasibility & commercial analysis

Whether a project works on the numbers, under stated assumptions and scenarios.

No-go in the base case · conditional only at the top of the market

On base assumptions the scheme loses MVR 42.7m and the break-even sale price of MVR 4,033/sqft sits in the top decile of the observed asking range. It turns positive only if land is acquired at the auction floor, construction lands at the bottom of the assessed cost band, and every unit sells at the very top of the market — three favourable assumptions at once, none of which is in the developer's control.

Decision confidenceThree readings
Evidence
Low-medium confidence

Land pricing is official (HDC auction floors) and financing is official (MMA lending rates). But sale prices are asking-tier with no transaction register behind them, and construction costs are a Nyra assessment because no official cost survey exists.

Model
Medium-high confidence

The structure is a standard development appraisal with no exotic assumptions, and it is externally validated: the sale prices it uses reproduce the 6–11% gross rental yields independently observed in the same market.

Recommendation
Medium-high confidence

Higher than the evidence alone would support, because the result is not marginal. The base case fails by MVR 833/sqft — roughly a quarter of the sale price — so it survives being wrong about any single input.

Questions this paper set out to answer
  1. Does a market-priced residential development in Greater Malé clear its cost of capital?
  2. Which input is actually binding — land, construction, finance or price?
  3. What would have to be true for the answer to change?

The decision

A landowner holds a 10,000 square foot plot in Hulhumalé Phase 2. They can build and sell apartments, or they can do nothing. The question is not whether housing is needed — the People programme establishes that it is, at roughly 3,270 net new households a year. The question is whether a market-priced development returns its cost of capital.

This analysis answers it from the public record alone, and shows exactly where the public record runs out.

The scheme

A deliberately unremarkable scheme, so the result is about the market rather than about the design.

ParameterAssumptionBasis
Site10,000 sq ftHypothetical, sized to a typical Phase 2 plot
Storeys8 at 80% coverageGross floor area 64,000 sq ft
Efficiency80%Saleable area 51,200 sq ft [OWN-CALC assumption]
Units60 × 850 sq ft, 2-bedConsistent with observed Hulhumalé 2BR sizing
Programme30 months to full sell-down[OWN-CALC assumption]
Capital structure60% loan-to-cost[OWN-CALC assumption]
Debt cost11.29%MMA weighted-average local-currency lending rate, March 2026 [OFFICIAL]
LandMVR 4,250/sq ftMidpoint of the observed HDC Phase 2 auction floor of MVR 4,000–4,500 [OFFICIAL]
ConstructionMVR 2,000/sq ft GFAInside Nyra’s assessed mid-rise band of MVR 1,500–2,300 [OWN-CALC]
Sale priceMVR 3,200/sq ftMid of the observed asking range of MVR 2,400–4,200 [ASKING]

Two of those inputs are official, one is a Nyra assessment, and one is an asking price. That distribution is the honest state of Maldivian development evidence, and it is why the confidence ratings on this analysis differ across evidence, model and recommendation.

What it costs

What a saleable square foot costs before anyone makes a rufiyaa
Base-case development cost per saleable square foot · MVR · hypothetical scheme
Land at the observed Hulhumalé Phase 2 auction floor midpoint of MVR 4,250/sqft [OFFICIAL — HDC plot sales]. Construction at MVR 2,000/sqft of gross floor area, inside Nyra's assessed mid-rise band of MVR 1,500–2,300 [OWN-CALC]. Soft costs at 12% of construction and sales costs at 3% of revenue [OWN-CALC assumptions, stated]. Finance at the MMA weighted-average local-currency lending rate of 11.29%, 60% loan-to-cost, average 50% drawn over 30 months [OFFICIAL rate; structure OWN-CALC]. Gross-to-saleable efficiency 80%.

Per saleable square foot, before any developer profit: construction MVR 2,500, land MVR 830, soft costs MVR 300, finance MVR 307, sales MVR 96. Break-even is MVR 4,033 per saleable square foot [OWN-CALC].

Note what dominates. Construction is 62% of the break-even cost — not land, which most commentary treats as the Maldivian constraint. Land matters, but at Phase 2 auction floors it is roughly a fifth of construction.

What the market pays

The break-even price sits at the top of the market, not in it
Required sale price against observed Hulhumalé asking evidence · MVR per saleable sqft
Observed asking range: Hulhumalé 2BR and 3BR listings, roughly MVR 2,400–4,200/sqft [ASKING — dossier D07]. Break-even and target-return prices: Nyra model [OWN-CALC]. Asking prices are not achieved prices; no transaction register exists, so the observed band is an upper bound on what the market actually clears.

Observed Hulhumalé asking prices run roughly MVR 2,400–4,200 per square foot [ASKING — dossier D07]. The break-even of MVR 4,033 therefore sits in the top decile of what units are advertised at — and asking prices are an upper bound on achieved prices, since no transaction register exists to show what anyone actually paid.

To earn a 20% margin on cost, the scheme needs MVR 4,901 per square foot — above the entire observed asking range.

The scenarios

Four scenarios, one positive
Project profit or loss on a hypothetical 60-unit scheme · MVR m
Nyra model [OWN-CALC]. Upside: land MVR 4,000/sqft, build MVR 1,700, sale MVR 4,000. Base: 4,250 / 2,000 / 3,200. Downside: 4,500 / 2,400 / 2,700. Severe: 4,500 / 2,900 / 2,400 — the severe build cost reflects the 29% input-cost rise construction firms told the MMA they expect for 2026, and parallel-rate sourcing of imported materials. Every scenario assumes full sell-down within 30 months; slower absorption makes all four worse.
ScenarioLandBuildPriceProfitMargin on revenueEquity IRR
Upside4,0001,7004,000+MVR 23.1m+11.3%13.0%
Base4,2502,0003,200−MVR 42.7m−26.0%negative
Downside4,5002,4002,700−MVR 101.3m−73.3%equity impaired
Severe4,5002,9002,400−MVR 155.1m−126.2%loss exceeds equity

All figures MVR per square foot for inputs; Nyra model [OWN-CALC]. Equity commitment in the base case is MVR 74.3m against MVR 111.5m of debt.

One scenario of four is positive, and it requires all three favourable assumptions simultaneously: land at the auction floor, construction at the bottom of the assessed band, and every unit selling at the top of the market. None of the three is within the developer’s control, and the third is a market outcome rather than a decision.

The severe case is not a fantasy stress. Its construction cost reflects what construction firms themselves told the MMA they expect — a 29% rise in non-wage input costs during 2026 [OFFICIAL — QBS Q1-2026] — combined with sourcing imported materials at the parallel exchange rate.

Which input is actually binding

Solving backwards is more informative than the scenarios, because it shows how far from viable the base case is:

  • Break-even land price, holding build at MVR 2,000 and price at MVR 3,200: MVR 316 per square foot [OWN-CALC]. The auction floor is MVR 4,000–4,500. The land would have to be roughly thirteen times cheaper.
  • Break-even construction cost, holding land and price at base: MVR 1,451 per square foot of GFA — below the entire assessed mid-rise band of MVR 1,500–2,300.
  • At top-of-market pricing of MVR 4,200, the scheme supports land at MVR 4,895/sq ft — just above the auction floor. This is the only configuration in which private development pencils at all.
What the market would have to pay at each build cost
Sale price required for a 20% margin on cost, by construction cost · MVR per saleable sqft
Nyra model [OWN-CALC], holding land at MVR 4,250/sqft and the capital structure constant. The construction range spans Nyra's assessed mid-rise band and its parallel-rate-sourcing extension. The observed asking ceiling of MVR 4,200/sqft is marked for comparison — every column exceeds it.

At every construction cost in the assessed band, the price required for a 20% margin exceeds the highest price anything in the market is advertised at.

Why this result matters beyond the scheme

The break-even land price is the finding. If land has to be near-free for mid-market apartments to clear, then the only entities that can build them are those that do not pay for land.

That is precisely what the Maldives does. The Binveriya scheme allocated 1,407 Phase 2 plots free of charge, at an audited cost to HDC of MVR 14.85–15.93bn [OFFICIAL — Auditor General]. Hiyaa units were built at roughly MVR 1.6m each with about 40% subsidy [OFFICIAL — World Bank]. State schemes dominate effective ownership transitions not because of policy preference alone, but because the private arithmetic does not close at market land prices.

The model, in other words, derives the market structure that is actually observed — including the fact recorded in our Hulhumalé dossier that private flats have been slow to arrive while state programmes proceed. That correspondence is the strongest validation available for an analysis with no transaction data behind it.

A second cross-check: the sale prices used here imply gross rental yields of 6.1–11.0% against observed Hulhumalé 2BR rents of MVR 18,000–25,000 a month [LISTING]. Our independent Hulhumalé dossier puts indicative gross yields at 6.2–12.0%. The model reproduces an observed ratio it was not fitted to.

What we would recommend

No-go on a build-to-sell basis at these inputs. Not because the margin is thin, but because it is absent by roughly a quarter of the sale price, and the single configuration that works requires three favourable outcomes at once.

Two alternatives are worth modelling before abandoning the site, and both change the question rather than the answer:

  1. Build to rent rather than to sell. Gross yields of 6–11% on the same asset may clear a hold strategy that a sale strategy cannot, particularly for an owner whose land is already held rather than being acquired at auction. That is a different model with different financing and a different risk profile, and we have not run it here.
  2. Re-test if the land is already owned. An owner who inherited or long-held the plot is not paying MVR 830 per saleable square foot. Removing that line alone does not rescue the base case, but it moves break-even to roughly MVR 3,200 — the base sale price assumption — which puts the scheme at the margin rather than clearly under water.

The general conclusion for the market: at current construction costs, financing costs and observed prices, market-rate residential development in Greater Malé is not obviously viable for a developer who must buy land at auction and borrow at commercial rates. Everything being delivered at scale is either subsidised, on free land, or both.


Sources

  • HDC — Hulhumalé Phase 2 plot sales and auction floors (via Atoll Times, plus.mv)
  • MMA — Monthly Statistics May 2026, Table 8 (weighted-average lending rate)
  • MMA — Quarterly Business Survey Q1-2026 (input-cost expectations)
  • MBS — construction producer price index
  • Nyra — real-estate programme dossiers D07, D09, D12, D13
  • Hulhumalé residential listings, property.mv and Handy Maldives
What would change our mind06
  • Evidence that achieved transaction prices materially exceed asking prices — the reverse of the usual pattern, but unobservable without a register
  • Land acquired well below the auction floor, through inheritance, an existing holding or an off-market transfer, which removes MVR 830/sqft of cost the model charges
  • A contractor tender at or below MVR 1,700/sqft of gross floor area with the exchange-rate risk contractually transferred
  • Access to the 5% state Hiyaavehi facility or equivalent concessional finance in place of 11.29% commercial debt
  • A build-to-rent structure rather than build-to-sell: at these prices gross yields of 6–11% may clear a hold strategy that a sale strategy cannot
  • Resolution of the parallel exchange-rate premium, which would move the whole construction cost band down
What we would need to go furtherCommissioned evidence

This analysis is built entirely from the public record. To convert it from a preliminary assessment into investment-grade due diligence, Nyra would need evidence that only the party making the decision can open:

  • Verified title and the actual acquisition price, rather than the published auction floor
  • Two or three priced construction tenders from named contractors, with the exchange rate each has assumed for imported inputs stated explicitly
  • Committed financing terms — margin, fees, drawdown profile and covenants — rather than the system weighted-average rate
  • An architectural scheme and unit schedule, to replace the assumed 80% gross-to-saleable efficiency with a measured one
  • Recent achieved transaction prices for comparable units, which no public source contains
  • An absorption assessment: how many units of this type actually sell per quarter in this location
  • The developer's own foreign-exchange access, which determines whether the build sits at the bottom or the top of the cost band

Nyra publishes independent market and economic research. This material is general information and commercial analysis only. It is not investment, legal, tax or accounting advice, is not a recommendation to buy or sell any security or asset, and does not take account of any reader's objectives or circumstances. Nyra is not a licensed investment adviser. Figures are sourced and tiered; estimates and scenarios are labelled as such and may change as new data is published.