Joint venture India: governance, control and exit that hold

India attracted gross FDI of USD 81.04 billion (provisional) in FY 2024-25, according to the Press Information Bureau, up 14% from USD 71.28 billion a year earlier. Manufacturing FDI grew 18% to USD 19.04 billion. Over the eleven years to 2025 the country drew USD 748.78 billion, which the government says is nearly 70% of everything received in 25 years. The headline for a CEO is straightforward: the market is open and the capital is plentiful.

The more useful reading is where the money did not go. Outlook Business, reporting Department for Promotion of Industry and Internal Trade (DPIIT) data, notes that the number of source countries has risen from 89 in 2013-14 to 112 in 2024-25. A wider field of foreign investors means more competition for the same well-placed Indian partners. Capital alone is therefore a weak differentiator; what a foreign group brings beyond capital is what earns it a seat.

Industrial policy sharpens the point. India Briefing reports that production-linked incentive (PLI) investments of US$20.3 billion had been realised across 14 sectors by March 2025. In electronics, production rose from US$24.6 billion in FY 2020-21 to US$60.7 billion in FY 2024-25 and mobile phone exports from US$2.6 billion to US$23.1 billion. These are the sectors in which technology transfer and local scale were combined, usually through partnerships. A vendor publication, the figures are consistent with the official PLI outlay described by the PIB.

Policy history explains why the partner universe looks the way it does. The Carnegie Endowment records that Chinese FDI amounts to USD 2.51 billion, or 0.32 percent of India's cumulative equity inflows since 2000, even though bilateral trade is large. Restrictions shaped who could partner with whom. As rules ease at the margin, Indian groups gain more options and foreign entrants face new competitors for the same partners, which makes an early and well-documented offer more valuable.

What to do: decide before approaching any partner what the venture needs from India that you cannot build alone, whether that is channel reach, plant capacity, licences or customer access. Write it as a one-page contribution statement for each side. That document becomes the spine of the term sheet, and it is the first defence against the vague roles that later turn into deadlock.

A conglomerate partnership is attractive because of breadth, but breadth is not the same as control of the specific route you need. The PIB shows Singapore supplying 30% of FDI in FY 2024-25, Mauritius 17% and the United States 11%, which tells you that much foreign capital arrives through holding structures. The Indian group on the other side of your venture is often a family-led business group whose subsidiaries compete with each other, and whose interests may not align with your product.

Geography adds a second filter. Outlook Business reports that Maharashtra took USD 19.6 billion of FDI equity in FY 2024-25 and 39% of the total, against USD 6.61 billion for Karnataka and USD 6 billion for Delhi. Capital clusters where infrastructure, ports and finance sit. A partner strong in one state may add little in another, so the commercial footprint of the partner should be mapped against your customer map, not accepted from a brochure.

Sector mix matters too. Services attracted 19% of equity inflows, computer software and hardware 16% and trading 8%, according to the same reporting. A group that is dominant in trading may be an excellent distributor and a poor technology host. The reverse also holds. Partners tend to claim both capabilities; the diligence task is to test which one is real.

Group governance is a risk that rarely appears in a financial model. Many Indian groups are controlled by families, and decisions that look like operating choices may in fact be family decisions on succession, borrowing or related companies. Ask who inside the group has authority to commit it, how disputes between family branches are handled and whether the person you negotiate with will be present when the venture needs a decision. A partner whose own governance is unclear cannot deliver clear governance to the venture.

One practical test separates distribution partners from financial ones: ask the group to name the five largest customers it would introduce to the venture in the first year, and to say which group company holds each relationship. A partner that answers with specifics, and accepts a target tied to them, is selling something real. A partner that answers in general terms about its brand and reach is offering access to its network at best. The difference should show up in the price you are willing to pay for the stake.

What to do: score three to five candidate groups on four measurable questions: customers they already serve, capacity they already own, regulatory approvals already held, and conflicts with other group companies. Ask for evidence, such as customer contracts and licence copies, instead of presentations. Treat any group that cannot show its distribution reach as a financial investor, and price the venture accordingly.

Control in an Indian joint venture is bounded by rules that many foreign boards discover late. Law.asia notes that, for restricted sectors, FDI caps are typically set at 26%, 49% and 74%, and explains the governance consequence: a 26% holder can block a special resolution, a 49% holder can block special resolutions but not ordinary ones, and a 74% holder cannot pass a special resolution alone. A cap is a governance fact, not only a ceiling on ownership.

Where a foreign partner wants equal say but the cap prevents a 50:50 split, the compromise is often a 49:51 structure with board committees and reserved matters preserving decision parity. Law.asia describes a 49:51 joint venture as usually a compromise for exactly this reason. The foreign investor should accept the compromise only with the reserved-matters list in the agreement and the articles, since shareholder agreements alone can be hard to enforce against the company.

Policy is also moving. The Carnegie Endowment reports that on 10 March 2026 the Union Cabinet approved amendments to Press Note 3, allowing investors with non-controlling beneficial ownership of up to 10 percent to invest through the automatic route, and introducing a 60-day processing timeline. King Stubb & Kasiva adds that the follow-on FEMA amendments clarified that standard affirmative rights in joint venture governance do not automatically constitute control. For investors from land-border countries this is a material change; others gain clearer treatment of veto rights.

Timing deserves attention in the plan. The 60-day processing timeline introduced by Press Note 3 is an administrative commitment that lets a deal team build a closing calendar, but it applies to proposals that need approval, and an application that is incomplete will stop the clock. Prepare the beneficial ownership chart, the investor background and the business plan before filing, and treat Indian counsel's sign-off on those papers as a condition precedent in the term sheet, not a task after signing.

What to do: before choosing 50:50 or 49:51, obtain written advice from Indian counsel on the sector route, the cap and whether your affirmative rights could be read as control. Build the structure around that answer, and recheck it whenever the foreign ownership chain changes, since beneficial ownership tests look through intermediaries.

The first governance document to settle is not the board composition; it is the allocation of functions. In a typical venture with an Indian group, the foreign partner contributes technology, product standards or brand, and the Indian partner contributes distribution, local operations and regulatory relationships. Written clearly, that split determines who nominates the chief executive, who owns the sales target and who signs off on product changes.

Reserved matters should then protect each function rather than veto everything. A foreign partner needs consent rights over changes to product specification, intellectual property use, quality failures and related-party pricing. The Indian partner needs comparable protection over channel policy, local hiring and customer terms. A short, specific list is easier to administer than a long one, and the Cyril Amarchand Mangaldas guidance on 50:50 ventures warns that over-broad definitions risk trivialising a mechanism while over-narrow ones leave dangerous gaps.

Related-party dealing needs special attention in a conglomerate setting. If the Indian group supplies raw materials, logistics or premises to the venture, the agreement should fix an arm's-length pricing method, an audit right and a benchmark review. Without these, the venture can transfer margin to the group while appearing to perform. Where the group is also a competitor in a nearby segment, add a non-compete tied to the venture's field, drafted to be enforceable under Indian law.

Board mechanics deserve the same precision as the responsibility matrix. Quorum rules should require at least one director from each partner so that neither can stall by staying away, while a second meeting on the same agenda with a lower quorum is a common safeguard. Casting votes are usually unavailable in a balanced venture, so reserved matters and the escalation ladder replace them. Nominee rights for the chief executive and finance head should be explicit, with removal rights tied to defined performance failures.

A responsibility matrix also helps with the hardest conversation in a joint venture, which is performance. When one party is accountable for the sales plan and another for the product specification, a shortfall can be traced to a function and discussed with data. When both share every decision, shortfalls turn into disputes about fault. For that reason we would rather see a venture with an imperfect but explicit allocation than one with a perfectly balanced board and no allocation at all.

What to do: draft a one-page responsibility matrix first, with one accountable party for each of ten to twelve functions. Convert it into reserved matters only for the functions where a wrong decision would damage the other partner's contribution. Appoint a joint steering committee below board level to resolve operating disagreements before they reach the shareholders.

Cyril Amarchand Mangaldas calls deadlock provisions the centrepiece of any 50:50 joint venture agreement, and explains why: there is no natural majority to force a resolution and no inherent leverage to compel an exit. In practice that means a disagreement over the budget or a capital call can freeze the venture, with each side waiting for the other to blink. Courts are slow and arbitration is expensive, so the contract has to do the work.

A workable ladder has stages that escalate in seniority and narrow in time. A first stage sends the issue to the board, a second to named senior executives from each group, a third to a mediator, and a fourth to an exit mechanism triggered only after the earlier stages fail. The firm's guidance cautions that neither party may act unilaterally, except where a stage is demonstrably futile. The timings in our exhibit are typical drafting ranges, not rules, and should be set against your operating calendar.

The definition of deadlock deserves as much attention as the mechanism. A trigger limited to board votes misses disputes at shareholder level, while one that counts any disagreement as a deadlock invites tactical use. Tie it to a defined list of matters, such as the annual budget, business plan, appointment of the chief executive and capital increases, with a rule that failure on two consecutive meetings counts.

The Russian roulette and Texas shoot-out mechanisms are a last rung that boards often fear, and Cyril Amarchand Mangaldas reports that the Bombay and Delhi High Courts have affirmed the enforceability of options such as these. That is a useful signal, but enforceability in court is not the same as workability in practice. A buy-sell clause rewards the party with deeper financial resources, so the weaker partner will resist it. Pair it with a valuation floor or an independent valuer so that it cannot be used to squeeze a smaller partner.

Deadlock can also be prevented by design choices that sit outside the dispute clause. An annual budget approved on a rolling basis, with the previous year's budget continuing if no agreement is reached, avoids freezing operations. Pre-agreed authority limits for management allow routine spending to continue. Capital calls can be capped or staged so that neither partner is forced into a dilution contest. Each choice removes a potential trigger and leaves the formal ladder for matters that really are strategic.

What to do: agree the ladder in the term sheet, not in the long-form agreement, because it is hardest to negotiate once relationship pressure has built. Name the individuals or offices on each rung. Test it by running a simulated dispute with both teams during the drafting stage. Joint ventures where the ladder is untested often fail at the third rung.

The exit clause is where a good partnership can still turn costly. Cyril Amarchand Mangaldas notes that valuation methodology in options with a foreign element must satisfy the fair value requirements prescribed by the Reserve Bank of India. Law.asia puts the practical effect plainly: a foreign shareholder cannot sell its stake to the Indian partner at a price above fair value, while a foreign holder with a call option may pay a price equal to or above it.

The Tata-DoCoMo dispute shows the risk. Law.asia records that a dispute arose because the pricing guidelines did not permit sale of DoCoMo's stake at a price above fair value, so a negotiated assured-return exit could not be performed as written. Maheshwari & Co adds that put options are enforceable in India as long as the pricing does not violate FDI valuation norms, and must avoid characteristics of a forward contract.

A foreign partner that already holds the maximum permitted stake will also find a call option hard to exercise, because it cannot lawfully acquire more. The same constraint can apply to change-of-control rights where the cap would be breached. Exit planning therefore needs to start from the sector cap and the cap table, not from the commercial wish to leave at a particular multiple.

Dispute resolution clauses should be written with the exchange control rules in mind. Cyril Amarchand Mangaldas warns that arbitral awards may face the wall of Indian foreign exchange regulation, which means that a favourable award is not self-executing if the price it orders breaches the pricing rules. Choose the seat and rules for arbitration on enforceability grounds, and keep the governing law and the pricing provisions consistent with each other so that an award does not contradict regulation.

What to do: draft exit rights in layers. Put a pre-agreed valuation formula and an independent valuer mechanism first, then a right of first offer, then a right to sell to a third party with transfer restrictions that protect confidential technology. Test every layer against the pricing rules with Indian counsel before signing. Where assured returns are commercially necessary, use instruments other than the equity price, such as licence fees or supply agreements.

Governance documents do not run a business. The first hundred days decide whether the allocation of roles works. The foreign partner should second a small team into the venture for technology transfer, quality systems and reporting, and the Indian partner should place sales and regulatory leaders with defined targets. A venture without named full-time leaders is a legal structure, not an operating company.

Performance reporting should be agreed up front and tied to the incentive schemes the venture uses. PLI schemes cover 14 key sectors with an outlay of Rs. 1.97 lakh crore according to the PIB, and eligibility often depends on investment and output thresholds. If the venture's business plan relies on PLI income, the shareholder agreement should say who bears the risk if those thresholds are missed and who controls the application process.

Intellectual property needs the same discipline. The licence from the foreign partner should be limited in field and territory, subject to quality control, and terminable on defined events. Know-how that is not registered still needs documentation, trade-secret handling and access limits. Where the Indian group is also a manufacturer for other brands, ring-fence the venture's product data.

Reporting rhythm is the last control. Monthly management accounts, a quarterly steering committee and an annual strategy session give both partners a predictable view of delivery. Share the same dashboard with both sides, so that a missed target is visible to the person accountable and to the partner who depends on them. Agree on audit rights in advance, including access to supplier records where the Indian group is a related-party provider, because disputes about data are often disputes about trust.

Finally, plan the second year before the first has ended. A venture that meets its start-up targets usually faces a new set of questions on capacity, pricing, new products and the use of retained profits. If the shareholder agreement has no mechanism for reviewing the business plan and the contributions of each side, the answers will be reached through argument. A scheduled review in month eighteen, with the original contribution statement on the table, keeps the partnership aligned with the commercial case that created it.

What to do: set a 100-day plan with six measures, such as first revenue, plant approval, hiring against plan, supplier contracts, quality pass rate and reporting calendar. Review it at the first board meeting and again at day 100 against the contribution statement drafted before the term sheet. If a function is not being delivered by the partner who agreed to deliver it, use the steering committee to escalate early.

Karter McKinley's International Expansion and Market Entry work supports CEOs and strategy teams who are weighing an Indian partnership. We help define the contribution statement, screen and shortlist partners, map the commercial footprint, coordinate the work of Indian counsel and tax advisers, and run the first hundred days. We are not a law firm and do not provide legal or tax advice.