NovA high return built on one fragile assumption deserves a second look
How to stress-test an attractive property return by identifying the assumption that carries the model and comparing it with a sturdier base case.
This article reflects the named expert’s practical perspective. See NovAsia’s editorial policy for how material is prepared and reviewed.
A high projected return is not a warning sign by itself. Nor is a modest return automatically credible. The more useful question is how much of the conclusion depends on one assumption behaving perfectly.
A property model can contain twenty rows and still be fragile if one row does nearly all the work. Perhaps the unit must be rented for twelve months every year. Perhaps the asking rent has to become the achieved rent. Perhaps a resale price is expected to rise by a fixed percentage before completion. Perhaps the apartment is assumed to become income-producing immediately at handover, with no furnishing period or leasing gap. The spreadsheet may look detailed while the investment case is effectively a single bet.
Find the line that can overturn the whole result
Take a hypothetical USD 100,000 apartment. A marketing model assumes USD 1,000 monthly rent for twelve months, producing USD 12,000 of gross annual rent. Dividing that by the advertised purchase price gives a visually satisfying 12%. Yet the number has not answered several basic questions. Is USD 1,000 supported by comparable achieved leases or is it an asking figure? Is a tenant assumed from day one? Does the buyer need another USD 8,000 for furnishing? Which recurring expenses are paid by the owner? Is management included? What happens between tenants?
I do not lower those assumptions automatically. I isolate them. If the rent is well evidenced for the exact unit type, that strengthens the case. If occupancy history is available and comparable, use it. The discipline is simply to make the evidence travel with the variable.
Change one assumption at a time
The fastest stress test is often the simplest. Keep everything else constant and move the most important variable.
Suppose the apartment is rented for nine months instead of twelve. Gross rent falls from USD 12,000 to USD 9,000 before any expenses. Suppose the achieved rent is USD 850 rather than USD 1,000. Nine occupied months then produce USD 7,650. Now add a hypothetical USD 8,000 fit-out that was excluded from the original capital base. None of these alternatives predicts what will happen. They show how dependent the headline result is on the original scenario.
A model that survives ordinary deviations is more useful than one that produces an impressive number only when every assumption is favourable. “Survives” does not mean remains profitable under every imaginable shock. It means the buyer can see a range of outcomes without the investment thesis changing identity after one normal inconvenience.
A boring base case can be more decision-useful
I like base cases that leave room for reality. A nine-month rental assumption, a documented cost allowance and an explicit leasing gap may produce a lower number, but it tells the owner what capital has to do without relying on perfect execution.
The upside case can still exist. If the apartment rents faster, achieves a higher rate or requires less fit-out, the result improves. The important change is psychological as much as mathematical: the upside becomes additional performance rather than the minimum condition needed for the purchase to make sense.
This also improves the conversation with a seller or developer. Instead of arguing about whether their 10% or 12% number is “realistic,” the buyer can ask which variable creates it and what evidence supports that variable. The discussion becomes testable.
Return percentages need labelled numerators and denominators
Another source of fragility is the definition of return itself. USD 10,000 of gross rent divided by a USD 100,000 purchase price is not the same measure as USD 10,000 of net cash after expenses divided by USD 115,000 of total capital deployed. Both calculations can be mathematically correct and economically different.
I prefer to write the fraction in words before trusting the percentage: gross annual rent divided by advertised purchase price; net operating cash after listed expenses divided by total invested capital; realised sale proceeds after transaction costs divided by equity invested. Once the components are named, an attractive figure often becomes much easier to understand.
A model should also avoid mixing realised and hypothetical values. An existing lease is evidence of a current contractual cash flow, subject to its terms. An advertised rent is evidence of an asking position. A future resale price is a scenario. They do not become equally certain because they occupy adjacent cells.
The break point is often more useful than the expected return
A buyer usually learns more from the point at which the decision stops working. How low can rent fall before the owner must add cash? How many vacant months can the annual budget tolerate? What fit-out overrun would exceed the available reserve? How long can a resale take before the capital is needed elsewhere?
Those questions turn the spreadsheet into a decision tool. They are not designed to frighten the buyer. They define the relationship between the asset and the buyer’s own constraints.
Let the strong case prove itself after the base case survives
There are legitimate reasons why one property may outperform a conservative model. A specific tenant contract, an unusually useful layout or a documented operating arrangement can change the economics. I have no problem with an upside case that is built from evidence.
What I distrust is a high number whose only defence is that “the market should support it.” If one assumption carries the entire purchase, that assumption deserves the most evidence in the file. If it cannot be evidenced, the model should show what happens without it.
A less exciting calculation can therefore be the stronger investment argument. It tells the buyer that the property does not need a perfect year to remain recognisable as the asset they intended to own. High return can be a result. It should not be the single fragile premise holding the whole decision together.