The Rupee Tariff Trap

In 1986, Bhutan’s first great dam began selling electricity to India at 27 paise a unit. Today Chhukha earns Rs 2.55 — the tariff has been renegotiated upward again and again, more than nine-fold in all. Every revision was announced as a win, and in rupees, every revision was.

Convert those same tariffs into dollars and the nine-fold rise flattens to roughly two-fold — over four decades. The rest never reached Bhutan. It dissolved between the currencies: the rupee in which the power is priced lost value faster than the tariff was raised. Measured in the money Bhutan needs for everything it imports from beyond India, Chhukha’s real earnings have fallen by most of their original value while the plant itself ran perfectly.

That is the rupee tariff trap. It has three moving parts, and none of them is hidden — they are simply never added up.

The mechanics

Part one: the price is in rupees, for decades. Every export PPA — Chhukha, Tala, Mangdechhu, now Punatsangchhu-II — denominates its tariff in Indian rupees, for terms of 25 to 35 years. There is no exchange-rate clause in any of them: no dollar indexation, no depreciation trigger, no automatic re-opener. The rupee has gone from about ₹45 to the dollar in 2010 to roughly ₹95 today; the contracts carry on as if that were nobody’s business.

Part two: the revisions chase, and never catch. Tariffs do get renegotiated — but nominally, episodically, and from behind. Chhukha’s 9.4× rise across four decades trailed the rupee’s slide plus Indian inflation combined. Even the newest deal repeats the pattern: PHPA-II’s escalation steps — 7%, then 5%, every five years — sit below recent Indian inflation, a tariff designed to decline gently in real terms from the day it was signed.

Part three: the risk runs entirely one way. Bhutan bears the revenue risk (rupee income), and much of its project cost is effectively hard-currency (imported turbines, steel, expertise). India bears — nothing. Its rupee payments cost it less in real terms every year; even the loans it extended are repaid in rupees whose value declines with its own currency. The buyer’s discount compounds automatically; the seller’s loss does too. No one has to renegotiate anything for Bhutan to earn less each year. That is what makes it a trap rather than a bad deal: it worsens by default.

The ledger

Add it up across the operating fleet — every plant, every tariff era, converted at the rates of its day — and the catalogue’s estimate of the realised loss to rupee depreciation since 1988 is about USD 1.85 billion, Chhukha alone accounting for some USD 700–800 million. For scale: that is well above Bhutan’s entire gross international reserves — the loss exceeds everything in the vault.

The forward ledger is worse, because the pipeline is bigger than the fleet. Run the same erosion over the existing plants plus Punatsangchhu-I for their remaining contract lives and the projected loss is roughly USD 3–3.7 billion in present value. Extend it across the full 25-gigawatt build-out Bhutan plans by mid-century — ten more projects on the same model — and the projection reaches USD 12–20 billion in present value: four to six times Bhutan’s current annual GDP. These are constructed estimates, not audited figures — they depend on discount rates and on the rupee behaving as it has for fifty years. But the direction is not in doubt, and the magnitude embarrasses every other number in the water economy.

The honest complication

There is a real hedge buried in the structure, and an honest account has to show it. Roughly seven-tenths of Bhutan’s external debt is itself rupee-denominated — mostly the hydropower loans. While a plant is repaying its loan, the same depreciation that erodes its revenue also shrinks its debt service: losses on one side, relief on the other. The trap half-offsets itself — during the loan years.

Then the loan ends, and the hedge ends with it. A 35-year PPA with a 15-year loan leaves two naked decades: the very years a dam should become pure profit are the years its rupee income stands fully exposed, with nothing left to offset. The trap tightens precisely when the plant was supposed to start paying for the country.

The clauses Bhutan has never asked for

None of this requires abandoning the rupee, the peg, or the partnership. Cross-border power trade elsewhere handles currency risk with ordinary contractual furniture: indexation clauses that adjust tariffs against a reference rate; part-payment in hard currency for a share of the offtake; shorter re-opener cycles tied to depreciation thresholds rather than diplomatic cycles. Bhutan’s PPAs contain none of these — including the one signed in May 2026, which suggests the trap is not yet treated as a negotiating item at all.

And there are the structural escapes, which Bhutan has quietly begun without naming them as such. Using more power at home — the 2025 crossover — is a natural hedge: a unit consumed by Bhutanese industry is a unit not sold for depreciating rupees. Converting surplus power into assets that are not rupees is another. Each shrinks the exposed share of the river.

The real deadline is the pipeline. The ten planned projects will roughly triple the contracted volume; whatever currency terms they sign will govern Bhutan’s export earnings into the 2070s. Sign them on the old template and the USD 12–20 billion projection stops being a warning and becomes a schedule. Ask for the clauses — any of them — and the trap, for the first time in forty years, stops being automatic.

The rupee will keep doing what it does. The only question the tariff trap ever asks is whether Bhutan keeps agreeing, in writing, not to notice.