The NEM 3.0 Logic No Longer Applies
Under Malaysia's previous Net Energy Metering (NEM 3.0) scheme, the sizing logic for C&I rooftop solar was straightforward: install the largest system your roof could support. Every kilowatt-hour exported to the grid earned credits at the full retail tariff — the same rate you paid TNB for imported electricity. Oversizing carried no economic penalty because exported and self-consumed energy were valued identically.
ATAP changes this equation. Under the Solar Accelerated Transition Action Programme, exported energy is credited at the System Marginal Price (SMP) — Malaysia's wholesale electricity clearing price, typically RM 0.19–0.24/kWh. Self-consumed energy still displaces the retail tariff at approximately RM 0.334/kWh. The gap between these two rates — roughly 34% — means every kWh exported instead of consumed represents a loss of RM 0.11 in value.
The Two-Rate Problem
To understand why oversizing hurts under ATAP, consider what happens to each unit of solar generation:
| Destination | Effective Rate | Value per 1,000 kWh |
|---|---|---|
| Self-consumed (displaces TNB import) | RM 0.334/kWh | RM 334 |
| Exported (SMP credit) | ~RM 0.22/kWh | RM 220 |
| Forfeited (excess credits) | RM 0.00/kWh | RM 0 |
The third row is critical. Under ATAP, export credits that exceed your monthly TNB bill are forfeited — they do not roll over. An oversized system that consistently generates more export credits than it can offset against monthly consumption effectively gives away electricity for free during those surplus periods.
Quantified Comparison: Load-Matched vs Oversized
Consider a factory in Selangor with a contracted Maximum Demand (MD) of 280 kW, operating Monday–Saturday on a standard industrial schedule (7am–6pm). The facility's annual electricity consumption is approximately 1,460,000 kWh. Two system sizes are proposed:
Option A: Load-Matched (250 kWp)
| Parameter | Value |
|---|---|
| System size | 250 kWp |
| Annual generation | 325,000 kWh |
| Self-consumption ratio | 85% |
| Self-consumed generation | 276,250 kWh |
| Exported generation | 48,750 kWh |
| Self-consumption savings | RM 92,268/year |
| Export credit value | RM 10,725/year |
| Forfeiture loss | ~RM 0/year |
| Total annual savings | RM 102,993/year |
| System cost (estimated) | RM 525,000 |
| Simple payback | 5.1 years |
Option B: Oversized (400 kWp)
| Parameter | Value |
|---|---|
| System size | 400 kWp |
| Annual generation | 520,000 kWh |
| Self-consumption ratio | 58% |
| Self-consumed generation | 301,600 kWh |
| Exported generation | 218,400 kWh |
| Self-consumption savings | RM 100,734/year |
| Export credit value | RM 48,048/year |
| Forfeiture loss (est.) | ~RM 8,400/year |
| Total annual savings | RM 140,382/year |
| System cost (estimated) | RM 840,000 |
| Simple payback | 6.0 years |
The Incremental Economics
The key metric is not total system payback — it is the return on each incremental kWp above the load-matched baseline. This is where the ATAP penalty becomes visible:
| Metric | First 250 kWp | Next 150 kWp |
|---|---|---|
| Capital cost | RM 525,000 | RM 315,000 |
| Annual savings | RM 102,993 | RM 37,389 |
| Savings per kWp | RM 412/kWp/yr | RM 249/kWp/yr |
| Effective blended rate | RM 0.317/kWh | RM 0.192/kWh |
| Simple payback | 5.1 years | 8.4 years |
| 25-year ROI | 390% | 197% |
The first 250 kWp earns an effective blended rate of RM 0.317/kWh because 85% of generation displaces retail tariff. The next 150 kWp earns only RM 0.192/kWh — below even the current SMP average — because nearly all incremental generation is exported, and a portion is forfeited entirely on weekends and public holidays.
Where Forfeiture Occurs
Credit forfeiture is not a theoretical risk — it is a structural feature of ATAP that activates under specific, predictable conditions:
- Weekends and public holidays. Factories operating Monday–Saturday still generate solar on Sundays. A 400 kWp system produces approximately 1,600 kWh on a clear Sunday — 100% of which is exported at SMP. For facilities closed on weekends, the figure doubles.
- Festive shutdowns. Malaysian factories typically close for Hari Raya (1–2 weeks), Chinese New Year (3–5 days), Deepavali, and year-end periods. During these windows, the solar system generates at full capacity with zero self-consumption. All generation exports at SMP, and if accumulated credits exceed the reduced monthly bill, the excess forfeits.
- Low-production months. Facilities with seasonal demand variation — common in food processing, textiles, and electronics — may have months where production drops below baseline. If the solar system was sized for peak demand, these low-production months trigger export surges.
- Billing cycle misalignment. TNB billing cycles do not always align with calendar months. A facility may have a billing period that spans a festive shutdown, concentrating forfeiture risk into a single bill.
The Self-Consumption Curve
The relationship between system size and self-consumption is not linear — it follows a declining curve. For a typical Selangor C&I facility with 280 kW MD operating on day-dominant schedules:
| System Size | % of MD | Self-Consumption | Effective Blended Rate | Simple Payback |
|---|---|---|---|---|
| 140 kWp | 50% | ~95% | RM 0.328/kWh | 4.7 years |
| 210 kWp | 75% | ~88% | RM 0.321/kWh | 4.9 years |
| 250 kWp | 89% | ~85% | RM 0.317/kWh | 5.1 years |
| 280 kWp | 100% | ~80% | RM 0.311/kWh | 5.3 years |
| 350 kWp | 125% | ~68% | RM 0.283/kWh | 5.8 years |
| 400 kWp | 143% | ~58% | RM 0.270/kWh | 6.0 years |
What EPC Proposals Often Miss
Standard EPC proposals in the Malaysian market typically present solar economics using one of two flawed methodologies:
- Roof-maximum sizing. The system is sized to fill the available roof area, regardless of the facility's load profile. This approach was rational under NEM 3.0 but is suboptimal under ATAP. A factory with 600 m² of usable roof and 180 kW MD does not need a 400 kWp system.
- 100% self-consumption assumption. The financial model assumes all generated electricity is consumed on-site. This overstates savings by 10–25% for systems above 80% of MD, because it ignores the export rate differential and forfeiture risk entirely.
Neither approach reflects ATAP economics. A credible financial assessment under the current framework must separate savings into self-consumed (retail tariff) and exported (SMP) streams, model forfeiture risk based on the facility's operating calendar, and present the incremental economics of each additional kWp above the load-matched baseline.
How to Evaluate a Proposal Under ATAP
When assessing an EPC solar proposal for your facility, apply these five checks:
- Does it disclose the self-consumption ratio? If the proposal shows only total generation and total savings without splitting self-consumed vs exported, the economics are incomplete.
- What SMP rate is used? The export credit should reference the latest published Monthly Average SMP from Single Buyer Malaysia — not a generic estimate. As of February 2026, the 12-month average is approximately RM 0.218/kWh.
- Is forfeiture modelled? For systems above 80% of MD, the proposal should account for credit forfeiture during weekends, holidays, and low-production periods.
- What are the incremental economics? Ask for the payback on the last 50 kWp of the proposed system, not just the blended total. If the incremental payback exceeds 8 years, the system is likely oversized.
- Is the system sized to MD or to roof? If the proposed capacity exceeds 100% of your contracted MD, ask for the load-profile justification.
The Strategic Conclusion
Solar under ATAP is an optimisation problem, not a maximisation problem. The goal is not the most kWh generated — it is the highest proportion of kWh consumed on-site at retail tariff displacement rates. Every kWp above the load-matched sweet spot earns less, risks forfeiture, and extends payback.
For CFOs and facility managers evaluating proposals: the right question is not "how much solar can we install?" but "what system size maximises the self-consumption ratio while maintaining acceptable payback?" The answer is almost always smaller than the roof-maximum proposal on your desk.