Which profit level indicator should I use for a distributor?
The one that matches what the entity does and what drives its result. For a reseller the return is usually measured against sales, because sales are what its activity turns over. For a service provider or a manufacturer under low risk, costs are the better denominator, since the activity is the cost base itself. Where the entity is asset-heavy, a return on assets can be more informative. The choice carries more weight than the database you search, so state the reasoning: an indicator picked because it produced a convenient answer is visible in the file as easily as one picked on the functions.
Who should be the tested party in a TNMM analysis?
The less complex of the two — the entity whose functions are routine and which owns no valuable intangibles, because that is the entity for which comparable companies actually exist. Test the party that owns the brand and directs the group and you are looking for comparables to something unique, which is why those searches return companies resembling nothing in the transaction. The selection is a conclusion from the functional analysis, not a preference, and it should be written down with reasons. Choice of tested party and choice of indicator between them decide most of the outcome.
Does it matter which database the comparables came from?
Less than most people expect. Coverage differs between sources and that is worth noting in the file, but the outcome is driven by the decisions you make inside whichever database you use: which entity you tested, which indicator you chose, how wide the initial screen was, and which candidates you rejected on manual review. Two analysts on the same database will produce different ranges if those decisions differ. So record the source, then spend the effort on the search strategy and the rejection reasoning, which is where a reviewer's questions will actually land.
Our operating margin is below the range — what happens now?
First check the analysis rather than the result. Confirm the tested party is the right one, that the indicator suits its functions, and that the comparable set survives a proper manual review — a range built on loose screens can be wrong in either direction. If the analysis holds, the shortfall is real and the file should address it: what happened in the year, which of the causes an independent party in the same position would also have borne, and whether the pricing policy needs to change going forward. An explained shortfall is a far better position than an unexplained one.
Can I use a multi-year average instead of a single year's margin?
Multi-year data is commonly used on both sides of the comparison, because a single year can be distorted by timing — a large contract, a start-up phase, a market disruption. The requirement is consistency: if the comparables are measured over a period, measure the tested party over the same period, and say which years are in it. What does not work is choosing the span after seeing which one puts the result inside the range. Set the period on the facts of the business, state it, and keep to it in later refreshes.
Do both sides of an intercompany transaction have to use the same method?
Nothing makes it automatic, and where the two sides document differently the risk is that both countries tax the same profit. That is the practical case for one analysis the group can stand behind in both places: a single tested party, a single indicator, a stated search, and consistent figures. Where the local requirements differ, the presentation can differ while the underlying analysis does not. A margin test is well suited to this, because operating margins survive the accounting differences between jurisdictions that gross margin comparisons do not.
How would a foreign tax authority know I am resident there?
Mostly from information you or your bank already provided. Account-opening forms ask you to self-certify tax residence, and that certification is reported between authorities under the Common Reporting Standard or, for US accounts, under the FATCA framework. Beyond that: employer and payroll filings, property registries, immigration records and the tax filings of anyone who paid you. The realistic planning assumption is that the data arrives. See FATCA and information reporting.
How many days can I spend in a country before I become tax resident?
It depends on the country, and a day count is only ever the start. Many use a threshold in a tax year, some also look at averages across several years, and some have no day test at all and decide on where your home and life are. Two countries can both conclude you are resident, which is what the treaty tie-breaker exists to settle. Counting days without checking the tie-breaker is how people end up filing as resident nowhere. See the residency tie-breaker.
