Buy High, Sell Low: The National Trading Strategy

The full numbers

Bhutan’s largest hydropower assets — Tala (1,020 MW), Chhukha (336 MW), Kurichhu (60 MW), Basochhu (64 MW) — sell power to India under PPAs signed in the 1980s–2000s at tariffs of INR 2.12–2.55/kWh. These look like simple commercial contracts. They are not. The structure is debt-bundled:

• Government of India financed the dam construction via concessional bilateral loans (~70% debt-funded)
• Bhutan repays via fixed-tariff electricity supply over 30+ years
• The “INR 2.12/kWh” effectively bundles principal + interest + operating margin
• Net retained after debt service: ~Nu 0.60–1.00 per kWh Meanwhile, the same electricity sold to a Bhutanese cement or ferro-alloy factory costs Nu 1.60/kWh under the HV1 industrial tariff (ERA 2022 revision) — and that revenue is fully retained, with no debt-service offset. Per kilowatt-hour retained: | Channel | Realised to Bhutan (Nu equivalent) | Net retention | |---|---:|---:| | Tala legacy PPA (INR 2.12) | 2.12 | ~0.60–1.00 (after debt) | | Mangdechhu newer PPA (INR 3.50) | 3.50 | ~2.00 | | PHP-II PPA (INR 4.12) | 4.12 | ~2.50+ | | Indian spot market (DAM, lean season) | 4.00–6.00 | full | | HV1 domestic industrial tariff | 1.60 | full |

Imagine this

A 35-year-old plant manager at a ferro-silicon facility in Pasakha receives her monthly BPC bill. Her plant consumed 4 million kWh that month. At Nu 1.60/kWh, the bill is Nu 6.4 million — paid in ngultrum, fully retained by BPC, contributing to BPC’s dividend to DHI. The same 4 million kWh, sold to India under the Tala PPA, would have netted Bhutan after debt service Nu 2.4–4.0 million — less than what her plant pays. She doesn’t know this. Nor do most Bhutanese. The country’s largest electricity revenue stream is also its lowest-margin one — and the asymmetry has compounded silently for two decades while INR has depreciated against USD by ~47%.

Where this came from

The original PPA structure made economic sense in the 1980s and 1990s. India needed power; Bhutan had no domestic market; concessional Indian financing was the only realistic source of capital for mega-hydropower. The fixed-tariff PPA was a reasonable bargain at the time.

But two things changed:

1. INR depreciated 47%+ against USD since 2010 (and ~85.5% since Chhukha COD in 1986) — the real value of fixed-INR tariffs has been eroding every year
2. Domestic industrial demand emerged — Bhutan now has buyers willing to pay Nu 1.60/kWh fully retained The PPA structure never accommodated either change.

The contracts are bundled with debt service; renegotiating tariffs requires renegotiating the loan structure; the legacy is locked in.

The full FX loss, sized. A project-by-project ledger of every operational Bhutanese hydropower plant against its respective COD INR/USD rate produces a mid-scenario realised cumulative USD-equivalent loss of ~USD 1.85 billion to date (range 1.65–2.05bn) from FX depreciation alone, separate from debt-service or tariff under-pricing. Forward PV loss over remaining tenures of operational plants plus PHP-I to ~2060 (2.5% p.a. INR depreciation, 6% discount): another USD 3.0–3.7 billion. Total realised + forward: ~USD 4.85–5.55 billion, roughly 1.5× current annual GDP in present-value terms. If the full 25 GW DGPC pipeline through 2040 (Sankosh 2,585 MW, Kuri-Gongri 2,800 MW, Wangchhu, Chamkharchhu-I, Dorjilung, etc.) is signed under the same INR-only PPA framework, the pipeline-inclusive PV loss reaches USD 12–20 billion by 2050 — 4–6× current annual GDP. Annual realised FX-loss flow currently runs at USD 50–80 million per year, equal to 20–25% of annual hydropower export earnings in USD terms. The asymmetry is structural: Bhutan bears FX risk on all revenue (PPA, royalty, equity), India bears FX risk on essentially none (INR-loan principal and coupon are both INR-denominated and decline in USD value automatically). See Bhutan Hydropower PPA INR Pricing Loss for the full project-by-project ledger.

The domestic-below-export gap, sharpened. The earlier framing — “Bhutan sells to India at a lower net price than to its own factories” — was about net retention after debt service. The newer finding is sharper: even on the gross tariff, the domestic HV1 rate (Nu 1.60/kWh) sits below the cheapest operational export PPA (Tala Rs 1.98/kWh) and roughly one-third of the latest reference price (PHP-II Rs 5.10/kWh, April 2026 protocol). The implicit subsidy to HV1 industrial customers, measured against foregone PHP-II-equivalent export revenue at 7,000 GWh/year of HV1 throughput, is ~USD 281M/year — comparable in size to the annual Government of India grant. The marginal MWh of new generation has an opportunity cost equal to the highest available export tariff; the gap between that opportunity cost (Rs 5.10/kWh) and the Nu 1.60/kWh HV1 charge is the marginal subsidy per MWh that Bhutan chooses to route domestically rather than sell to India. See Bhutan vs Global Electricity Costs §6 for the implicit-subsidy ledger and the lean-season import-resale loss companion (paradox #63).

Why this matters now

The Tala PPA renewal in 2026–2027 is the single largest electricity-policy decision Bhutan will make this decade. The legacy structure will either:

• Be rolled forward at marginal tariff increases (India’s preferred outcome) — locking in another 20 years of low-net-retention export
• Be renegotiated to reflect current market realities (higher tariff, shorter duration, currency-indexed pricing, domestic optionality clauses) The renewal terms will set the template for Chhukha, Kurichhu, and Basochhu renewals over the following decade. Together these four legacy PPAs represent ~4,500 GWh/year of Bhutan’s export volume — roughly 57% of total exports — locked in INR at below-market rates.

What it should be

A renewal structure that reflects:

• Headline tariff closer to current market (INR 4.50–5.00/kWh, or USD-denominated equivalent)
• Shorter duration (15 years vs 30 years)
• Annual escalation tied to INR/USD reference rate
• USD-denominated tranches (at least 30% of output)
• Domestic-priority clauses (DGPC retains right to redirect up to 20% to higher-value domestic off-take) This is not “anti-India” framing — it is currency and contract diversification consistent with international best practice. The bilateral relationship can survive (and benefit from) honest renegotiation.

How others do it

• Norway → EU electricity exports — market-priced, daily clearing, FX-neutralised
• Quebec → New York / New England — market-priced PPAs with periodic renegotiation; export tariffs above wholesale rates
• Laos → Thailand — newer PPAs at THB 2.4–3.0/kWh with explicit escalation clauses
• Paraguay → Brazil + Argentina (Itaipu) — historic PPA was below market for decades; renegotiation in 2023 finally adjusted terms; Paraguay’s economy was distorted for 50 years by the original structure
• Bhutan: legacy PPAs at INR 2.12–2.55/kWh fixed for 30 years; net retention below domestic tariff after debt service

The question we should be sitting with

Why does a country sell its single most valuable export — clean electricity — to its largest neighbour at a price lower than it charges its own factories? And why are we still doing this 20 years into the contracts?