A Japanese investor wrote a ₹12,700 crore cheque into an Indian telecom venture in 2009, with a negotiated exit: if performance targets were missed, it could leave at no less than half its acquisition price. When the clause was invoked in 2014, the Reserve Bank refused the price; FEMA does not permit a non-resident an assured exit. The money was eventually paid in full, through a London arbitration award enforced in Delhi over RBI's objection. Why does India's exchange-control architecture refuse the promise but allow the damages, and where does downside protection now actually live?
Every foreign investor entering an unfamiliar market wants the same thing: a way out if things go wrong. The standard tools are written into the shareholders' agreement. A put option, which is the right to make the promoter or the company buy your shares back. A buyback obligation. An exit at a guaranteed rate of return. Through the 2000s these clauses were standard in Indian deals, especially in real estate and infrastructure, where foreign money entered assets that could not be sold easily and wanted a promised door out.
India's exchange-control law reads these clauses through one simple principle. Under the Foreign Exchange Management Act, 1999 (FEMA), the pricing rules work in both directions. When shares move from an Indian resident to a foreign investor, the price cannot be below fair value. When the foreign investor sells back to a resident, the price cannot be above fair value. The two rules are the same idea seen from opposite ends: foreign equity must take equity risk. If a foreign investor is promised a fixed exit price regardless of how the company performs, that money enjoys the upside while being protected from the downside, which is how a loan behaves, not equity. And a loan has its own door into India, the external commercial borrowing route, with its own limits and rules; an assured-return equity exit is a loan trying to use the equity door. What the rule protects is the boundary between the two doors.
The law did eventually make room for exit clauses, but only up to that boundary. Options in private share agreements sat in legal doubt until SEBI validated them in October 2013. The Reserve Bank of India (RBI) followed in January 2014: a foreign investor may hold a put option, subject to a minimum holding period of one year, provided the exit happens at the fair value prevailing at the time of exit, with no assured return. The settlement this created is precise, and it is still widely misunderstood. FEMA guarantees the exit; it does not guarantee the price. The put option is legal in India as a right to leave. It is illegal as a right to leave at a number fixed on the day the cheque was written.
The collision between a real contract and this rule played out in full public view in the Tata and NTT Docomo matter, and it remains the reference case for every term sheet since. In 2009, the Japanese telecom company Docomo invested about ₹12,700 crore for roughly a quarter of Tata Teleservices. The agreement said that if performance targets were missed, Docomo could exit at no less than half its purchase price. The targets were missed. Docomo invoked the clause in 2014, and by then the fair value of the shares had fallen far below the promised floor. The exchange-control system held its line: permission to pay the contracted price was refused, because paying a foreign investor above fair value on exit is exactly what the rule forbids. Docomo took the dispute to arbitration in London, and in June 2016 the tribunal awarded it 1.17 billion US dollars. Not as the price of the shares. As damages, for Tata's failure to keep its promise.
The next step is the one that matters most. Docomo brought the award to the Delhi High Court for enforcement. RBI itself applied to the court, objecting that paying the award would achieve precisely the assured exit FEMA prohibits. The court heard the objection and rejected it in April 2017. Damages for breaking a contract, the court held, are not a share price, and paying them needs no exchange-control permission. The money moved. The same month, in the Cruz City and Unitech matter, the same court went further: even where a clause may breach FEMA, that breach alone does not make a foreign arbitration award unenforceable in India, because FEMA violations are civil matters that can be settled through compounding. Between the two judgments, the route was paved. The promise that is refused at the exchange-control counter is paid at the enforcement counter, as damages, and both outcomes are the system working as designed.
This is not one arm of the state defeating another. Each arm protected its own mandate. RBI protects the grammar of the capital account: equity is equity, and no transaction crossing a bank's desk may breach the pricing rule. The courts protect something different: India's reliability as a country where foreign arbitration awards are honoured, which is itself an asset India holds before global capital. The cost of the route is paid by the parties. Docomo invoked its clause in 2014 and was paid in 2017, and that was the fast version, run between two of the most creditworthy counterparties in the world.
For an investor structuring downside protection today, the practical map follows from that story. The lawful core is what the 2014 settlement allows: a contractual exit after the lock-in, at fair value on the date of exit. This is genuine protection against being trapped in an investment, and no protection at all against the investment having lost value. It secures liquidity, not price. Inside the instrument, compulsorily convertible preference shares and debentures allow the conversion formula to be fixed upfront, so an investor can receive more shares if valuations fall; the protection works through the share count, while the eventual sale of those shares still meets the fair-value ceiling. Beyond the instrument, the price promise has moved to where Docomo found it: obligations written as indemnities whose breach produces a damages claim, backed by parent guarantees or assets sitting outside India, with the arbitration venue chosen at the term sheet stage. And the limits must be priced honestly. A damages claim is only as strong as the balance sheet behind it; the Cruz City award was followed by years of chasing the counterparty's assets. The route takes years even when it works.
So the exit clause in an Indian shareholders' agreement is really two promises fused into one sentence. The first is the promise of an exit, which India honours: the option is legal, the lock-in is defined, the fair-value mechanics are prescribed. The second is the promise of a price, which India will never let the Indian counterparty pay as a price, and which survives, if at all, as a damages claim travelling through a foreign arbitration seat to an Indian court, against whatever assets stand behind the signature. Every foreign investor in India signs both promises at once; the diligence that matters is knowing which one FEMA will keep, and what the other one is actually worth.