Notes

The one-month line in Indonesia's accommodation tax that nobody writes down

Indonesia’s regional tax on certain goods and services — Pajak Barang dan Jasa Tertentu, PBJT — covers accommodation services. Pasal 58 ayat (1) of UU No. 1 Tahun 2022 caps the rate at 10% and leaves each regency and city to set the rate in force by local regulation. In Badung, Gianyar and Denpasar, between them effectively the whole Bali villa market, it sits at the cap.

The scope of “hotel services” in that law is deliberately wide. It covers hotels, hostels, villas, guest houses, glamping and, in terms, private residences functioning as hotels.

Then the explanatory memorandum to Pasal 54 ayat (1) adds a line that almost nobody in this market repeats: where such a residence is let for more than one month, it is not an object of the tax.

What that actually means

Nothing about the building changes. The owner does not change. The classification the villa holds does not change. The furniture does not change.

The length of the booking changes, and one of the two Indonesian charges on the letting stops applying.

That is not a loophole, in the sense that it is not an unintended gap that somebody found. It is a policy line drawn in published law between accommodation and residential letting, and it is drawn in a fairly ordinary place — most jurisdictions draw one somewhere.

What it does not change

The other Indonesian charge is unaffected. PP No. 34 Tahun 2017 charges final income tax at 10% on income from letting land and buildings, on the gross rental amount, and it is final: not reduced by expenses, not recomputed at year end. That applies to a twelve-month lease exactly as it applies to a three-night stay.

So the one-month line moves one charge, not the tax position as a whole. Anyone presenting it as a way to let a villa tax-free is either confused or selling something.

Why it matters more than one deduction

On its own, removing a 10% charge is worth 10%. What makes the line more interesting than that is what else tends to travel with a stay of over a month.

A long let is usually not booked through a platform, so the platform commission goes too. Housekeeping and laundry drop to a fraction of what nightly turnover requires. Utilities often become the tenant’s. The management job shrinks from running a calendar to holding a deposit and handling maintenance, which should mean a smaller fee, and often does not unless somebody asks.

Put those together and the gap between gross and net narrows far more than the gross revenue falls. That is the arithmetic worth running, and it is the reason the comparison everyone makes — nightly gross against long-let gross — is the wrong comparison.

Where this is a live question and where it is not

In Seminyak, where three to five nights is the shape of the market, it is academic. In Ubud, where wellness and retreat stays run to weeks, and in Sanur, where a health-tourism zone established by government regulation produces guests on treatment cycles, it is not academic at all. Canggu sits in between, with a substantial month-plus remote-working population alongside the holiday market.

The honest caveat

This post tells you where the line is written and what it says. It does not tell you how it applies to your villa, because that depends on how the villa is held, what it is registered as, which regency it sits in, and what the arrangement with the occupant actually is in substance rather than in name.

Those are questions for an Indonesian adviser licensed to answer them, with your documents in front of them. What this post gives you is a specific question to ask instead of an open one, which is usually the difference between a useful hour of advice and an expensive one.

All posts · Yield estimator · Fees, line by line