Doubled for the Nation, Untouched for the Few

The full numbers

The 2025–2028 Bhutan Power Corporation tariff application — filed in December 2025, on the mandated 3-year revision cycle — proposed raising the Low Voltage rate from Nu 2.66 to an unsubsidised Nu 5.73 per unit (the final consumer rate, after the subsidy the Royal Government must still approve, would be lower). The breakdown:

• Low Voltage customers: 99.96% of total customer count, ~10% of total domestic electricity consumption — face a 115% tariff increase
• High Voltage customers: 23 industrial entities (66kV and above), 88% of total domestic consumption — bear their own transmission/substation/demand-charge costs but face no comparable percentage burden in the revision
• Highland subsidy preserved: 200 free units per month
• Lowland subsidy preserved: 100 free units per month
• Regional comparison: Bhutan’s proposed unsubsidised Nu 5.73 still lowest vs West Bengal Nu 7, Assam Nu 8, Bangladesh Nu 7.5, Tibet Nu 8 The export-domestic asymmetry that makes this sharper: Legacy PPAs (Tala, Chhukha, etc.) net Bhutan only Nu 0.60–1.00/kWh after debt service. Tala’s contract price to India is INR 2.12/kWh; Mangdechhu INR 2.55/kWh. The proposal would put Bhutanese LV consumers paying Nu 5.73 — roughly 3× the price India pays for the same electricity, and 5× what Bhutan retains net after debt service. Prime Minister Tshering Tobgay intervened: ERA was instructed to look at affordability and the tariff proposal was sent back to ERA. As of late May 2026 no final ERA determination has been issued — the proposal must still clear ERA, then the Ministry of Energy and Natural Resources and Cabinet (which sets the subsidy) — and the current Nu 2.66 LV / Nu 1.60 HV tariff remains in force.

The global-comparator dimension. The same 23 HV1 customers receiving the lighter percentage increase already enjoy what is, on a fully-loaded basis, the 3rd–4th cheapest industrial electricity tariff in the world (Bhutan HV1 = $0.019/kWh; only subsidised Iran and renewable Ethiopia are lower; Iceland aluminium-smelter contracts are 1.85× higher; US Texas industrial 4.3× higher; Germany industrial 9.5× higher; world industrial average 8.5× higher). The LV residential tariff at $0.031/kWh is 18% of the US residential average and 9% of Germany’s. So the structure being defended in the 2026 proposal is not “industrial customers paying market rates while LV gets a break” — it is “industrial customers receiving the world’s deepest electricity subsidy while LV customers (already globally cheap) are asked to absorb a 115% increase to fund the system.” See Bhutan vs Global Electricity Costs §3 for the full industrial benchmark table and paradox #64 for the framing of this as a structural inverted cross-subsidy.

Imagine this

A schoolteacher in Thimphu opens her electricity bill in March 2026. Last month: ~Nu 600 for 250 units (after 100 free units). Under the proposed revision: ~Nu 1,290 for the same consumption — a 115% increase on what is already a tight household budget. The PDP manifesto promised affordable energy for all Bhutanese. The proposed tariff would more than double her monthly burden. She looks across the road at the cement factory on the industrial estate. The factory consumes 50,000 units a month. Under the existing structure, the factory has been negotiated to absorb its own transmission costs but the per-unit tariff revision proposed doesn’t materially change its bill. The 23 industrial customers like that factory account for 88% of all electricity used in the country. The teacher and 99.96% of other Bhutanese households are being asked to absorb the system’s recovery burden; the 23 customers consuming 88% of the electricity are not. At the same time, the electricity flowing through her wall socket was probably exported to India at Tala’s contract price of INR 2.12 per unit. She is now being asked to pay roughly 2.7× that price for the same electricity, generated in her own country.

Where this came from

Bhutan’s hydropower sector was built since the 1980s as an export industry financed by Indian bilateral loans. The legacy PPAs (Tala, Chhukha, Kurichhu, Basochhu) locked in maximum-export framing for 25–30 years at INR-denominated tariffs that were generous at signing but are now far below market. Domestically, the tariff structure evolved separately. Low Voltage customers (households, small commercial) were always heavily subsidised — both with free units and below-cost tariffs. The actual cost of generation and distribution rose with inflation, fuel prices, transmission upgrades, and the post-PHPA-II grid integration costs. The accumulated under-recovery had to land somewhere. The 2026 proposal landed it on LV customers — the 99.96%. The 23 HV industrial customers operate as wholesale buyers; they negotiate their own infrastructure costs and demand charges; their pricing is largely beyond the per-unit retail tariff. So when the regulator looks for cost recovery, the LV pool is structurally where the tariff lever moves.

Why this matters now

Three things converge in 2026:

• Fuel prices are squeezing households (see paradox #1 — petrol subsidy lifted, diesel still subsidised)
• GST is straining consumer prices (no GST paradox written but the pattern is clear)
• Now electricity tariff is proposed to double A 99.96% household electricity doubling, on top of fuel and GST pressures, in an economy where wages have not doubled — is a household-affordability shock at exactly the moment when emigration is already at record levels (paradox #54, #55). The PM’s intervention to send the proposal back to ERA is a recognition that the political-economic math doesn’t work.

What it should be

• Tariff revisions should not load the entire recovery burden onto LV customers
• HV customers consuming 88% of electricity should bear a proportional share of the system-cost increase
• Bhutan’s export tariffs to India should be renegotiated upward at every renewal — at minimum to recover the gap between PPA price and what Bhutan’s own LV customers are paying for the same electricity
• The “Bhutan has the lowest tariff in the region” comparison is true but misleading at the household level — what matters is affordability relative to Bhutanese wages, not relative to neighbours

How others do it

• Norway — hydro-rich economy with similar export-domestic dual structure. Norwegian household LV electricity is generally affordable relative to wages; large industrial buyers pay separately negotiated rates.
• Iceland — kept domestic LV tariffs low through structural subsidy from large industrial users (aluminium); the cross-subsidy goes the OTHER way (industrial subsidising LV) compared to what Bhutan’s 2026 proposal does
• Paraguay — hydro export pattern similar to Bhutan; domestic tariffs are politically protected; the burden of cost recovery is often borne by tariff structure renegotiation, not just domestic LV doubling
• Quebec — Hydro-Québec uses progressive rate structures; large industrial users face different tariffs than residential LV

The question we should be sitting with

If the cement factory and 22 other large industrial customers consume 88% of our electricity, why are the 99.96% of households and small businesses being asked to bear 100% of the tariff revision burden? And why is the country with the world’s cleanest grid asking its own people to pay 3× the price the same electricity is sold to India for?