Vietnam C&I rooftop solar: the return is anchored to the tariff table — and so is the risk
Composite-board score: 5.4 · Where it is strong: Net cash flow, Tax friendliness · methodology
The model is clean: put solar on a factory roof, self-consume against the factory's power bill (owner model), or fund the system and sell the power to the factory at a discount (ESCO model). Every anchor of the arithmetic is checkable: MOIT's tariff decision (1279/QD-BCT) publishes manufacturing prices (1,833 VND/kWh normal hours, 3,398 at peak), Vietnam yields 1,100–1,550 kWh/kWp a year (south over north), and installers publish prices around 8.5–9.0M VND/kWp. This site's model (1 MWp in the south, 90% self-consumption) gives first-year bill savings of about 28% of capex gross, and about 12–20% net cash return after O&M, inverter replacement and degradation — consistent with the academic review's 5–8-year paybacks. Note what that is: a model, not any operator's audited performance, graded operator-estimate on this page. The first risk is not technical but the tariff regime itself: the two-part tariff pilot from July 2026 shifts bills toward capacity charges and cuts the energy component by roughly 25–35%, rewriting the savings base; surplus feed-in is capped at 20%; and the government is studying a tariff cut. Every change to the tariff table re-opens this arithmetic — hence do, on conditions, the condition being the tariff table in force in the month you sign.
In short: four lines first
About 8.5–9.0M VND/kWp; 100 kWp about 900M VND (roughly US$35K), 1 MWp about 8.5bn VND
12–20%（Vietnam C&I rooftop solar net cash return (modelled)）
Medium (ESCO is hands-off after contracting; the owner model carries its own O&M)
The two-part tariff pilot is the largest policy variable: Decree 146/2025 applies two-part tariffs from July 2026 to customers at 22 kV and above using 200,000+ kWh a month — shifting bills toward capacity charges and cutting the energy component by roughly 25–35%. Solar saves on the energy component; every point it falls shrinks self-consumption savings, and the largest customers — exactly the ESCO target set — are hit first. Re-run the full model on the post-pilot tariff table before signing anything.
The numbers that matter
Why we think it makes money
Both sides of the fraction are verifiable: the tariff is a government pricing document, the yield is a solar-resource atlas, the capex is installers' published price lists — their product (net 12–20%, payback 5–8 years) is a model, but no line of it comes from a seller's pro forma.
The demand is a real factory cost line: manufacturing tariffs have risen through repeated adjustments, peak pricing (3,398) overlaps solar output hours closely, and self-consumption displaces the most expensive electricity the factory buys.
The asset/contract shape is comparatively simple: 25-year panel life against a 5–8-year payback makes cash flows front-loaded and relatively determinate for an industrial asset — provided the tariff regime holds still.
How to actually get in
- Owner model: the factory owner funds the system and offsets its own bill — an equipment investment, not a financial one, suited to readers who already run a Vietnamese factory.
- ESCO model: the investor funds the system and sells power to the factory at an agreed discount (PPA-style); roof rights, contract term versus factory life, and termination handling are the contractual core.
- Licensing path: generation is a conditional business line; self-consumption systems register under Decree 58/2025 with grid-connection procedures case by case — verify with local counsel before entry; this page makes no "no-barrier" claim.
What it costs, what it earns
Entry amount: About 8.5–9.0M VND/kWp; 100 kWp about 900M VND (roughly US$35K), 1 MWp about 8.5bn VND
How much work for you: Medium (ESCO is hands-off after contracting; the owner model carries its own O&M)
- Capex: about 8.5–9.0M VND/kWp for commercial systems (installers' published lists, >100 kWp basis); 1 MWp is about 8.5bn VND (roughly US$330K).
- O&M: about 1–1.5% of capex a year (cleaning, inspection, insurance); one inverter replacement around years 10–12 must be reserved.
- Silent drags: panel degradation of about 0.5% a year; self-consumption below design (factory shutdowns, holidays, roof changes) cuts savings directly — the 90% self-consumption in the model is a discipline, not a default.
Basis: this site uses full-cost net returns only — income minus vacancy, management, maintenance, taxes and insurance, then financing interest, with entry costs in the denominator. See the methodology。
Identity & tax
Status (residency / citizenship)
This asset does not count toward the channel This asset leads to no status
Threshold: DT3 注册资本 30 亿 VND 起（居留卡最长 3 年）
资产本身不产生居留：越南的投资居留走的是对企业注册资本的 DT 分级（DT3 30 亿–<500 亿 VND 给最长 3 年居留卡），买设备、买物业或持股外围资产都不计入；以投资人身份设立运营公司是另一条独立通道，须以律师核实的注册资本口径为准。
Source: 越南《出入境法》及投资签证 DT 分级规定
Tax (investor level)
自发自用的「回报」是省下的电费（避免成本），不是售电收入；ESCO 售电则按经营收入纳税。电价制度本身是最大的税务面变量。
- 企业所得税标准税率 20%：ESCO 模式的售电收入按经营收入计税；自用模式的节省额体现为成本下降、利润上升，间接进税基。
- 两部制电价试点（146/2025 号议定，2026-07 起对 ≥22 kV、月用电 ≥20 万 kWh 客户）：电费向容量费转移、电量电价部分下降约 25–35%——直接改写自用节省额的计算基础，签约前必须以最新电价表重算。
- 余电上网按 58/2025 号议定封顶 20%（工贸部已提议放宽至 50%，尚为草案），按上年平均市场价结算。
Double-tax treaties: 越南与新加坡有税收协定；屋顶光伏的许可与并网程序按 58/2025 的自发自用登记口径，入场前以当地律师核实为准。
Source: 越南工贸部电价决定与 58/2025、146/2025 号议定
This section is general information only — not immigration or tax advice. Program rules change frequently; before acting, rely on the official texts and a licensed adviser.
Risks: what could go wrong
What we have not verified yet
The following key figures have not yet been verified; the reason is noted for each item.
- Actual generation and savings data for specific projects: this site holds a model and an academic-review range; no audited operating series for Vietnamese C&I rooftop projects is publicly citable.
- The final parameters of the two-part tariff: the pilot basis is published, the final design and provincial implementation are not — savings models must be refreshed when it lands.
FAQ
- Why is 12–20% graded only "operator estimate"?
- Because every input is real but the combination is this site's model: tariff (official), yield (resource atlas) and capex (market lists) are individually checkable, while their product has no operator's audited performance behind it. Evidence grades follow the weakest link — this site's uniform treatment of all modelled numbers.
- Owner model or ESCO — which suits an outside investor?
- The owner model suits people who already have the factory (the saving is your own money). ESCO is the external-investment form, but it stacks factory credit, tariff policy and roof tenure into one contract, and generation is a conditional business line needing local compliance. The returns share a source; the risk structures are entirely different.
- Can solar go on a leased factory's roof?
- Yes, but the contract is the investment: you need the roof owner's consent, a remaining lease term covering the 5–8-year payback, and agreed handling of equipment and compensation on early termination. A lease shorter than the payback rules the project out under either model; ESCO contracts must equally lock roof rights and the factory-closure case.
Related opportunities
Get the free one-page brief
Get the one-page brief / subscribe for updates
Leave a contact and this page will immediately show a time-limited download link for the brief (valid 24 hours). This site only sends information — it will never pitch you a specific project or pass your contact to any project party.
Received — thank you.
Your brief download link: Download the one-page brief (PDF/HTML) (valid for 24 hours; submit again after it expires).
Updates are sent only after the monthly briefing launches, and every email carries an unsubscribe link. To have your contact details deleted, contact us as described in the Privacy Policy.