Thailand’s electrification:
who pays,
who gains?

How to read this page

Each finding carries a grade. Supported: official data or documents back it. Partly supported: the facts exist but the conclusion is one reading among several. Not yet evidenced: a fair question, but no public evidence yet. Figures are as of 27 October 2026, before the official rules of the 50,000-baht scheme were published. This page names no person or company as acting in bad faith; it shows where to look.

Rising demand, and data centres

Supported. Data centres worldwide used about 415 TWh in 2024, roughly twice Thailand’s total use, and the IEA expects about 945 TWh by 2030. Thailand’s planned data centres (about 2.87 GW) could add roughly 10% to national demand at 80% use (our estimate). Few permanent jobs come with them, and they compete for water and grid upgrades. Full detail: Data centres: the real costs and benefits.

Why data centres want clean power

Partly supported. Thailand has no general law that every data centre must use a set share of clean power. The expectation comes from three places:

The reasons: climate pledges and emissions reporting, customers’ supply-chain rules (including EU carbon rules for exporters), long-term price stability, and competition between countries for investment. The gap to watch: about 30% of data-centre power worldwide still comes from coal, and Thailand’s Energy Minister has said operators publicly demand 100% clean energy yet often ask for “any power immediately”.

A fair share: who gets to sell clean power?

Supported on access; partly supported on profit. The direct-sale rules are written for large plants, so a household rooftop cannot take part. Per unit, households and big solar farms are paid about the same; the difference is scale and contract length.

One large listed company signed 25-year contracts with EGAT for 12 solar projects totalling 644.8 MW under the 2022 scheme; a household gets at most 5 kW. The larger money is in fossil capacity: Thailand’s reserve margin was 34–38% in 2025 against a 15% norm, and under some private contracts availability payments are made even when plants are not used, costs that TDRI says flow into tariffs.

What a fairer system could look like: reserve a share of data-centre green-power demand for households, cooperatives and community solar; let aggregators pool many rooftops to reach the 1,000 kVA threshold; publish the full chain of prices; and give households contracts as long as large producers get.

The 200,000-baht tax deduction: who benefits?

Supported: it favours higher incomes. It reduces taxable income, so its value rises with the tax bracket: about 6,100–50,000 baht saved on the full 200,000 baht (SCB EIC). Rules: new on-grid systems up to 10 kWp, installed 3 March 2026 – 31 December 2028, claimant must own the meter and hold an e-tax invoice from a VAT-registered seller.

Side effects to watch: a 2026–2028 rush means panels reaching end of life together while recycling is still being planned, and deadline pressure invites poor installations.

The 50,000-baht scheme: what is still unclear

“Unclear” is supported; a 25-year debt trap is not supported by the terms reported so far; hidden intent is not yet evidenced. Reported terms: 50,000 baht for each of 1 million homes; PEA’s example of a 5 kW system at about 140,000 baht leaves about 90,000 baht; loans at 2.5% for 7 years; surplus bought at 2.20 baht. Still unclear: the vendor list and its criteria; the interest rate after the promotion and the penalty for missed payments; whether the subsidy goes to the installer or the household; the buy-back term (10 or 20 years); how the 90,000-baht balance is financed; and whether batteries and ground-mounted systems get the same terms.

Does a loan pay for itself? Our arithmetic: 90,000 baht at 2.5% over 7 years ≈ 1,169 baht a month. A 5 kW system makes about 600 units a month.

The poorest households pay the lowest tariff, use little power in the day and depend on selling at 2.20 baht, so their margin is thin; a cloudy month, a rate rise or an inverter fault can push them below zero. Why the rules are unclear has ordinary explanations (Cabinet approves money first, then regulators write rules; political timing), and the vendor list, once published, will show who benefits.

What people say

We read about 175 public Facebook comments on four widely shared posts: clear worries outnumbered support about six to one. The main worries were debt, distrust of who supplies the equipment, the low 2.20-baht buy-back price and rising installation prices; supporters said it pays for homes that use power in the daytime. Full summary and a reader poll: What people on Facebook are saying.

Who carries the costs?

Agnotology, and being realistic

A flood of complex detail – panel types, IEC and Thai standards, sizes and scales, the positive published and the negative left out – can leave the public confused enough to give up. Researchers call the deliberate production of confusion agnotology. The answer is not to be negative or positive, but realistic: ask who pays, who gains, for how long, and who checks. There is no free lunch: the 50,000 baht is borrowed public money, the deduction is tax given up, and grid upgrades are paid through tariffs.

• Is the full rulebook published, with dates?
• Can I choose my own installer, and is the vendor list public?
• What is the loan rate after the promotion, and the penalty if I miss a payment?
• Is the buy-back price fixed, and for how many years?
• Does my daytime use make the numbers work without relying on selling power?
• Who supplies the equipment, and is the warranty in writing?

Further reading

1. Atlas of AI – Kate Crawford (Yale University Press, 2021). AI as an extractive industry.
2. Hyperscale – Paris Marx (Profile Books, 2026). The physical footprint of data centres and who bears it.
3. Power and Progress – Daron Acemoglu & Simon Johnson (PublicAffairs, 2023). Who gains from new technology.
4. Agnotology: The Cultural Production of Ignorance – Robert N. Proctor & Londa Schiebinger (eds.) (Stanford University Press, 2008). How confusion and doubt are produced on purpose.
5. Merchants of Doubt – Naomi Oreskes & Erik M. Conway (Bloomsbury, 2010). How doubt is used to delay action.
6. The Shock Doctrine – Naomi Klein (Metropolitan Books, 2007). How a crisis is used to rush policy through.
7. Code Dependent – Madhumita Murgia (Henry Holt / Picador, 2024). AI’s effect on ordinary people and workers.

This is an analysis by Solar ESS, based on the public sources listed below. Where the evidence is thin we say so. We will update it when the official rules of the 50,000-baht scheme are published.

Sources

1. IEA – Energy and AI: energy demand from AI
2. DCD – Google plans 24/7 carbon-free energy for data centers by 2030 (15 Sep 2020)
3. Mayer Brown – German Energy Efficiency Act and data centers (Feb 2024)
4. Hunton – Thailand’s draft direct PPA regulation for data centers (Oct 2025)
5. Green Tech Festival – Thailand gives data centres their own power tariff (14 Aug 2026)
6. BOI – Utility Green Tariff (UGT1/UGT2) presentation (Jul 2024)
7. SolarQuarter – Thailand launches UGT2 (20 Mar 2026)
8. MEA MyEnergy – People’s Solar notices
9. PEA – โซลาร์ภาคประชาชน registration site
10. Watson Farley & Williams – Thailand’s 5 GW renewable FiT scheme 2022–2030 (Oct 2022)
11. Reccessary – 25-year solar PPAs for 12 projects, 644.8 MW (Dec 2023)
12. Oxpeckers – Wired for waste: Thailand’s energy surplus (Oct 2025)
13. The Nation – September 2026 household tariff, first 200 units capped at 3 baht
14. Thansettakij – Royal decree on the 200,000-baht solar tax deduction (4 Mar 2026)
15. SCB EIC – Solar rooftop tax benefits (Jul 2025)
16. Prachachat – PEA plan for the 50,000-baht scheme (21 Oct 2026)
17. Asia News Network / The Nation – The true price of the cloud (Sep 2026)
18. Brookings – New evidence on data center employment effects (2026)
19. IEEFA – PJM capacity prices and data centers (Jul 2025)
20. The Conversation – AI’s invisible and exploited workforce

← All news on the 50,000-baht scheme