A10.04 — Electricity Costs Behind Malaysia’s Data Centres – Projected Electricity Revenue Over 15 Years
Key Takeaways
- A10.04 — Electricity Costs Behind Malaysia’s Data Centres – Projected Electricity Revenue Over 15 Years establishes the 15-year electricity revenue profile for the 300 MW reference data centre, translating its assumed load take-up into an illustrative electricity consumption and revenue stream.
- The reference case uses an analytical load take-up schedule of 100 MW, 150 MW, 200 MW and 250 MW in Years 1 to 4, followed by 300 MW from Year 5 to Year 15. These values are modelling assumptions and are not prescribed data-centre load-pick-up requirements.
- Applying the assumed 90% effective MD in Years 1–4, 85% thereafter and a 70% load factor, the reference case produces approximately 21.063 TWh of electricity consumption and RM12.798 billion of nominal electricity-bill revenue over 15 years.
- The projected revenue comprises energy, capacity, network and retail charges under the assumed UHV Time-of-Use tariff structure, with zero AFA for the purposes of this analytical reference case.
- The calculation holds the selected tariff rates constant over the 15-year period to isolate the effect of the assumed load profile. In practice, tariffs, running costs and other regulatory parameters may change according to the applicable regulatory period.
- An illustrative low-load scenario produces approximately 11.589 TWh of electricity consumption and RM7.042 billion of nominal electricity revenue, demonstrating the material effect that lower load take-up can have on the revenue stream.
- The low-load scenario also produces an illustrative RM51.510 million of Connected Load Charge over the first six years under the assumptions stated in the article. CLC is a separate charge and should not be treated as a substitute for electricity revenue that is not generated.
- Nominal cumulative revenue is not the value used directly for Connection Charge assessment. TNB’s 1st Principle methodology deducts running costs and relevant consumer-related charges from revenue before determining the present value of future cash flow.
- The resulting cash flow provides the revenue-side input for the next stage of the A10 analysis, where it will be compared with Project Cost to examine the applicable Connection Charge or Refundable Connection Charge (RCC) under TNB’s published methodology.
Parent article: A10 — Who Pays for the Electricity Behind Malaysia’s Data Centres?
From Project Cost to Projected Revenue
→A10.03 established the infrastructure and cost side of the reference case:
Infrastructure → Asset Cost → Attributable Asset Cost → New-Build Cost → Project Cost
This article develops the corresponding revenue side:
Contractual Maximum Demand → Electricity Consumption → Monthly Charges → Annual Revenue → 15-Year Revenue Stream
The distinction is important because the two calculations answer different questions.
The infrastructure analysis (→A10.03) asks what electricity-system assets are required to serve the reference data centre and what portion of their economic value or cost can reasonably be attributed to it. The revenue analysis asks how much electricity-related revenue the same consumer would generate over the 15-year supply period.
The reference case uses a hypothetical 300 MW data centre in Iskandar Puteri, Johor, with a staggered Maximum Demand declaration reaching a Final MD of 300 MW. For the analytical revenue model, the assumed effective demand is set at 90% of the declared Staggered MD in Years 1–4 and 85% of the Final MD from Year 5 onward, providing a more conservative reference than assuming the data centre will consistently achieve 100% of its planned demand. These levels are informed by, but are not requirements under, TNB’s published Connected Load Charge (CLC) framework, which uses 85% of declared Staggered MD for Years 1–4 and 75% of Final MD for Years 5–6. A consistent 70% load factor is assumed in calculating the electricity that would be consumed over the 15-year period. Both the effective demand profile and the load-factor assumption are MDCO analytical assumptions rather than TNB-prescribed operating requirements.
These figures are not, however, the final economic return used in a 1st Principle Connection Charge calculation.
Under TNB’s published 1st Principle methodology, annual electricity revenue is reduced by the applicable running cost and consumer-related charges to derive the relevant annual cash-flow amount. The future cash flows are then converted into present value using the applicable WACC over the 15-year period. A Connection Charge arises where the Present Value of Future Cash Flow is lower than the applicable Project Cost. TNB’s 2025 worked example uses a 15-year period and a 7.3% WACC, subject to changes according to the applicable regulatory period.
Accordingly, the cumulative revenue figure should not be confused with present value. It is the nominal sum of the annual electricity bills under the assumptions used in this article.
The analytical sequence is therefore:
Projected electricity revenue → Running Cost deduction → Annual net revenue → Present Value → comparison with Project Cost → Connection Charge / RCC
A10.04 establishes the first part of this sequence — the projected electricity revenue stream — and examines how that stream changes under a materially lower load-take scenario. The present-value calculation and comparison with Project Cost are left to the subsequent article.
MDCO Insight: This article establishes the 15-year electricity revenue stream and its sensitivity to load take-up, while leaving the conversion of future net revenue into present value and the resulting Connection Charge calculation to the subsequent article.
The 15-Year Reference Data Centre
The revenue calculation uses the same reference data centre established in the preceding articles.
The reference case is a hypothetical 300 MW data centre in Iskandar Puteri, Johor, supplied through a high-voltage transmission connection. The analysis uses a 15-year supply period, consistent with the period used in TNB’s published 1st Principle methodology.
The reference data centre is assumed to operate continuously, with high utilisation of its electrical infrastructure once the relevant load has been commissioned. However, the analysis does not assume that the full 300 MW is consumed from the first year.
Instead, the calculation follows a staggered Maximum Demand profile. The final contractual MD is 300 MW, but the assumed demand increases progressively during the 15-year period.
The distinction between Maximum Demand and electricity consumption is fundamental.
Maximum Demand represents the electrical demand imposed by the consumer on the electricity system. It is the demand measure used in the relevant demand-related billing components and is also important in planning the electricity supply infrastructure.
Electricity consumption, by contrast, is the amount of electrical energy used over time, measured in kWh.
For the purposes of this analysis:
Declared Maximum Demand → Effective Maximum Demand → Load Factor → Electricity Consumption
The declared or contractual Maximum Demand (MD) establishes the demand profile against which the supply is planned. For this analysis, however, it is not assumed that the data centre will consistently achieve 100% of its planned MD. Instead, an effective MD is derived using reference levels informed by TNB’s Connected Load Charge (CLC) framework: 90% of the declared Staggered MD during Years 1–4 and 85% of Final MD from Year 5 onward. These are MDCO analytical assumptions, not TNB operating requirements. TNB’s published CLC framework uses 85% of declared Staggered MD in Years 1–4 and 75% of Final MD in Years 5–6.
A consistent 70% load factor is assumed when calculating the electricity that would be consumed over the 15-year period.
The reference assumptions are:
| Parameter | Reference Assumption |
|---|---|
| Location | Iskandar Puteri, Johor |
| Final contractual MD | 300 MW |
| Supply | High-voltage transmission connection |
| Supply period | 15 years |
| Operating pattern | Continuous operation |
| Effective MD | 90% of declared Staggered MD in Years 1–4; 85% of Final MD thereafter |
| Load factor | 70% |
| Load development | Staggered contractual load pick-up |
| AFA | RM0 for this analysis |
| Electricity consumption | Derived from effective MD and load factor |
The reference case is an analytical model. It does not represent the electricity consumption profile, tariff arrangement or contractual terms of any particular data-centre project.
Establishing the Contractual Load-Pick-Up Profile
The most important input to the 15-year revenue calculation is the contractual Maximum Demand profile.
A 300 MW Final Maximum Demand does not necessarily mean that the data centre will require or achieve 300 MW of actual demand from the first day of supply. For a large new consumer, TNB’s current Connected Load Charge (CLC) framework provides for Staggered MD declarations for Years 1–4, with the Final MD used for the CLC reference in Years 5–6. TNB also states that the declared Final MD is used to determine the required supply infrastructure.
For the present 300 MW reference case, this analysis therefore adopts an illustrative staggered contractual MD profile that reaches the Final MD within the first five years:
| Year | Declared Staggered / Final MD (MW) | % of Final 300 MW |
|---|---|---|
| 1 | 100 | 33.3% |
| 2 | 150 | 50.0% |
| 3 | 200 | 66.7% |
| 4 | 250 | 83.3% |
| 5–15 | 300 | 100% |
This profile is an MDCO analytical assumption, not a TNB-prescribed load-development schedule. The analysis then applies an additional margin to the declared demand when estimating the effective MD actually achieved for electricity consumption. Rather than assuming that 100% of the planned MD will be consistently achieved:
Effective MDₜ = 90% × Declared Staggered MDₜ for Years 1–4
and:
Effective MDₜ = 85% × Final MD for Years 5–15
The resulting effective MD profile is therefore:
| Year | Declared MD (MW) | Effective MD (MW) |
|---|---|---|
| 1 | 100 | 90 |
| 2 | 150 | 135 |
| 3 | 200 | 180 |
| 4 | 250 | 225 |
| 5 | 300 | 255 |
| 6 | 300 | 255 |
| 7 | 300 | 255 |
| 8 | 300 | 255 |
| 9 | 300 | 255 |
| 10 | 300 | 255 |
| 11 | 300 | 255 |
| 12 | 300 | 255 |
| 13 | 300 | 255 |
| 14 | 300 | 255 |
| 15 | 300 | 255 |
Thus, the analytical sequence used in the revenue model is:
Declared / Contractual MD → Effective MD → Load Factor → Electricity Consumption → Electricity Revenue
This distinction is important because the declared MD is relevant to supply planning and CLC, whereas the effective MD is the analytical demand used in this article to estimate electricity consumption and the associated electricity bill.
The reference profile should not be interpreted as a prediction of when any particular data-centre building will be commissioned or reach a particular operating load. It is a reference contractual demand profile constructed for the financial analysis.
The later low-load scenario will further distinguish between the effective reference demand and a materially lower actual MD, allowing the sensitivity of electricity revenue to lower-than-expected demand to be examined separately.
MDCO Insight: The reference case assumes a 300 MW Final MD, reached through a staggered declaration over the first five years, with 90% of declared MD in Years 1–4 and 85% of Final MD thereafter used as the effective MD for the revenue analysis.
Establishing an Appropriate Data-Centre Load Factor
Maximum Demand alone does not determine the quantity of electricity consumed.
A data centre may have a high effective Maximum Demand while its average electrical demand over a year remains below that level. The relationship between the two is represented in this analysis by the load factor.
For each year, annual electricity consumption is calculated as:
Annual Energyₜ = Effective MDₜ × Load Factor × 8,760 hours
where:
- Annual Energy is measured in kWh;
- Effective MD is measured in kW;
- Load Factor represents the ratio between average demand and Maximum Demand; and
- 8,760 represents the hours in a non-leap year.
A consistent 70% load factor is assumed when calculating the electricity that would be consumed over the 15-year period.
For example, an effective MD of 255 MW at a 70% load factor implies an average demand of 178.5 MW and annual electricity consumption of approximately:
255 MW × 70% × 8,760 = 1,563 GWh per year
The 70% load factor is an MDCO analytical assumption. It is not a TNB requirement, nor is it intended to represent a universal operating characteristic of data centres.
A data centre may operate continuously and still have a load factor below 100%, because its average electrical demand need not equal its measured Maximum Demand.
For the reference calculation, the 70% load factor is held constant across the 15-year analytical period. This isolates the effect of the effective MD profile on electricity consumption and revenue.
The consequence is straightforward:
Lower Load Factor → Lower kWh Consumption → Lower Energy Revenue
while, under the reference tariff structure:
Maximum Demand → Capacity / Network Charges
is treated separately in the electricity-bill calculation.
This distinction also becomes important in the low-load scenario. There, the load factor remains at 70%, while the actual MD is reduced. This isolates the effect of lower actual demand rather than simultaneously changing the assumed operating pattern.
MDCO Insight: The reference calculation uses a 70% load factor to convert effective Maximum Demand into annual electricity consumption, providing a separate analytical assumption from the contractual demand profile.
Establishing the Electricity-Tariff Basis
The revenue calculation requires a defined electricity-tariff basis so that the effect of the reference demand profile can be translated into an annual electricity bill.
For the 300 MW reference case, this analysis uses the Non-Domestic Ultra High Voltage (UHV) Time of Use (ToU) tariff applicable under the electricity tariff schedule introduced from 1 July 2025. The UHV category is specifically identified in the tariff schedule and is the appropriate tariff basis for the reference case’s assumed 275 kV high-voltage transmission connection. The published UHV rates are:
| Tariff component | Reference rate |
|---|---|
| Peak energy charge | RM0.5518/kWh |
| Off-peak energy charge | RM0.5109/kWh |
| Capacity charge | RM21.76/kW of peak-period MD/month |
| Network charge | RM23.06/kW of peak-period MD/month |
| Retail charge | RM250/month |
| AFA | RM0 for this analysis |
The UHV tariff applies separate energy rates during peak and off-peak periods. Under the revised ToU schedule, the peak period is 2:00 pm to 10:00 pm on weekdays, while weekends are treated as off-peak. The revised schedule substantially increases the number of off-peak hours, with the peak period representing approximately 22% of annual hours when the published treatment of 15 public holidays is taken into account. For the reference case, 22% of annual electricity consumption is therefore assumed to occur during the peak period and 78% during the off-peak period. This is an analytical allocation based on the published ToU time structure rather than an assumption taken from TNB’s earlier High Voltage worked example.
The tariff rates are held constant throughout the 15-year calculation. This does not represent a forecast that these rates will remain unchanged. The current tariff structure forms part of Regulatory Period 4 (RP4), which commenced on 1 July 2025, and electricity tariffs and related regulatory parameters may be revised in subsequent regulatory periods.
The purpose of holding the rates constant is to isolate the effect of the load-development and electricity-consumption assumptions on the revenue profile. A future analysis may examine the effect of tariff revisions and other variable components separately.
For this article, AFA is assumed to be RM0. This is an analytical simplification rather than an assertion that AFA will remain at zero throughout the 15-year period. AFA is a monthly mechanism under the Incentive-Based Regulation framework that adjusts generation-related charges for changes in fuel and other specified generation costs.
The resulting annual electricity bill consists of four principal components:
Energy Charges + Capacity Charges + Network Charges + Retail Charge
The energy charges depend on the electricity consumed in kWh and its allocation between peak and off-peak periods. The capacity and network charges are based on Maximum Demand during the applicable peak period, while the retail charge is fixed monthly.
The electricity bill represents the gross revenue stream used as the starting point. Under TNB’s 1st Principle methodology, running costs and relevant consumer-related charges are subsequently taken into account before determining the present value of future cash flow for comparison with Project Cost.
The tariff basis used here should therefore be understood as a current-rate analytical reference, not a 15-year tariff forecast. It provides a consistent basis for examining how the assumed demand and consumption profile translates into electricity revenue before the analysis proceeds to running costs, discounting and Connection Charge assessment.
MDCO Insight: The reference revenue model applies the UHV ToU tariff applicable to the 300 MW data-centre reference case, using the published UHV rates, a 22% peak / 78% off-peak consumption allocation, and AFA of RM0 for analytical purposes.
Calculating the Monthly Electricity Bill
With the contractual Maximum Demand profile, effective demand, load factor and applicable tariff established, the next step is to translate the demand assumptions into an illustrative electricity bill.
The reference case uses the Non-Domestic Ultra High Voltage (UHV) Time-of-Use (ToU) tariff effective from 1 July 2025 provided in the previous section.
For the continuous-load reference case, 22% of annual electricity consumption is allocated to the peak period and 78% to the off-peak period. This reflects the applicable ToU structure and provides a consistent analytical basis for the 15-year calculation.
The 70% load factor established in Section 4 is used to determine electricity consumption. The Effective MD, rather than the Final MD or declared Contractual MD, is therefore the demand input used when converting demand into energy consumption.
The AFA is set at RM0 for this article. This is an analytical simplification that isolates the underlying relationship between the reference data centre’s electricity consumption, Maximum Demand and the applicable tariff structure. Future tariff changes and generation-related adjustments can be examined separately.
Monthly Energy Consumption
For a given Effective MD:
Monthly Energy = Effective MD × Load Factor × 730 hours
At the Final MD stage, the reference case has an Effective MD of 255 MW, based on the 85% assumption established in Section 3.
Therefore:
255 MW × 70% × 730 hours = 130.305 GWh/month
Applying the assumed 22% peak and 78% off-peak allocation:
Peak-period consumption = 130.305 GWh × 22% = 28.667 GWh/month
Off-peak consumption = 130.305 GWh × 78% = 101.638 GWh/month
Monthly Energy Charge
The peak-period energy charge is:
28.667 million kWh × RM0.5518/kWh = RM15.818 million
The off-peak energy charge is:
101.638 million kWh × RM0.5109/kWh = RM51.927 million
Therefore:
Total monthly energy charge = RM67.745 million
Monthly Capacity and Network Charges
The capacity charge is based on the Maximum Demand during the applicable peak period:
255,000 kW × RM21.76/kW = RM5.549 million/month
The network charge is:
255,000 kW × RM23.06/kW = RM5.880 million/month
The fixed retail charge is:
RM250/month
No AFA is included in the reference calculation.
6.4 Representative Monthly Bill at the Final MD Stage
The resulting illustrative monthly electricity bill when the reference case has reached its 300 MW Final MD, corresponding to an Effective MD of 255 MW, is:
| Component | Monthly amount |
|---|---|
| Peak energy charge | RM15.818 million |
| Off-peak energy charge | RM51.927 million |
| Total energy charge | RM67.745 million |
| Capacity charge | RM5.549 million |
| Network charge | RM5.880 million |
| Retail charge | RM0.00025 million |
| AFA | RM0 |
| Total monthly bill | RM79.174 million |
Thus, under the stated assumptions, the reference data centre would generate an illustrative electricity bill of approximately RM79.2 million per month, or approximately RM950.1 million per year, once the 300 MW Final MD stage has been reached.
This illustrates why electricity revenue cannot be estimated using a simple RM/kWh rate alone. The bill also contains capacity and network charges linked to Maximum Demand, while the energy component depends on actual electricity consumption.
The same methodology is applied to each year of the 15-year supply period using the corresponding Effective MD.
MDCO Insight: The reference electricity bill combines energy charges based on consumption, capacity and network charges based on Maximum Demand, and a fixed retail charge. At the 300 MW Final MD stage, the reference case has an Effective MD of 255 MW and an illustrative annual electricity bill of approximately RM950 million, before AFA or other adjustments.
Projected Annual Electricity Revenue
The monthly calculation can now be applied to the 15-year contractual load-pick-up profile established earlier.
The reference case assumes the following declared Contractual MD:
| Year | Contractual MD | % of Final 300 MW | Effective MD |
|---|---|---|---|
| 1 | 100 MW | 33.3% | 90 MW |
| 2 | 150 MW | 50.0% | 135 MW |
| 3 | 200 MW | 66.7% | 180 MW |
| 4 | 250 MW | 83.3% | 225 MW |
| 5–15 | 300 MW | 100.0% | 255 MW |
The intervening values are MDCO analytical assumptions, not a prescribed TNB or ST load-pick-up schedule. The published ST example demonstrates that staggered Maximum Demand declarations can be structured over a long supply period, but it does not establish a universal schedule for data centres.
Annual electricity consumption is calculated from the Effective MD:
Annual Energy = Effective MD × 70% × 8,760 hours
The resulting annual electricity revenue is then calculated using the UHV ToU tariff established in Section 6.
Annual Revenue by Component
| Year | Contractual MD (MW) | Effective MD (MW) | Annual Energy (GWh) | Energy Revenue (RM m) | Capacity Revenue (RM m) | Network Revenue (RM m) | Retail (RM m) | Total Revenue (RM m) |
|---|---|---|---|---|---|---|---|---|
| 1 | 100 | 90 | 551.88 | 286.921 | 23.501 | 24.905 | 0.003 | 335.330 |
| 2 | 150 | 135 | 827.82 | 430.382 | 35.251 | 37.357 | 0.003 | 502.993 |
| 3 | 200 | 180 | 1,103.76 | 573.843 | 47.002 | 49.810 | 0.003 | 670.658 |
| 4 | 250 | 225 | 1,379.70 | 717.303 | 58.752 | 62.262 | 0.003 | 838.320 |
| 5 | 300 | 255 | 1,563.66 | 812.944 | 66.586 | 70.564 | 0.003 | 950.097 |
| 6 | 300 | 255 | 1,563.66 | 812.944 | 66.586 | 70.564 | 0.003 | 950.097 |
| 7 | 300 | 255 | 1,563.66 | 812.944 | 66.586 | 70.564 | 0.003 | 950.097 |
| 8 | 300 | 255 | 1,563.66 | 812.944 | 66.586 | 70.564 | 0.003 | 950.097 |
| 9 | 300 | 255 | 1,563.66 | 812.944 | 66.586 | 70.564 | 0.003 | 950.097 |
| 10 | 300 | 255 | 1,563.66 | 812.944 | 66.586 | 70.564 | 0.003 | 950.097 |
| 11 | 300 | 255 | 1,563.66 | 812.944 | 66.586 | 70.564 | 0.003 | 950.097 |
| 12 | 300 | 255 | 1,563.66 | 812.944 | 66.586 | 70.564 | 0.003 | 950.097 |
| 13 | 300 | 255 | 1,563.66 | 812.944 | 66.586 | 70.564 | 0.003 | 950.097 |
| 14 | 300 | 255 | 1,563.66 | 812.944 | 66.586 | 70.564 | 0.003 | 950.097 |
| 15 | 300 | 255 | 1,563.66 | 812.944 | 66.586 | 70.564 | 0.003 | 950.097 |
| 15-year total | — | — | 21,063 | 10,951 | 896.952 | 950.538 | 0.045 | 12,798 |
Under the reference assumptions, the data centre would consume approximately 21.06 TWh of electricity over the 15-year period.
The corresponding nominal electricity-bill revenue would be approximately:
Energy charges: RM10.951 billion
Capacity charges: RM896.9 million
Network charges: RM950.5 million
Retail charges: RM45,000
Total nominal bill revenue: RM12.798 billion
Energy charges account for approximately 85.6% of the total nominal bill revenue, while capacity and network charges together account for approximately 14.4%.
These are cumulative nominal bill values. They do not represent the present value of the revenue stream or the amount ultimately available to recover the Project Cost.
MDCO Insight: Under the reference assumptions, the 300 MW data centre consumes approximately 21.06 TWh over 15 years and generates approximately RM12.80 billion in nominal electricity-bill revenue, with the annual revenue increasing as the declared and Effective MD move through the assumed load-pick-up profile.
Worst-Case Load-Take Scenario
The reference case assumes that the data centre’s declared Maximum Demand increases from 100 MW in Year 1 to a Final MD of 300 MW, with the corresponding Effective MD increasing from 90 MW to 255 MW.
A different question arises if the data centre does not take up the expected load as planned.
For this article, a worst-case load-take scenario is used to illustrate the financial effect of materially lower actual electricity demand:
- Years 1–3: actual Maximum Demand of 30 MW, equivalent to 10% of the 300 MW Final MD; and
- Years 4–15: actual Maximum Demand of 150 MW, equivalent to 50% of the 300 MW Final MD.
The 70% load factor, UHV ToU tariff, 22% peak / 78% off-peak consumption allocation and other revenue assumptions remain unchanged from the reference case.
The scenario is not intended as a forecast. It is an analytical stress case designed to isolate the effect of lower actual electricity demand on the electricity revenue stream.
| Period | Actual MD | % of 300 MW Final MD |
|---|---|---|
| Years 1–3 | 30 MW | 10% |
| Years 4–15 | 150 MW | 50% |
Annual Electricity Consumption
At the assumed 70% load factor, annual electricity consumption is:
30 MW × 70% × 8,760 hours = 183.96 GWh/year
and:
150 MW × 70% × 8,760 hours = 919.80 GWh/year
Over the 15-year period, cumulative electricity consumption would therefore be approximately:
3 × 183.96 + 12 × 919.80 = 11,589.48 GWh
or approximately 11.59 TWh.
This compares with approximately 21.06 TWh in the reference case.
Annual Electricity Revenue
Applying the UHV ToU tariff and the same peak/off-peak allocation established in Section 6, the illustrative annual electricity bill is approximately:
- RM111.779 million per year at 30 MW actual MD; and
- RM558.881 million per year at 150 MW actual MD.
The resulting cumulative nominal electricity revenue is:
| Revenue Component | Years 1–3 | Years 4–15 | 15-Year Cumulative |
|---|---|---|---|
| Energy charges | RM286.921m | RM5,738.426m | RM6,025.347m |
| Capacity charges | RM23.501m | RM470.016m | RM493.517m |
| Network charges | RM24.905m | RM498.096m | RM523.001m |
| Retail charges | RM0.009m | RM0.036m | RM0.045m |
| Total electricity revenue | RM335.336m | RM6,706.574m | RM7,041.910m |
The low-load scenario therefore generates approximately RM7.042 billion of nominal electricity-bill revenue over 15 years, compared with approximately RM12.798 billion in the reference case.
The scenario also produces approximately 11.59 TWh of electricity consumption, compared with 21.06 TWh in the reference case.
MDCO Insight: Under the assumed worst-case load-take scenario, the data centre would consume approximately 11.59 TWh and generate approximately RM7.04 billion of nominal electricity-bill revenue over 15 years, substantially below the reference case.
Consequences of Lower Load Take-Up
Lower electricity consumption has two distinct financial effects.
The first is the loss of electricity revenue because the data centre purchases less electricity than under the reference case.
The second is the possibility of a Connected Load Charge (CLC) where the actual Maximum Demand recorded falls below the applicable Reference Maximum Demand during the CLC period.
These are separate mechanisms and should not be conflated.
Lost Electricity Revenue
The reference case generates approximately RM12.798 billion of nominal electricity-bill revenue over 15 years.
The low load-take scenario generates approximately RM7.042 billion.
The resulting nominal revenue shortfall is therefore:
RM12.798 billion − RM7.042 billion = RM5.756 billion
This RM5.756 billion is not a penalty or charge. It represents electricity revenue that would not arise because the data centre consumes less electricity under the low load-take scenario.
Connected Load Charge
TNB’s current framework provides for a Connected Load Charge (CLC) for applicable Medium Voltage and High Voltage consumers. For a new consumer, CLC applies for six years from the date the supply is connected. TNB describes CLC as a mechanism intended to discourage over-declaration of load requirements.
For a new consumer, TNB’s published methodology provides:
- Years 1–4: Reference Maximum Demand (RMD) = 85% of the higher of the declared staggered MD for the relevant year or the highest recorded MD; and
- Years 5–6: RMD = 75% of the higher of the declared Final MD or the highest recorded MD.
The current published CLC rate is RM8.50/kW for each kW of shortfall, calculated monthly and subject to prevailing changes. The shortfall is between the actual Maximum Demand recorded and the applicable RMD. CLC therefore does not apply to the entire declared MD; it applies only to the portion by which the actual recorded MD falls below the RMD.
This is distinct from the 90%/85% Effective MD assumption used in the electricity-revenue model. The latter is an MDCO analytical assumption for estimating electricity consumption and revenue; it is not the CLC threshold.
For the low load-take scenario, the declared MD used for CLC follows the reference case:
| CLC Year | Declared MD | Applicable RMD |
|---|---|---|
| Year 1 | 100 MW | 85% × 100 = 85.0 MW |
| Year 2 | 150 MW | 85% × 150 = 127.5 MW |
| Year 3 | 200 MW | 85% × 200 = 170.0 MW |
| Year 4 | 250 MW | 85% × 250 = 212.5 MW |
| Year 5 | 300 MW Final MD | 75% × 300 = 225.0 MW |
| Year 6 | 300 MW Final MD | 75% × 300 = 225.0 MW |
The low load-take scenario assumes an actual MD recorded of 30 MW in Years 1–3 and 150 MW from Year 4 onward.
For this analytical scenario, the stated actual MD is assumed to represent the highest actual MD recorded in a month during the relevant year.
The CLC shortfall is therefore:
- Year 1: 85.0 − 30.0 = 55.0 MW
- Year 2: 127.5 − 30.0 = 97.5 MW
- Year 3: 170.0 − 30.0 = 140.0 MW
- Year 4: 212.5 − 150.0 = 62.5 MW
- Year 5: 225.0 − 150.0 = 75.0 MW
- Year 6: 225.0 − 150.0 = 75.0 MW
After Year 6, the CLC period for a new consumer has ended under the stated framework.
MDCO Insight: The CLC is calculated only on the shortfall between actual recorded MD and the applicable RMD. It is therefore separate from both the declared MD and the Effective MD assumption used for electricity-revenue modelling.
Low Load-Take Scenario — Revenue and CLC
The combined calculation makes the distinction between declared MD, actual recorded MD, Reference Maximum Demand and CLC clear.
The following table applies the published CLC framework to the simplified low load-take scenario. For illustration, the stated actual MD is assumed to represent the highest MD recorded in a month during the relevant year. In an actual account, CLC is calculated from the recorded monthly MD and the applicable highest-recorded-MD provisions.
| Year | Declared MD | Actual MD Recorded | RMD | CLC Shortfall | Annual CLC | Electricity Revenue |
|---|---|---|---|---|---|---|
| 1 | 100 MW | 30 MW | 85.0 MW | 55.0 MW | RM5.610m | RM111.779m |
| 2 | 150 MW | 30 MW | 127.5 MW | 97.5 MW | RM9.945m | RM111.779m |
| 3 | 200 MW | 30 MW | 170.0 MW | 140.0 MW | RM14.280m | RM111.779m |
| 4 | 250 MW | 150 MW | 212.5 MW | 62.5 MW | RM6.375m | RM558.881m |
| 5 | 300 MW | 150 MW | 225.0 MW | 75.0 MW | RM7.650m | RM558.881m |
| 6 | 300 MW | 150 MW | 225.0 MW | 75.0 MW | RM7.650m | RM558.881m |
| 7–15 | 300 MW | 150 MW | — | — | RM0 | RM558.881m/year |
For example, the Year 5 CLC shortfall is:
225 MW − 150 MW = 75 MW
or:
75,000 kW
At RM8.50/kW per month:
75,000 kW × RM8.50/kW = RM637,500/month
or:
RM637,500 × 12 = RM7.650 million/year
Applying the same methodology across the first six years gives:
| CLC Year | Annual CLC |
|---|---|
| Year 1 | RM5.610m |
| Year 2 | RM9.945m |
| Year 3 | RM14.280m |
| Year 4 | RM6.375m |
| Year 5 | RM7.650m |
| Year 6 | RM7.650m |
| Total Years 1–6 | RM51.510m |
The highest-recorded-MD provision is important. TNB provides that where the actual recorded MD exceeds the declared MD, the higher recorded MD can replace the declared MD in determining the applicable RMD. The revised declaration framework also provides that the revised Reference MD applies from the subsequent billing cycle.
For this analytical scenario, however, the assumed actual recorded MD does not exceed the relevant declared MD in any of the first four years, and the 300 MW Final MD remains higher than the assumed 150 MW actual MD in Years 5–6. The simplified calculation therefore does not require an upward revision of the RMD.
The RM51.510 million is consequently an illustrative application of the published CLC framework, rather than a determination of an actual customer’s bill.
The important distinction remains:
Electricity revenue falls because less electricity is consumed. CLC is a separate charge arising from a shortfall of actual recorded MD below the applicable RMD during the CLC period.
MDCO Insight: In the illustrative low load-take scenario, the six-year CLC totals approximately RM51.5 million, while the associated reduction in nominal electricity revenue is approximately RM5.76 billion. The two figures arise from fundamentally different mechanisms.
Comparing the Reference and Low Load-Take Scenarios
The two scenarios can now be compared on a common 15-year basis.
| Measure | Reference Case | Low Load-Take Case | Difference |
|---|---|---|---|
| 15-year energy consumption | 21.063 TWh | 11.589 TWh | −9.474 TWh |
| Energy revenue | RM10.951bn | RM6.025bn | −RM4.926bn |
| Capacity revenue | RM896.952m | RM493.517m | −RM403.435m |
| Network revenue | RM950.538m | RM523.001m | −RM427.537m |
| Retail revenue | RM0.045m | RM0.045m | — |
| Total electricity revenue | RM12.798bn | RM7.042bn | −RM5.757bn |
| CLC | — | RM51.510m | RM51.510m |
| Revenue shortfall plus CLC | — | — | -RM5.706bn |
The comparison shows four distinct effects.
First, the low load-take scenario reduces cumulative electricity consumption by approximately 9.47 TWh over 15 years.
Second, the lower demand reduces not only energy-charge revenue but also capacity and network-charge revenue. The resulting nominal electricity-revenue reduction is approximately RM5.757 billion.
Third, the CLC is a separate charge arising only during the six-year CLC period. Under the simplified assumptions used here, it amounts to approximately RM51.510 million.
Fourth, adding the CLC to the nominal revenue shortfall gives approximately RM5.706 billion. This combined figure should not, however, be interpreted as a single financial measure: the first component is electricity revenue not generated, while the second is a separate charge payable under the CLC mechanism.
The CLC should therefore not be interpreted as compensation for the electricity revenue that would have been generated under the reference case. It is a specific charge associated with the consumer’s failure to meet the applicable Reference Maximum Demand during the CLC period.
This distinction is particularly relevant to the economics of infrastructure serving a large data centre. Infrastructure may need to be planned and constructed around a declared Final MD and staggered MD profile, while the electricity revenue subsequently realised depends on the actual load taken and maintained.
The nominal revenue comparison alone, however, is not sufficient for the subsequent 1st Principle Connection Charge analysis.
The relevant next question is how the future net revenue stream, after applicable running costs and other relevant deductions, compares in present-value terms with the attributable Project Cost. TNB’s Connection Charge framework expressly bases the 1st Principle calculation on the Present Value of Future Cash Flow over the 15-year supply period.
That calculation is taken forward to the next stage.
MDCO Insight: The low load-take scenario reduces 15-year electricity-bill revenue by approximately RM5.76 billion, while the illustrative six-year CLC is approximately RM51.5 million. The two should remain separate when assessing the economics of infrastructure serving a large data centre.
Preparing the Revenue Stream for the Next Article
The preceding sections have established the projected electricity revenue associated with the reference data centre and examined how that revenue changes under a materially lower load-take scenario.
The next analytical step is to convert the projected electricity revenue into the form required for the 1st Principle Connection Charge calculation.
TNB’s published methodology describes the sequence as:
Revenue − Running Cost = Annuity Amount
The resulting future cash flow is then converted into present value using the applicable Weighted Average Cost of Capital (WACC). TNB states that a Connection Charge is payable where the Present Value of Future Cash Flow is less than the Project Cost.
Conceptually, the sequence is therefore:
Projected electricity revenue → Running Cost deduction → Annual net revenue → Present Value → comparison with Project Cost → Connection Charge / RCC
The distinction between these stages is important.
The RM12.798 billion calculated for the reference case in this article is cumulative nominal electricity revenue over the 15-year period. It is not the amount that would be used directly in the Connection Charge calculation.
Similarly, the RM7.042 billion generated under the illustrative low-load scenario is a nominal cumulative revenue figure. It does not by itself represent the economic return available to recover the Project Cost.
The next article will therefore take the annual revenue stream established here and deduct the applicable running costs and consumer-related charges before discounting the resulting future cash flows.
Because the reference data centre’s load increases progressively over the 15-year period, the analysis should also preserve the annual profile rather than treating the entire revenue stream as a single constant annuity. This allows the timing of the revenue to be reflected in the present-value calculation.
The annual revenue stream established in this article is therefore the starting point for the next calculation, rather than its final economic measure.
This article stops at the point where the revenue side of the analysis has been established.
The next article brings that revenue stream together with the Project Cost developed in the preceding infrastructure-cost analysis.
MDCO Insight: This article establishes the 15-year electricity revenue stream; the next stage converts that annual revenue into net future cash flow and present-value terms for comparison with Project Cost under TNB’s 1st Principle methodology.
Limitations and Analytical Boundaries
The calculations in this article are intended to establish an analytical reference case rather than forecast the actual electricity revenue of any particular data centre or consumer.
Several assumptions and limitations should therefore be kept in view.
Load-pick-up profile. The reference load profile is an analytical application of the published regulatory and connection framework. The annual values used in this article — 100 MW, 150 MW, 200 MW and 250 MW in Years 1 to 4, followed by 300 MW from Year 5 — are modelling assumptions. They are not prescribed load-pick-up requirements for data centres. TNB’s connection framework is based on the consumer’s declared demand and the required supply scheme, while published examples of staggered Maximum Demand declaration illustrate possible arrangements rather than establishing a universal schedule.
Effective Maximum Demand. For the reference calculation, the effective Maximum Demand is assumed to be 90% of the declared Staggered MD in Years 1 to 4 and 85% of the Final MD thereafter. These levels are MDCO analytical assumptions, informed by the relationship between declared demand and TNB’s Connected Load Charge framework. They are not TNB requirements or prescribed operating levels.
Actual electricity consumption. A data centre’s actual Maximum Demand and energy consumption may differ from the assumed profile because of construction timing, commissioning, IT deployment, operational conditions, efficiency improvements and other factors.
Load factor. A consistent 70% load factor is assumed when calculating the electricity that would be consumed over the 15-year period. This is an analytical assumption for the reference case and should not be interpreted as a universal data-centre operating norm or as a requirement under TNB’s connection-charge methodology.
Tariff rates. The calculation holds the selected UHV Time-of-Use tariff rates constant for analytical purposes. In reality, electricity tariffs and their components may change between regulatory periods. TNB’s published Connection Charges Book notes that tariffs and running costs are subject to revision in line with the applicable regulatory period.
AFA. The analysis assumes zero AFA so that the underlying relationship between demand, electricity consumption and the principal tariff components can be examined without introducing a separate generation-cost adjustment. AFA and related fuel-cost or subsidy questions are outside the scope of this article.
Connected Load Charge. The CLC calculation is illustrative and based on the published framework and assumptions stated in this article. Actual CLC depends on the consumer’s declared MD, actual recorded MD and the applicable Reference Maximum Demand methodology, together with the prevailing CLC rate. TNB currently states RM8.50/kW for each applicable shortfall, subject to prevailing changes. CLC is calculated monthly rather than as a fixed annual charge.
Revenue versus economic return. Electricity revenue is not equivalent to economic profit, cash profit or TNB’s return on investment. TNB’s 1st Principle methodology separately deducts running costs and consumer-related charges before determining the present value of future cash flow.
Nominal revenue versus present value. The cumulative figures presented in this article are nominal amounts. They do not incorporate the time value of money and should not be compared directly with Project Cost for the purpose of determining a Connection Charge.
No Connection Charge calculation. This article does not calculate a Connection Charge or RCC. Those calculations require the projected future cash flows, after the relevant deductions, to be converted into present-value terms and compared with the applicable Project Cost.
These boundaries are important because a 15-year calculation can create an appearance of precision that exceeds the certainty of the underlying assumptions.
MDCO Insight: The 15-year figures are analytical estimates, not a forecast of TNB’s actual revenue or return, because load, tariffs, running costs and regulatory parameters can change over time.
The Observatory Perspective
The cost of connecting a very large data centre cannot be assessed against its infrastructure cost alone.
A large data centre creates a continuing electricity demand over many years, and that demand generates a corresponding stream of electricity revenue. The economic significance of the connection therefore depends not only on the capital cost of the infrastructure required to serve the consumer, but also on the revenue that the electricity system can reasonably expect to recover from supplying that consumer over the relevant period.
This article has established that revenue side.
For the 300 MW reference data centre, the assumed load profile and 70% load factor produce approximately 21.063 TWh of electricity consumption and RM12.798 billion of nominal electricity revenue over 15 years. The illustrative low-load scenario produces approximately 11.589 TWh and RM7.042 billion, demonstrating that the revenue outcome can change materially when the consumer takes substantially less electricity than the reference profile.
The analysis has also shown why several concepts should remain separate.
A declared Maximum Demand is not the same as actual Maximum Demand. An analytical effective Maximum Demand is not the same as either of those measures. Electricity revenue is not the same as a CLC. CLC is a separate charge and is not a substitute for lost electricity revenue. And cumulative nominal revenue is not the same as the present value of future net cash flow.
These distinctions become particularly important when a large electricity connection involves substantial existing and new infrastructure. →A10.03 established the infrastructure cost side of the reference case, including the distinction between existing attributable asset value and new or modified infrastructure costs.
A10.04 has now established the revenue side.
The next stage is therefore to bring the two sides together:
Project Cost → projected net revenue → Present Value → Connection Charge / RCC
That is the point at which the analysis can begin to examine how the infrastructure investment compares with the future net cash flow associated with supplying the data centre, and how the applicable connection-charge mechanism addresses any resulting difference.
That calculation belongs to the next article.
Selected References
Electricity Tariff and Connection-Charge Framework
- Tenaga Nasional Berhad (TNB) – Connection Charges Book 2025, setting out the 1st Principle and Refundable Connection Charge (RCC) methodologies, Project Cost treatment, the 15-year revenue period and the application of WACC. https://www.tnb.com.my/assets/files/TNB_Connection_Charges_Book_2025_ENG.pdf
- Tenaga Nasional Berhad (TNB) – Connection Charge, providing the current connection-charge framework for High Voltage consumers, including the 1st Principle methodology, Project Cost, WACC and the distinction between dedicated and non-dedicated supply. https://www.tnb.com.my/connection-charge/
- Tenaga Nasional Berhad (TNB) – Charges & Penalties – Connected Load Charge, setting out the current CLC rate, Reference Maximum Demand methodology and monthly application of CLC. https://www.tnb.com.my/commercial-industrial/charges-penalties/temporary-supply/
Electricity Tariff and Regulatory Framework
- Suruhanjaya Tenaga (ST) – Components of IBR, explaining the Base Tariff and Automatic Fuel Adjustment (AFA) mechanisms, including CAPEX and OPEX for transmission and distribution and returns on regulated assets. https://www.st.gov.my/pricing/electricity-pricing-framework/components-ibr
- Tenaga Nasional Berhad (TNB) – Electricity Tariff Schedule, providing the applicable UHV Time-of-Use energy, capacity, network and retail charges used as the tariff basis for the reference calculation. https://www.mytnb.com.my/tariff/index.html?lang=en&v=1.1.70
Supply and Demand Framework
- Tenaga Nasional Berhad (TNB) – Electricity Supply Application Handbook (ESAH), covering supply schemes, demand estimation, PMU requirements, transmission routes, land allocation and right-of-way/wayleave requirements for electricity-supply infrastructure. https://www.tnb.com.my/esah
Citation
Malaysia Data Centre Observatory (MDCO). A10.04 — Electricity Costs Behind Malaysia’s Data Centres – Projected Electricity Revenue Over 15 Years. MDCO Analyse Series.
MDCO Note
This article forms part of the Malaysia Data Centre Observatory (MDCO) Analyse Series, including the A10 and A10.xx series on the electricity costs and cost-recovery mechanisms associated with data centre development. The series aims to improve public understanding through evidence-based, accessible and balanced analysis. It is intended for educational and informational purposes only and does not constitute legal, engineering, financial, regulatory or professional advice.
The cost of supplying electricity to large data centres involves multiple disciplines and perspectives, including electrical engineering, energy economics, utility regulation, infrastructure planning, finance and public policy. MDCO does not advocate for or against any particular stakeholder, project, technology, tariff or regulatory position. Its role is to facilitate transparency, reduce information asymmetry and support informed discussion by examining how electricity infrastructure is provided, paid for and ultimately recovered through the electricity system, using publicly verifiable information and clearly stated assumptions.