Rental deal in Osaka: ¥81,860,000 purchase, ¥249,100/month — sanity check?

[SteadyPebble]

The headline return is already fairly narrow, and tenant turnover is my main concern. The property is a 5-bed duplex in Osaka priced at ¥81,860,000, with projected combined rent of ¥249,100 a month, giving about 3.7% gross before costs.

I have modelled only eleven paid months and allowed for management, routine maintenance, vacancy and one larger repair. Property tax and insurance still need firmer figures, and the repair reserve may be too optimistic.

If the actual tax, insurance and management costs leave a reasonable margin, I could keep investigating. If one turnover or significant repair wipes out most of the annual return, I would pass. What figures would you verify first?

 

[SimpleLane]

Eleven months gives ¥2,740,100 before any operating costs, only about 3.35% of the purchase price. From there, property tax, insurance, management and repairs leave very little room. I would not rely on an estimated tax figure: ask for the actual recent property-tax bill and confirm exactly what land and buildings are included.

 

[emery_repairs]

How old is the building, and is ¥249,100 the combined rent for both units or an assumption based on full occupancy? With a duplex, one empty unit can remove a large share of income at once. I’d model tenant turnover separately from routine vacancy, including the period between tenancies and any work needed before reletting.

 

[porch.easy]

I partly disagree that a particular net-yield target can settle this. Financing sensitivity may dominate when the gross yield starts at 3.7%. Run the deal unlevered first, then test the loan separately at a higher interest cost and with no rent growth. Also split the repair reserve by likely timing: annual minor work is different from one major building expense arriving early.

 

[sailsAndQuill]

Before deciding, request the tax history, insurance quote, management proposal, current rent evidence and a schedule of past major repairs. Then run three cases: full expected rent, eleven months’ rent, and one unit vacant for an extended period. If the deal only works in the first case, the margin is probably too thin regardless of the headline yield.