IGCEP and illusion of long-term planning

In a recent session of Parliament, the Minister for Power, while responding to a question about the energy sector planning, quoted an old Chinese saying: “You cannot discuss the ocean with a well frog.” It was meant to dismiss the criticism by opposition; suggesting that those raising concerns do not see the bigger picture.

But sometimes, the frog inside the well genuinely believes that the well is the ocean, or he might have jumped into a swimming pool considering it as an ocean.

And that is exactly our problem. We think we are planning, seeing the full energy landscape, making data-driven decisions. But the truth is, we are still stuck inside the same old planning mindset. And the well we have mistaken for the ocean is built around one comfortable but misleading idea: the Levelized Cost of Electricity (LCOE).

The IGCEP 2025-35 is an approval away to be submitted to NEPRA and as per publicly available information, it is coming with the same old methodology i.e., LCOE.

Every time a new IGCEP is announced, the hope returns that maybe this time it will be different. That it will be based on better data, sharper modelling, and a more realistic understanding of demand and generation needs. But the same core mistake keeps repeating itself. In this article we shall discuss why this IGCEP is also likely to misguide investment decisions, and what are simplistic workable solutions which can make this document an accurate and reliable planning tool.

LCOE looks neat on paper. One number that tells you the cost of electricity from any project, no matter the technology. It feels scientific. But that is the problem, it only feels that way. In reality, LCOE ignores the most critical parts of our power system. It does not care when electricity is produced, whether it is during peak summer hours or chilly winter nights. It does not care where the plant is located, whether near a load centre or hundreds of kilometres away, where expensive transmission lines will be needed.

And it does not care how the plant contributes, whether it is flexible, dispatchable, seasonal, or completely dependent on weather. It treats all kilowatt-hours as equal, even when the grid does not. On top of that, it hides the real impact of financing terms. A slight change in interest rate can make a project appear cheap or expensive overnight. So, when we say LCOE is misleading, it is not an overstatement, it simply was not designed to answer the kinds of questions we need to ask today. Yet it remains at the centre of our planning documents, giving a false sense of clarity where thoughtful analysis is needed.

One of the biggest flaws in using LCOE is that it ignores the timing of generation. In a country with growing reliance on solar generation, where demand can swing sharply between hours, days, and seasons, timing is everything. A project that generates reliably during peak summer afternoons has a much higher value than one that delivers most of its energy at night or during the winter.

But LCOE treats them the same. It averages everything out without asking when that electricity will be needed. A simple fix exists: instead of counting all units equally, we should assign weights to generation based on monthly or even hourly demand profiles. A unit delivered during peak summer should count more than one in low-demand winter months. This is not complex modelling; it is just a more honest way to value electricity.

Another critical blind spot is location. LCOE does not include the cost of transmitting electricity from the project site to where the demand is. A project with a low LCOE built in a remote valley may end up being far more expensive once transmission lines, grid losses, and system upgrades are added in.

But none of this shows up in the LCOE figure. As a result, such projects look attractive in planning documents, until transmission costs and losses make those savings vanish.

The solution is again straightforward: calculate the delivered cost of electricity, not just the generation cost. If NEPRA can assess the full tariff of each project, including transmission and net delivered energy, during planning, the IGCEP will automatically become more accurate and aligned with system needs.

Then there is the question of function. Not all power plants serve the same role. Some are base-load, designed to run around the clock. Others are peaking plants, called in only during the highest demand hours. Some are flexible, adjusting output based on real-time grid conditions. Comparing all of them on one flat metric like LCOE is fundamentally wrong.

A base-load plant cannot be expected to perform the job of a peaking plant, and vice versa. What is needed is clear segmentation. Projects should be grouped according to their operational role, base-load, load-following, or peaking, and only compared within those categories. This simple adjustment would prevent apples-to-oranges comparisons and allow planners to select the most suitable technology for each type of grid requirement.

LCOE also fails to account for flexibility and system value. A project that can ramp up or down quickly, respond to frequency changes, or provide backup support during grid stress has a quite different value than a project that just injects fixed power into the system.

But LCOE does not see that. It assumes all generation is equal, regardless of system services provided. To fix this, we need to introduce system-adjusted evaluation tools. These do not need to be overly technical, simply basic metrics that reflect a project’s ability to support grid stability. Even assigning flexibility scores or adding a simple adjustment factor in tariff calculations would go a long way in capturing this missing piece.

There is another issue that often goes unnoticed. It is the quality and validity of the data being used to calculate LCOE. For the IGCEP 2025–35, project sponsors were asked to submit their cost and generation details back in 2023. That data, already outdated by now, is still being used to decide which projects are considered “affordable.” In some cases, especially for public sector projects, updated numbers from 2025 have quietly been accepted.

Meanwhile, all other projects, mostly from the private sector, are still being evaluated based on the original data submitted back in 2023. They were not allowed to revisit their assumptions or submit revised costs. Instead, an arbitrary inflation index was applied to update their figures, which does not truly reflect market conditions or financing realities. This selective flexibility creates a clear bias in favour of a few while sidelining others.

The fair and transparent approach would be for NEPRA to first determine tariffs based on verified and current data, across the board. Only then should projects be shortlisted as candidates for inclusion in IGCEP. That is the only way to ensure equal treatment and build confidence in the planning process.

Lastly, LCOE is overly sensitive to financing assumptions, especially the discount rate. A slight change in interest rate can dramatically shift the final number, making one project look cheaper than another purely because of how it is financed, not how it performs. In Pakistan, where financing terms vary widely and risks are high, this creates a distorted picture.

The better approach is to standardise tariff calculations using real, project-specific financing structures. If a project expects concessional financing, reflect that. If it carries risk premiums, add them in. Avoid hiding those differences behind one average discount rate. Only then can we get a real sense of what each project will cost the country.

We do not need to follow the world blindly. We have our own grid realities, demand patterns, financial limitations, and seasonal challenges. The idea that only foreign consultants or global models can guide our planning has already caused enough damage. It is time we start listening to our own experts, people who have worked within this system, who understand the practical issues, and who can offer grounded solutions.

The continued reliance on outdated metrics like LCOE is not just a technical oversight, it is a habit. Even the U.S. Energy Information Administration, in its 2025 Annual Energy Outlook, clearly states that using LCOE individually is not suitable for system planning. If they are acknowledging the limitations, why are we still using it as the backbone of our national planning?

We have the talent, the institutions, and the data to do better. The only thing missing is the willingness to move beyond shortcuts and start planning like a country that actually wants to fix its energy future. It is time to acknowledge that LCOE has served its purpose but is no longer fit to lead our investment planning.

The cost of repeating the same mistake is too high. If we want a power system that is affordable, reliable, and aligned with actual national needs, then our planning methodology must evolve. And that begins by asking more from the models we use, and more from those who design them.

Copyright Business Recorder, 2025