New Delhi (ABC Live): The CBDT announced on 31 March 2026 that it signed a record 219 Advance Pricing Agreements in FY 2025–26, taking the total since the scheme began to 1,034, including 750 UAPAs and 284 BAPAs. It also reported 84 BAPAs in one year, the highest ever, with bilateral outcomes involving 13 treaty partners, and first-ever bilateral APAs with France, Ireland, Indonesia, and Sweden. These figures show that India’s APA system is now working at a much larger scale.

This matters because transfer pricing has long been one of the most disputed parts of Indian international tax law. For multinational groups, such disputes not only lead to tax demands. They also create doubt in planning, reporting, business design, and, in cross-border cases, the risk that the same income may be taxed in two countries. India’s push for tax certainty also fits into a wider rewrite of direct-tax law and administration, a point discussed in ABC Live’s analysis of the Income-tax Act, 2025.

Still, the key issue is this: India now appears to be following a three-track model. Safe Harbour deals with standard cases. UAPAs give certainty in India. BAPAs help where two countries must align. That design makes sense. However, these three routes do not remove the same risk. So, taxpayers should not assume that more options mean equal protection. That is the core policy issue in India’s current cross-border transfer pricing system. This is an analytical reading based on the structure described in the official releases.

What UAPA, BAPA and Safe Harbour really mean

UAPA: a domestic certainty tool with cross-border limits

A Unilateral APA is, in substance, an advance agreement between the taxpayer and the Indian tax administration alone. It sets the pricing method or arm’s-length framework for certain international transactions in advance. In practice, that helps a taxpayer reduce future Indian transfer-pricing disputes about margins, comparables, tested-party choice, and key assumptions. CBDT’s latest figures show that UAPAs remain the majority, accounting for 750 of 1,034 total APAs.

The benefit of a UAPA is clear. It gives the taxpayer certainty in India. That can sharply reduce litigation, audit swings, and the repeated reopening of the same pricing issue year after year. For taxpayers whose main concern is Indian transfer-pricing exposure, this is a major benefit. However, the weakness is also clear. A UAPA binds only India. If the foreign tax authority does not accept the same pricing result, the multinational group may still face economic double taxation. In other words, a UAPA may solve the Indian dispute without solving the full international dispute. That is why UAPAs are best seen as domestic shields, not complete cross-border solutions. This conclusion follows from the tool’s unilateral design and the separate role of bilateral APAs.

BAPA: the strongest route for real cross-border certainty

A Bilateral APA is negotiated through treaty-based talks between India and the treaty partner. It seeks joint acceptance of the transfer-pricing treatment by both sides. In cross-border tax practice, this is a major difference. A BAPA is not just about reducing Indian assessment risk. It is about reducing the risk that two countries will tax the same income in different ways. CBDT’s release expressly notes that BAPAs offer protection against potential or actual double taxation.

That is why the rise to 84 BAPAs in FY 2025–26 is arguably more important than the overall 219 APA headline. For multinational groups with large cross-border service, royalty, financing, R&D, manufacturing, or distribution arrangements, a BAPA is usually more valuable than a UAPA because it addresses two-country tax risk rather than only India-side risk. When both jurisdictions agree in advance, the certainty is much stronger than a one-sided domestic arrangement. This inference is supported by the official explanation of the extra double-tax protection that BAPAs provide.

Still, BAPAs are not easy. They depend on treaty support, capable tax authorities, a cooperative treaty partner, and a taxpayer that can handle a longer and more detailed process. So while a BAPA is usually the strongest tool, it is also the most demanding. In addition, the official disclosures focus on signings, not on average disposal time, pending cases, renewal ratios, or failed bilateral cases. So, India’s bilateral progress is real, but the public record still does not fully show whether the system has become faster overall or simply more productive at the point of closure. That is a fair critical reading of what the official releases disclose and what they leave out.

Safe Harbour: a faster route to certainty, but through standard rules

India’s Safe Harbour Rules work differently from APAs. The CBDT release describes Safe Harbour as a faster, lower-cost alternative that uses fixed margins for certain categories of international transactions. The same release says the regime covers twelve categories, including IT and software services, IT-enabled services, KPO, contract R&D, intra-group financing, guarantees, auto components, low value-adding services, and some diamond-sector transactions.

The Finance Act 2026 changes make Safe Harbour much more important. Official budget communication says that several technology-service segments were merged into one “Information Technology Services” category with a common 15.5% margin, and that the threshold for using the safe harbour for IT services was raised from ₹300 crore to ₹2,000 crore. The same official material says safe-harbour approval for IT services will move to an automated, rule-driven process and can continue for five years at the company’s option.

This expansion is business-friendly. It may reduce compliance cost, officer contact, and case-by-case review for a large part of India’s technology-services ecosystem. Yet Safe Harbour remains a blunt tool. It gives certainty by using fixed margins and standard rules, which means it can lose case-specific accuracy. A standard margin may be easy for administration, but it may not reflect the true economics of every captive service provider or mid-sized technology structure. It also remains mostly one-sided: Indian acceptance does not automatically force matching foreign acceptance. So, Safe Harbour can reduce disputes, but it cannot always prevent cross-border mismatch. That point is an inference from the domestic, rule-based design of Safe Harbour and the bilateral design of BAPAs.

Comparison table: UAPA vs BAPA vs Safe Harbour

This table reflects the official features of the three routes and the practical difference between unilateral, bilateral, and rule-based certainty.

Why the present regime is a policy gain, but not a complete answer

India has improved certainty, but not all certainty is equal

There is no doubt that India’s transfer-pricing certainty system has matured. The latest official release shows both scale and wider bilateral reach, especially through the 13 treaty-partner outcomes and first-time bilateral APAs with France, Ireland, Indonesia, and Sweden. That suggests stronger administrative capacity and deeper treaty engagement than before.

But the deeper policy issue is that the three tools do not solve the same problem.

• UAPA solves Indian-side uncertainty.
• BAPA deals more directly with bilateral double-tax risk.
• Safe Harbour solves compliance burden through standard rules.

That means the regime is more advanced than before, but also more segmented. A taxpayer is not choosing between equal options. A taxpayer is choosing which kind of risk to remove and which kind of risk to keep. That is the real decision framework created by the present Indian regime. This is an analytical conclusion drawn from the distinct roles set out in the official material.

The output numbers are strong, but the missing metrics still matter

Officially, the story is one of success. However, the releases remain heavily focused on signing volume. They do not tell the public enough about pendency, median closure time, sector mix, rollback use, taxpayer drop-off, or post-agreement compliance burden. As a result, the government can credibly claim scale, but outside observers still cannot fully judge system efficiency or taxpayer experience. For a regime meant to improve ease of doing business, that lack of transparency remains an important weakness. This criticism arises from the gap between what the official releases highlight and what they leave unreported.

Safe Harbour reform is welcome, but standardisation has limits

The 2026 Safe Harbour expansion is a major pro-business step, especially for technology services. It broadens eligibility, sets a common 15.5% margin for the combined IT services category, raises the threshold to ₹2,000 crore, and promises an automated approval path. Still, this route works best when taxpayers are willing to trade some factual precision for administrative speed. That may be a sensible trade for routine captive operations. However, it is not always the right trade for complex cross-border structures, sensitive business models, or transactions where foreign tax symmetry is critical. So, the reform is important, but it should not be treated as a substitute for strong bilateral certainty.

Final assessment

India’s cross-border transfer-pricing system is clearly stronger today than it was a few years ago. Record APA signings, more bilateral outcomes, and a wider Safe Harbour framework all point in the same direction: India is trying to move away from long transfer-pricing disputes and toward advance tax certainty.

However, the real strength of the system does not lie in simply having more tools. It lies in knowing what each tool can and cannot do.

UAPA is useful for certainty in India.
BAPA is the strongest option for serious cross-border protection.
Safe Harbour is the quickest option for standard compliance relief.

So, the real question is not which route looks easiest at first glance. The real question is which risk the taxpayer wants to remove first: Indian tax dispute risk, double-tax risk across two countries, or compliance burden. Once that answer is clear, the choice becomes easier. For routine cases, Safe Harbour may be enough. For India-side certainty, UAPA works well. But for large and important cross-border transactions, BAPA remains the strongest option. This ranking is an analytical conclusion based on the official design and purpose of the three mechanisms.

How We Verified

We verified this article against the latest official government and tax-administration material:

• CBDT’s 31 March 2026 PIB release confirming 219 APAs in FY 2025–26, 1,034 cumulative APAs, 84 BAPAs, the list of treaty partners, and the Safe Harbour summary.
• Official budget and PIB material from 1 February 2026 confirming the Safe Harbour changes for Information Technology Services, including the 15.5% margin, the threshold increase from ₹300 crore to ₹2,000 crore, the automated approval process, and the five-year option.
• The Income Tax Department’s transfer-pricing page for the official framework context.
• ABC Live’s contextual analysis of the wider direct-tax shift in its Income-tax Act, 2025 piece, used here as an internal reference for the broader tax-simplification debate.

The interpretive conclusions in this article, especially the distinction between domestic certainty, bilateral protection, and rule-based compliance ease, are analytical judgments based on those official facts.