Exit Planning in India: The Downside-Risk Guide for US Companies (2026)

Figure: The operational detail behind the same decision  what each structure asks of you at every phase.

What the difference actually buys you

The employee obligation does not disappear with an EOR. What changes is the corporate workstream attached to the employment relationship.

• You avoid a liquidation process whose historical average time to final report was 403 days. No liquidation, no dissolution order, no 489-case queue waiting on the tribunal.
• You avoid the statutory two-year non-operation condition for the strike-off route. Section 248(2) requires that the company has not carried on business for two preceding financial years. That requirement simply does not arise if there is no company.
• You skip the compliance tail. At ₹3–8 lakh a year, eliminating 18 months of filings on a shell entity is a $5,000–$13,000 saving on its own  for a business that has already decided it doesn’t want to be in India.
• You may reduce one category of fixed-place PE exposure. Under Article 5 of the US–India Double Taxation Avoidance Agreement, a fixed place of business through which an enterprise is carried on can create a taxable presence. A wound-down subsidiary that had an office and staff gives a tax officer something to assess. An EOR arrangement changes the employment and premises position  but it does not determine permanent establishment by itself. The US parent still needs to assess where its business is actually conducted and what India-based personnel actually do, including any dependent-agent risk.

The difference in workstreams you have to close

An EOR does not eliminate the employee exit. It can eliminate an entire second layer of corporate wind-down from the US company’s exit project.

The honest caveat

An EOR narrows exit risk. It does not eliminate India risk, and it does not cover every situation. See “When is an EOR the wrong answer?” below  that section is deliberately unflattering, because a partner who won’t tell you when their product doesn’t fit is not a partner.

CTA  comparison moment. Not sure which column you are in? Book a 30-minute India entry-and-exit risk review. No pitch  including the cases where the answer is an entity.

At what headcount does an Indian entity become cheaper than an EOR?

Short answer: there isn’t one. The break-even moves with headcount, duration, business activity and how reversible you need the structure to be  and the cheaper structure is not the one with the lower monthly fee, it is the one whose total obligations match what you actually need to do in India. Model cost to enter, cost to operate and cost to exit separately.

It is tempting to ask, “At what employee count does an Indian subsidiary become cheaper than an EOR?” That is the wrong first question. The real comparison is:

EOR cost = EOR fees × employees × duration

Entity cost = setup + recurring compliance + payroll + accounting + audit + tax + governance + management time + eventual exit exposure

The break-even point changes with headcount, duration, business activity and how reversible the company needs the structure to be. A 30-person team that only needs employment infrastructure produces a different answer from a 10-person team that needs to contract with Indian customers, hold licences, own local assets or operate as a permanent local business.

Most India-entry comparisons focus on the cost of getting started. Model three separate costs instead.

The comparison should be made across the entire expected lifecycle, not by comparing an EOR’s monthly fee with the incorporation cost of a subsidiary. A structure that looks cheaper to enter can become more expensive to operate, manage or exit.

That also reframes what an EOR fee is. With your own entity you separately incur incorporation and corporate setup; ongoing accounting, tax and statutory compliance; payroll and employment administration; professional advisers; corporate governance and filings; and eventual wind-down and closure. With an EOR you pay for an employment infrastructure that already exists and is already operated. So the right question is not “Is the EOR fee higher than incorporation?” It is “What does that fee replace, and what obligations remain outside it?”

The cheaper structure is not necessarily the one with the lower monthly fee. It is the one whose total obligations make sense for what you actually need to do in India.

When is an EOR the right answer, the expensive one, and the wrong one?

When an EOR is the safer bet

An EOR is worth evaluating when the India strategy is still being tested and the company primarily needs employment infrastructure rather than a full Indian operating entity. The case is stronger when India hiring is the primary requirement; the expected long-term size of the operation is still uncertain; the company does not need to contract or invoice Indian customers through its own entity; no substantial local assets or regulated licences are required; and the company wants the ability to scale down without creating a separate corporate closure project.

In that situation, the EOR fee is not simply a payroll expense. It is the price of keeping the corporate structure reversible while you test the India thesis.

When an EOR becomes the expensive choice

The same logic works in reverse. An EOR becomes the less suitable structure when India has become a permanent operating centre; the company needs to enter commercial contracts or invoice locally; Indian licences or registrations are required in the company’s own name; the operation needs to own material local assets; the workforce is large and stable enough that a direct operating structure makes commercial sense; or the company needs direct control over the Indian entity and its operations.

At that point the question changes from “How do we avoid setting up an entity?” to “Why are we continuing to pay for a structure designed for reversibility?”

What an EOR changes, and what it doesn’t

An EOR does not make employee exit obligations disappear. Notice or notice pay, applicable retrenchment obligations, gratuity, leave settlement and other contractual or statutory dues still have to be addressed. What changes is the corporate layer.

An EOR is not a way to make employee obligations disappear. It is a way to avoid creating a separate corporate entity you may later have to dismantle.

When an EOR is the wrong answer

An honest exit strategy India conversation has to include the cases where the flexible model doesn’t apply. If the business needs an Indian legal entity to conduct its core activities, hold assets, obtain licences, sign regulated contracts or build a substantial permanent operation, evaluate an entity instead. There are four such situations.

1. India is becoming a revenue-generating operation. If the operation will contract with customers, invoice revenue or collect local payments, an EOR is not a substitute for an entity. Revenue-generating activity in India needs an entity, GST registration and a corporate tax presence. Husys’s own position is explicit: when a company is directly selling or directly generating income in India, an EOR is not the appropriate model. Trying to use one anyway is how you create the permanent establishment problem you were trying to avoid. Example: a US SaaS company hires 15 engineers through an EOR, then two years later begins selling directly to Indian customers and wants an India-based commercial organisation. The original structure is no longer the right operating model, and continuing with it simply because the employment arrangement works is the mistake.
2. The business needs an Indian licence, registration or regulated presence. Some activities require the business itself  not merely its employees  to hold a licence, approval or authorisation, and most public-sector procurement requires a locally incorporated bidder. Example: putting employees on an EOR does not create the eligibility a US company needs to bid on an Indian government tender. The question is not “Can an EOR employ these people?” but “Can the business legally perform the activity it wants to perform through the proposed structure?”
3. India is becoming a substantial, permanent operating centre. An EOR is practical for an initial team or a defined expansion phase. If you are building a 500-person captive with a ten-year horizon, both the per-head economics and the strategic case for owning your entity change. The right sequence is often EOR first, entity later, once the thesis is proven  not entity first on the assumption that it will be. Example: a company that starts with 20 engineers and grows to 200 across engineering, finance, HR and operations, with dedicated facilities and India-specific leadership, is no longer asking how to employ a team. It is asking what structure supports the business it has built. There is no universal headcount threshold here: the assessment should weigh activities, control, assets, contracts, regulatory requirements, commercial model and expected permanence.
4. The India operation needs to own or hold assets locally. Real estate, plant, locally owned IP registrations and capitalised infrastructure all need an owner. An EOR employs people; it does not hold your balance sheet.

The practical test. Before choosing an EOR, ask: “What does the India operation need to own, sign, licence, sell, regulate or legally perform in its own name?” If the answer requires an Indian legal entity, an EOR should not be presented as a substitute. An EOR solves an employment and workforce administration problem. It does not automatically solve a corporate, commercial, licensing or regulatory presence problem.

If you are in one of those four situations, the correct advice is to set up properly and plan the exit deliberately  which is a different article, and one we would rather write honestly than sell around. For any US company, the right structure can change as the India operation changes: EOR, subsidiary and other structures are operating choices to be reassessed against the business model, not permanent labels attached to the India strategy on day one.

What does the US parent actually need India to do?

Before deciding on an Indian subsidiary, separate what the US parent needs to accomplish from what it needs to own locally. For some companies the requirement is “We need to employ people in India.” For others it is “We need an Indian business that can contract, invoice, hold assets, obtain licences and operate commercially.” Those are fundamentally different requirements.

Don’t establish an Indian entity simply because you need Indian employees. Establish one when the business activities you need to perform in India require the infrastructure of an Indian entity.

What does it cost if your India thesis is wrong?

A US company entering India is making a bet. The plan may be to hire 10 people, then 50, then build a larger operation. But the structure should also be tested against the possibility that the plan does not work.

The Cost-of-Being-Wrong Test. Before committing to an Indian entity, ask: if our India plan is wrong 12 months from now, which structure leaves us with the smaller problem? Test it against the five scenarios that actually happen.

The point is not that an EOR makes every scenario effortless. It is that the structure determines how much of the exit remains with the US company after the workforce decision has been made. So the important question is not only “What will India cost us if the plan succeeds?” but “What will India cost us if the plan changes?”

Then ask the version of the question that cuts through everything: what still exists after the last employee leaves? If the answer includes an Indian entity, registrations, contracts, assets, liabilities or continuing statutory obligations, those items belong in the entry decision  not just the exit plan.

This does not mean an EOR is automatically the better structure. It means comparing the downside of each, not only the upside. The cheapest way to enter India is not always the cheapest way to discover that your original India plan was wrong.

Which India exit costs do entry models leave out?

Short answer: two, and neither appears on an adviser’s invoice  the management time to coordinate a wind-down across six or seven workstreams in another time zone, and the institutional memory to know what has to be coordinated at all. Both are real costs, and both are borne by the same small group of people who are also running the business.

Two categories reliably fall out of the model: the management time to coordinate an exit, and the institutional memory required to know what has to be coordinated in the first place.

Management time. With your own Indian entity, a wind-down requires coordination across US and India finance, HR and payroll, legal, tax and accounting advisers, a company secretary or corporate adviser, and banking and other local service providers. Each workstream is manageable on its own. The cost appears when someone in the US has to coordinate all of them, across a time zone, until the entity is actually closed. The real cost of an exit is not just what you pay advisers  it is the management time required to make all the advisers, filings and internal teams move together.

Institutional memory. If the person who set up and manages your India operation leaves, can someone else explain which employees are covered by which employment arrangements, which registrations and advisers are involved, what obligations remain with the Indian entity, what contracts, assets and liabilities need resolving, and what has to happen if the company scales down or exits? With your own entity, that knowledge sits scattered across internal teams and external advisers  which is why reconstructing it later is a project in itself: see how one company reassembled its India employment records into a diligence-ready package under a fundraise deadline. A structure is more resilient when the India strategy does not depend on one person remembering how everything works.

Who ends up owning which obligations?

The practical difference between an Indian entity and an EOR is not who runs payroll. It is who becomes responsible for the infrastructure created around the workforce, and who has to dismantle it.

The right question is not “Which structure is easier to set up?” It is “Which obligations are we prepared to own if our India strategy changes?”

Your exit should be executable, not theoretical

Before entering India, you should be able to answer five questions without starting a new research project:

• Who is the legal employer of each India employee?
• What employee obligations arise if we reduce or terminate the workforce?
• What corporate obligations remain after the last employee leaves?
• Who is responsible for completing each part of the wind-down?
• What will we still be paying until the structure is fully closed?

If those answers depend on finding the right person, reconstructing old decisions or discovering obligations only when the exit begins, the structure was never exit-ready. Exit readiness is not a document you create when you decide to leave. It is a property of the structure you choose when you enter.

Is the exit finished when the last employee leaves?

Short answer: no. An India exit runs on two separate clocks, and the corporate one keeps running after the employee one stops. With your own entity, the legal entity, tax and accounting obligations, statutory filings, registrations, bank accounts, assets, liabilities and contracts all survive the workforce. With an EOR there is no client-owned subsidiary sitting behind the team, so the sequence ends with contractual offboarding.

One of the easiest mistakes in exit planning in India is to treat the last employee’s departure as the end of the operation. It isn’t. An India exit runs on two separate clocks:

Employee clock: notice / termination → final settlement → employee transition

Corporate clock: tax and accounting closure → statutory filings → registration closure → asset and liability resolution → corporate closure

With an Indian subsidiary, both clocks run together  and the corporate clock continues long after the employee clock has stopped. What is still alive after your last India employee leaves may include the legal entity, tax and accounting obligations, statutory filings, registrations, bank accounts, assets and liabilities, and contracts.

With an EOR, employee obligations still have to be completed correctly and contractual obligations settled, but there is no client-owned Indian subsidiary sitting behind the workforce. The sequence is narrower: employee transition → applicable settlements → EOR offboarding → end of the relationship. And the commercial exit should be as transparent as the operational one: Husys does not charge a separate termination fee; where the $99 offboarding charge applies, it is a straightforward fee covering the offboarding process. See pricing.

That distinction matters most when India is a talent or delivery location rather than a separate commercial business. If the workforce is the reason you entered India, don’t create a corporate structure that survives the workforce unless the business actually needs it to.

When comparing structures, don’t ask only “How long does it take to exit our employees?” Ask “How long until we have no India structure left to manage?”

→ Read next: what this looks like when it works: a client scaled a team up and back down again without a termination-fee event or a corporate wind-down. Scaling a team up and down without termination fees.

How should the India structure change across the lifecycle?

Short answer: it should be reassessed at each stage rather than chosen once. Obligations differ at enter, hire, scale, reduce and exit  and the structure you pick at entry follows you through all five. Define the commercial, regulatory, scale, operational and control triggers that would make you change structures before you enter, not when you have outgrown the one you have.

A US company should evaluate its India structure across the full lifecycle: ENTER → HIRE → SCALE → REDUCE → EXIT. The obligations change at each stage.

The structure you choose at entry follows you through every later stage. Don’t optimise only for the first hire.

Decide the trigger before you need to change structures

Choosing an India structure is not a one-time decision. The right structure at 10 employees may not be the right one at 100, and an India strategy typically moves through Test → Validate → Scale → Operate. So the question is not “EOR or subsidiary forever?” but “What structure fits the stage we are at today, and what would make us reconsider it?”

Define those triggers before you enter:

• Commercial trigger: India requires local customer contracts or invoicing, or becomes a material revenue market.
• Regulatory trigger: the business needs a licence or registration that requires its own Indian entity.
• Scale trigger: the workforce or operating complexity changes the economics.
• Operational trigger: India becomes a permanent operating centre rather than a hiring market.
• Control trigger: the business needs direct ownership of local assets, contracts or operations.

A trigger does not mean “incorporate now.” It means the structure should be reassessed against what the India operation now needs. A good structure is not one you never change. It is one that gives you a clear reason, and a clear point, at which to change it.

Does the India entry decision change with company stage?

Short answer: yes, but not because of headcount. What changes is what the India operation has to be able to do. A five-person regulated operation may need an entity; a 30-person engineering team may not. Stage, activity and expected permanence matter alongside size, and there is no employee count at which a subsidiary automatically becomes the right answer.

A 30-person startup hiring its first five people in India should not approach India the same way as a company building a 300-person captive centre. There is no universal employee count at which an Indian subsidiary becomes the right answer  stage, activity and expected permanence matter alongside headcount.

A five-person regulated operation may need an entity. A 30-person engineering team may not. A 100-person permanent India centre may have a strong reason to own its infrastructure. Headcount is a signal, not the decision.

Seed and Series A: your first five India hires

This is where most US startups decide too early. The first question is not “Should we incorporate an Indian subsidiary?” but “Have we proven that we need a permanent Indian operating structure?”

Take a company with 5–15 total employees, its first 1–5 India hires  say three engineers, a product designer and a QA engineer  no Indian customers, no need to invoice locally, no licences or substantial assets. The immediate requirement is employment infrastructure, not a second corporate headquarters. Before incorporating, answer: are these five hires the start of a permanent India centre or an experiment? What happens if you hire five rather than fifty, or double the team in 12 months, or funding changes and hiring stops? What remains to close if you exit? And who inside the US company will own Indian compliance and corporate administration?

Your first five India hires should not automatically determine the corporate structure for your next fifty.

Series B: when the India question changes

By Series B the decision is no longer whether you can hire five people. It is whether India is becoming a durable operating capability  20–40 employees, dedicated engineering or product leadership, India-specific management, continued hiring plans, long-term office requirements, local vendors.

At this point the question is not “Is an EOR still possible?” It is “What does the business now need its India operation to own, control or legally perform?” If the answer remains employing and managing talent, an EOR may still fit  the conclusion reached by a Series B SaaS company that kept its 12-engineer India team on EOR rather than incorporating. If it includes commercial contracting, regulated activity, local assets or a permanent corporate presence, the case for an entity strengthens. The trigger should be the business requirement  not the headcount.

The 50–200 employee SMB: management bandwidth is the real constraint

A US SMB may have a CFO, an HR lead and outside counsel  but not a dedicated India finance, HR, tax and corporate team. With 5–20 India employees supporting the US business, evaluate five things separately: what the India team must actually do; who will own payroll, compliance, employment issues, local advisers and corporate administration; whether you are building a long-term centre or testing; what obligations remain if you cut the team by 50%; and whether the operation needs to contract, invoice, hold assets or obtain licences in its own name.

The answer may be an EOR  but reach it because of your operating requirements, not because “SMBs should use EOR.” For a lean company, the hidden cost of an India subsidiary is the management infrastructure needed to run it properly.

→ Read next: the mistakes at this stage are predictable and mostly avoidable. India Expansion Checklist for US Companies: mistakes to avoid.

What does an India exit look like from the chair you sit in?

Short answer: different in every seat. The founder’s constraint is attention, the CFO’s is forecastability, the General Counsel’s is what a wind-down reveals, and the CTO’s is that the reason they chose a hiring geography can be invalidated inside a fiscal year. Four scenarios below.

Abstract frameworks don’t get budget approved. Here is what this looks like from four specific chairs.

The YC-backed founder, 14 people, pre-Series A

You are: a founder out of a Y Combinator batch who raised a $3M seed and needs three engineers in Bengaluru because the same three people in San Francisco cost your entire runway. Your board wants global talent. Your lawyer quoted $8,000 to incorporate an Indian subsidiary and mentioned something about a resident director.

How it normally goes. You incorporate. You discover Section 149(3) of the Companies Act, 2013 requires at least one director who stays in India for 182 days or more during the financial year  so you appoint a nominee or promote your first hire into a statutory office, adding a permanent governance requirement. You register for PF, ESI and Professional Tax. You appoint an auditor. You start filing.

Eighteen months later the product pivots and the India team isn’t the right team any more. You now have three employees with statutory obligations to settle, a director who has to formally resign, an audit obligation, and a corporate wind-down whose available route depends on the company’s statutory eligibility  including a two-year dormancy condition before the fastest route is even open to you. Your seed-stage CFO-by-necessity is now managing an Indian corporate wind-down.

How it goes with an EOR. Three engineers onboarded as employees of the EOR’s Indian entity  Husys benchmarks onboarding at 8 working hours once documents are in. No Indian directorship, no audit, no ROC calendar, no PF code in your name. If the pivot happens, you serve contractual notice, fund the statutory settlements, and you are out inside the notice period. Your exit is a line item, not a project. The founder’s actual scarce resource is attention, and the entity path creates an ongoing compliance-coordination burden that does not disappear simply because the team is small.

The CFO of a Series B drone-tech company, 90 people

You are: VP Finance or CFO at a hardware-plus-software company that just raised $22M. You need an India team for firmware, ground-control software and 24/7 flight-operations support. The board wants a three-year cost model you can defend in a diligence room.

How it normally goes. Entity setup, then a cost stack you cannot forecast cleanly: statutory audit fees that scale with turnover, transfer pricing certification once intercompany transactions cross ₹1 crore, GST registration if you invoice locally, a corporate tax rate of 35% plus surcharge and cess for a foreign company or 22% (≈25.17% effective) for a domestic subsidiary electing the concessional regime under Section 115BAA, plus FEMA reporting. Every one has a variable component. The three-year model has error bars wide enough to make a board member uncomfortable.

How it goes with an EOR. The people cost is salary plus statutory employer contributions plus a flat $99 per employee per month service fee, with no setup fee, no minimums and no volume gate. For context, SelectSoftwareReviews puts typical global EOR fixed fees at $300–$1,000 per employee per month, with platform providers commonly around $500. Against a $500 benchmark that is roughly 80% less on the service-fee line  about $48,000 a year across a 10-person India team  and it is the one line in the model with zero variance.

The exit math is what the board actually cares about. Ask the diligence-room question  “if the India programme is cancelled at month 30, what is the wind-down cost and how long does it take?”  and the entity answer is a range with an 18-month tail. The EOR answer is statutory severance plus one notice period. You can put that on a slide.

The General Counsel of a Series C fintech, 260 people

You are: first in-house counsel. Your India ops team handles reconciliation and customer support, touching financial data. You have to sign off on an India structure and then defend it to auditors, regulators and an acquirer.

Your actual fear is not the exit. It is what the exit reveals. A wind-down is when unresolved positions crystallise: whether contractors were really contractors, whether the India team’s activity created a permanent establishment, whether termination clauses were enforceable, whether employee data was handled per the Digital Personal Data Protection Act, 2023.

How it normally goes. The subsidiary is dissolved, and two years later a reassessment notice arrives. Under Section 250 of the Companies Act, 2013, the liability of every director and officer who exercised management powers continues after dissolution and may be enforced as if the company had not been dissolved. Striking off the company does not strike off the exposure.

How it goes with an EOR. The employment relationship, payroll records, statutory filings and termination documentation sit with a provider whose job is to keep them audit-ready  as in the case of a Series C security platform that built its India team under DPDP-compliant contracts from day one, with access-control and IP clauses written in before the first hire, not after. Husys runs internal and external audits quarterly and holds ISO 9001 and ISO 27001 certification  which matters less as a badge than as evidence that the record you will need in year three exists in retrievable form. On permanent establishment specifically, Husys monitors client invoices and deliverables against the activity being performed, which is the practical way to catch a role drifting from “builds product” toward “closes deals” before a tax officer does.

What you still own: the DAPE question. If your India-based staff habitually negotiate or conclude contracts on behalf of the US parent, dependent-agent permanent establishment risk follows the conduct, not the employment paperwork. No EOR structure fixes that. Keep contracting authority at the parent.

The CTO of a 40-person AI company

You are: deciding between sponsoring an H-1B for a researcher and building a research pod in Bengaluru.

The relevant fact is volatility, not cost. Presidential Proclamation 10973, issued 19 September 2025, imposed a $100,000 fee on certain new H-1B petitions. On 8 June 2026 the US District Court for the District of Massachusetts vacated it as an unlawful tax; on 24 July 2026 the First Circuit declined to stay that ruling pending appeal, so the fee is not currently enforceable. Other suits are pending and the proclamation carries its own September 2026 sunset. (Verify current status before relying on this  it has moved several times.)

The lesson isn’t “hire in India instead of sponsoring.” It is that the reason you chose a hiring geography can be invalidated by events outside your control, in either direction, inside a single fiscal year. If US immigration policy can reverse twice in ten months, your India strategy should be built to reverse too. Commit to an entity because the India plan looks permanent, and you can end up locked into a structure designed for a world that lasted nine months. With an EOR, if the calculus flips you unwind inside a notice period rather than a liquidation. The technical work is identical. Only the reversibility differs.

→ Read next: the CTO’s decision above turns on a number, not a principle. Put your own salary and headcount into it: H-1B Alternative Cost Calculator.

🔗 Internal link: The True Cost of Hiring Employees in India · India Expansion Checklist: Mistakes to Avoid

Which US employment terms do not translate to India?

Short answer: almost all of the ones that matter. There is no at-will employment, severance is statutory rather than discretionary, the final paycheck is due in two working days rather than at the next payday, gratuity has no US equivalent, notice periods of 30 to 90 days are contractual and enforced, and director liability survives dissolution instead of being extinguished by it.

Every term below means something different from what a US reader assumes.

How do you test an India structure for reversibility before you commit?

Short answer: score the structure against ten questions before the first hire  exit speed, severance exposure per head, role classification, permanent-establishment conduct, offer-letter accuracy, run-rate to dissolution, thesis testing, succession, residual contracts, and whether you can separate employee obligations from entity obligations. Ten out of ten means you have optionality. Six or fewer means you are carrying an exit you cannot cost or schedule. It is a reversibility test, not a headcount test.

The strongest India-entry structure is not the one that is easiest to launch. It is the one that gives the company a clear, workable path to scale, restructure or exit if the business changes.

The more your India operation requires its own commercial, regulatory or asset-holding infrastructure, the less an employment-only structure solves. That is why the exit decision should be tested at Day 0, before the first employee is hired.

What questions belong on a Day-0 reversibility scorecard?

Short answer: the ten below. Score one point for each honest “yes”. Every “no” is exit cost you are pre-committing to, and the right-hand column tells you what that particular “no” will cost you.

Score 10/10 and you have optionality. Score 5/10 and you have a commitment you haven’t priced.