Skip to main content

Research and analysis across business, economics, artificial intelligence and technology

Energy & Commodities

Research Brief4 min readIssue 03

The Price of Cheap Power

Iran and Ethiopia sell the world’s cheapest electricity. Neither price reflects what the power actually costs — and both countries are now paying for that gap, one way or another.

$0.003
Iran residential electricity price per kWh
$0.006
Ethiopia residential electricity price per kWh
155x
difference between the lowest and highest prices shown
Residential electricity price
155x apart — and neither number tells you the real cost
Iran
$0.003
Ethiopia
$0.006
Global avg.
$0.176
Bermuda
$0.466
Bar length ∝ √price, so the smallest bars stay visible — exact values labeled.
Source: GlobalPetrolPrices.com, Q2 2026 (world average), Q4 2025 (Iran, Ethiopia), via Visual Capitalist for the Bermuda comparison. These are subsidized retail prices, not production cost — the gap between the two is the whole point of this piece.

Most “cheapest electricity” rankings measure what a household pays, not what power actually costs to produce and deliver. That gap is where this story lives.

In Iran, the gap is filled by a government subsidy on cheap domestic gas. In Ethiopia, it was filled — until recently — by simple neglect: tariffs that hadn’t been revisited in years while currency and costs moved on without them. Both are now facing a reckoning. Iran’s arrived as an unmanaged crisis in 2025–2026. Ethiopia’s is a deliberate, four-year price correction it’s about halfway through.

01 — Iran · Subsidized Fossil

Cheap because someone else is paying

Roughly 80% of Iran’s power comes from natural gas priced at about $34,000/mcm — a tenth of what U.S. households pay for the same gas. Hydro and renewables barely register.

Generation mix

Iran runs on gas — almost entirely

80%
GAS
Natural gas80%
Other fossil~15%
Hydro + renewables~5%
Source: Enerdata; Low-Carbon Power.

The IEA puts Iran’s power-sector subsidy alone at roughly $12.5 billion a year. A 2022 Energy Policy study found that removing it entirely would cut demand 16% and emissions 31% by 2050, and push renewables from 2.5% to nearly half the generation mix — simply by letting price do its job.

Worth understanding why the subsidy exists at all: by Iran’s own implicit economics, exporting gas as electricity is the worst-earning thing the country can do with it. A Stanford analysis found electricity exports return about $70,000 per million cubic meters of gas — versus $155,000 for petrochemicals, $270,000 for gas exports, and $450,000 for compressed natural gas sold as vehicle fuel. Domestic power, priced under the subsidy, earns even less. There’s never been much internal pressure to price power properly.

95k MW
Installed, on paper
~72k MW
Actually operational
25k MW
2026 forecast shortfall

That gap is now a lived crisis: rolling blackouts, industrial revenue lost to outages rising from ~300 to ~400 trillion tomans between 2024 and 2025, and emergency-market electricity trading at 4x the regulated price. A drought and the 2026 conflict with the U.S. and Israel have compounded things — but the underlying structure, a system built almost entirely on subsidized gas with a razor-thin hydro cushion, is the constant.

02 — Ethiopia · Underpriced Hydro

Cheap because nobody fixed the price

Ethiopia’s grid is ~94% hydropower — genuinely clean, not subsidized fossil fuel. But before the country’s last tariff reform in 2018, residential customers paid just 1.3–3.4¢/kWh against a real ~9.2¢ cost of service — and currency depreciation ate most of that 2018 increase anyway, so tariffs were still only around 3¢ by 2022, after Ethiopia had adjusted prices just five times in over sixty years.

Ethiopia’s four-year tariff reset — residential mid-band

ETB per kWh, 201–300 kWh/month tier

2.00
Before
update
3.84
Year 1
5.68
Year 2
7.52
Year 3
9.35
Year 4
A 4.7x increase for this tier by 2028; industrial customers see a 6x increase over the same period. Source: Ayele & Tesfaye, Energy for Growth Hub (July 2026).

The generation side just got a major boost, too: the Grand Ethiopian Renaissance Dam, Africa’s largest hydropower project, added 5,150 MW — roughly doubling national capacity — when it was inaugurated in September 2025, after a $5 billion build financed almost entirely by Ethiopians themselves.

The tariff trigger was a 2021 debt default and a 2024 IMF program requiring cost-reflective pricing.

5–12x
Utility revenue growth, 2019–2025
5.6M
Active connections, up from 3M
60M
Ethiopians still without power

Even after the full reform, tariffs will still sit below actual cost — and the Energy for Growth Hub flags five open risks that could still undercut the gains: the cost structure stays underpriced even post-reform; revenue can’t rely on tariffs alone; higher bills don’t automatically buy reliability; a 4–6x increase strains affordability against wages that haven’t kept pace; and each new rural connection costs $100–$1,000 against customers with little ability to pay, so access expansion remains the harder unsolved problem. One more wrinkle worth knowing: because prices rise every quarter, the $0.006/kWh ranking figure up top is already a few tariff updates behind Ethiopia’s actual current rate.

03 — What Durable Cheap Power Looks Like

Paraguay and Norway

Paraguay runs 99.7% on hydro and exports 75% of it — genuinely abundant, priced at 5.4–7.6¢/kWh with no subsidy involved. That surplus is now drawing serious AI-data-center investment (up to $50B floated by one firm alone), though combined transmission and distribution losses near 21% (4% + 17%) show that abundance alone doesn’t automatically reach everyone. Norway generates ~89% hydro and is about as clean a grid as exists — yet still saw real price spikes in 2021–22 once its market connected more tightly to the rest of Europe. The lesson: abundance helps, but no grid is fully insulated once it’s interconnected.

Why It Matters

The Business Read

Cheap, reliable power has become a decisive site-selection factor for energy-intensive investment — data centers most visibly, but also manufacturing and anything power-cost-heavy. Data-center demand can, under the right conditions, actually lower average costs by spreading fixed grid costs across more usage; but reporting on PJM, the largest U.S. grid operator, shows that same demand now pushing prices upward as buildout outpaces new generation. Malaysia offers a live preview — new tariffs there are already nudging data-center investment toward Vietnam and Thailand instead.

The first kind of cheap is an asset worth building around. The second kind is a bill that hasn’t arrived yet.

For anyone weighing where to build energy-intensive capacity, that’s the actual question behind any “cheap electricity” pitch: is the price backed by real, cost-reflective abundance — or is it a subsidy running on borrowed time?

TRIMLINE RESEARCH

Issue 03 in an ongoing series delivering business, economic and commodity insight — from the team behind the Trimline Group.

References

  1. Aryanpur, V., Fattahi, M., Mamipour, S., Ghahremani, M., Ó Gallachóir, B., Bazilian, M. D., & Glynn, J. (2022). How energy subsidy reform can drive the Iranian power sector towards a low-carbon future. Energy Policy, 169, Article 113190. https://doi.org/10.1016/j.enpol.2022.113190
  2. Ayele, Y. N., & Tesfaye, M. (2026, July). A cheap kilowatt-hour, a weak birr, and a debt crisis walk into a bar…. Energy for Growth Hub. https://energyforgrowth.org/article/a-cheap-kilowatt-hour-a-weak-birr-and-a-debt-crisis-walk-into-a-bar/
  3. Azadi, P., Nezam Sarmadi, A., Mahmoudzadeh, A., & Shirvani, T. (2017). The outlook for natural gas, electricity, and renewable energy in Iran (Working Paper No. 3). Stanford Iran 2040 Project.
  4. Enerdata. (n.d.). Iran energy information. Retrieved August 2026, from https://www.enerdata.net/estore/energy-market/iran/
  5. Ethiopian Electric Power. (2026). Generation [Data set]. https://www.eep.com.et/?page_id=1033
  6. Fortune. (2026, July 26). Data centers were actually making electricity costs cheaper, but the $7 trillion buildout is threatening the trend. https://fortune.com/2026/07/26/data-centers-electricity-costs-cheaper-7billion-buildout-ai-demand/
  7. GlobalPetrolPrices.com. (2026). Electricity prices around the world (Q2 2026 update). https://www.globalpetrolprices.com/electricity_prices/
  8. International Energy Agency. (2026). Iran [Country profile]. https://www.iea.org/countries/iran
  9. International Renewable Energy Agency. (2025). Ethiopia: Energy profile. https://www.irena.org
  10. Malay Mail. (2025, July 1). Malaysia’s data-centre operators face higher power costs under new tariff structure. https://www.malaymail.com/news/money/2025/07/01/malaysias-data-centre-operators-face-higher-power-costs-under-new-tariff-structure/182447
  11. Pappis, I., Centurion, C., Pereira Ramos, E., Howells, M., Ulloa, S., Ortigoza, E., Gardel-Sotomayor, P. E., & Alfstad, T. (2021). Implications to the electricity system of Paraguay of different demand scenarios and export prices to Brazil. Energy Systems, 12, 911–939. https://doi.org/10.1007/s12667-020-00420-w
  12. Rio Times. (2026, July 6). Paraguay data centers: Cheap hydropower draws AI. https://www.riotimesonline.com/paraguay-data-centers-ai-hydropower-itaipu-2026/
  13. Statistics Norway. (2026). Electricity [Data set]. https://www.ssb.no/en/energi-og-industri/energi/statistikk/elektrisitet
  14. World Bank. (2023). Access to electricity (% of population) – Ethiopia [Data set]. https://data.worldbank.org/indicator/EG.ELC.ACCS.ZS?locations=ET
Trimline ResearchIssue 03 — short read. A full research edition with nine figures and the complete reference list is available separately.

Continue reading

Want the complete analysis?

Open the full research note for the expanded country comparisons, additional evidence, nine figures and complete supporting references.

Read the Full Research Note 13 min
Scroll to Top