Your TP policy. Documented.
Defensible. Done.

If value moves between related entities — your parent funding a subsidiary, a sub providing development or support, IP licensed across borders, even routine intercompany cost recharges — then yes, you have intercompany transactions that need an arm’s length basis and an agreement. Two entities is enough to create a TP obligation. The clearest exception is a genuine buy/sell flow with no embedded services or IP, which has limited TP implications initially. We’ll confirm which category your actual flows fall into rather than assume, since cost recharges and informal "funding" between entities almost always carry a TP question even when founders don’t realize it.

The right agreement follows what’s actually moving between the entities. A services agreement covers one entity providing development, support, or back-office work to another, typically on cost-plus (markup applied to the cost base — salaries, rent, operational costs). A distribution agreement covers reselling, where one entity keeps an arm’s-length operator margin and remits the rest. A royalty/licensing agreement covers IP. Many setups need more than one — for example, an Indian dev entity on a services agreement and a royalty arrangement on top, depending on how revenue flows. We map your actual transactions first, then put the right contracts around them rather than starting from a template.

The common pattern is the Indian entity providing development or support to the US entity on a cost-plus services basis, with an intercompany services agreement and benchmarking that sets the markup against comparable Indian service providers. The US side documents the markup as arm’s length; the Indian side needs documentation that satisfies Indian requirements as well, with India tying documentation expectations to transaction value. Indian documentation supports an annual filing cycle that runs separately from US deadlines. We handle both ends together so the position is consistent across the corridor rather than two providers telling slightly different stories.

The markup should reflect what an independent provider doing the same functions, with the same risks, would charge — established through benchmarking against comparable companies in recognized databases, with the tested party positioned within the resulting arm’s length range. We set it from data rather than picking a round number that "feels right," and document why it fits given the functional analysis (who does what, who owns what, who bears which risks). That data-backed basis is what makes the markup defensible at audit — an unsupported markup, however reasonable, is the weakest position in any TP inquiry.

The exposure is meaningful and compounds the longer it runs — undocumented or mispriced intercompany flows can lead to adjustments, double taxation, and penalties in any country examining them, and historical losses or profits in the wrong entity can be hard to unwind. The good news is the position can largely be fixed going forward: we build current documentation and benchmarking, put the right agreements in place, and establish a defensible go-forward basis. Prior years can’t be rewritten, but assessing the historical exposure lets you make informed decisions about whether and where to disclose proactively. Getting the next filing cycle right is the cleanest way to stop the bleed.