Insoluble dilemma of power forecasts?

As the NTDC (National Transm-ission and Dispatch Company) and Power Division are progressing with the next IGCEP (Indicative Generation Capacity Expansion Plan) iteration, they acknowledged the invalidity of the previous demand forecasts (used in IGCEP 2022-31) before Nepra (National Electric Power Regulatory Authority).

While revisions are essential, the impact on project selection is significant. Incorrect demand projections, leading to the initiation and subsequent halting of investor-backed projects, can severely undermine investor confidence.

This scenario not only affects current ventures but also deters future investments. Therefore, ensuring accuracy in IGCEP’s demand forecasting is crucial for internal efficiency and upholding the country’s reputation as a reliable and stable investment destination in the energy sector.

Pakistan’s electricity sector has been experiencing fluctuations since the 1990s due to flawed planning. Our situation consistently oscillates between being over-capacity and experiencing load shedding because of inaccurate demand forecasts.

The multiple regression analysis technique employed by the NTDC in preparing the IGCEP presents a substantial challenge despite its international recognition. It has failed to effectively predict and manage the electricity demand, highlighting a notable issue with its accuracy and reliability.

Using aggregate GDP as an independent variable can lead to inaccuracies as it does not account for the varying energy intensities of different sectors, overlooking distinct energy consumption patterns across sectors. Additionally, the methodology does not explicitly address the unique energy intensity of each sector.

Furthermore, the current method of forecasting energy needs introduces biases due to its reliance on historical data, which may not fully account for technological advancements, policy changes, and economic structure.

The model’s scope of independent variables may also be limited, overlooking factors like energy efficiency improvements and consumer behaviour. Additionally, the approach may not be flexible enough to account for economic fluctuations and external shocks.

A detailed strategy is required to improve energy demand forecasting accuracy, considering sector-specific GDP contributions, energy intensities, demographics, and other variables influencing future consumption trends. Scenario analysis should also be incorporated to accommodate future technological, policy, and global economic changes.

Energy intensity is the ratio of a sector’s energy consumption to its GDP share, indicating its energy consumption relative to its economic output. A higher ratio indicates a sector is more energy-intensive, while a lower ratio suggests greater energy efficiency, where the sector uses less energy for its shared output.

To calculate the energy intensity of each sector, the author employed a method that leverages the comparative analysis of two key metrics: the energy consumption and the GDP contribution of each sector (instead of using projected GDP linearly).

Initially, the total energy consumption, represented in Gigawatt-hours (GWh), was compiled for various sectors, such as agricultural, commercial, and industrial, for each fiscal year. This data provided a clear picture of the absolute energy use across different sectors of the economy.

The results of analysing the agricultural sector are particularly revealing. The data for the fiscal year 2013-14 shows that the agricultural sector accounted for 24.29% of the total energy consumption (agricultural, industrial, and commercial consumption) while contributing 24.87% to the GDP.

This near parity in energy consumption and economic output indicates a balanced energy intensity, with an energy intensity factor close to 1 (0.98), suggesting a relatively efficient energy use of economic production.

The agricultural sector’s energy intensity has maintained a balance over the years, averaging precisely 1 with a standard deviation of 0.05 for nine yearly observations. This minor deviation can be further calibrated by investigating sub-sectors.

The analysis of the industrial sector presents a different yet significant picture. In the fiscal year 2013-14, the industrial sector’s energy consumption was notably high, accounting for 61.38% of the total energy use (total of agricultural, industrial, and commercial consumption), while its contribution to the GDP was only 20.98%.

This disparity is reflected in the energy intensity factor of 2.93, indicating a high level of energy intensity. In other words, the industrial sector consumes a disproportionately large share of energy relative to its economic output.

The industrial sector’s energy intensity has averaged 3.08, with a standard deviation of 0.1 for nine yearly observations. This deviation can be further calibrated by investigating sub-sectors.

The commercial sector’s energy consumption patterns also offer valuable insights. In the 2013-14 fiscal year, the sector consumed 14.33% of the total energy, while its contribution to the GDP was substantially higher at 54.15%.

This results in an energy intensity factor of 0.26, indicating a relatively low energy intensity compared to its economic output. The commercial sector’s energy intensity has averaged 0.28, with a standard deviation of 0.02 for nine yearly observations.

These contrasting patterns across sectors underline the importance of a nuanced approach to energy policy and economic planning. A one-size-fits-all strategy will not suffice. Instead, sector-specific strategies are needed, tailored to address each sector’s unique challenges and opportunities.

The risks are manifold if we continue to rely on the current forecasting methodology. We face a future where discussions about overcapacity or undercapacity will resurface, leading to economic inefficiencies like overbilling due to higher capacity payments or the burden of load shedding on consumers. In addition to a sector-specific analysis, the overall demand scenario should be based on actual data rather than subjective opinions and hypothetical probabilities.

During recent public hearings by NEPRA, SOLRs (suppliers of last resort) and NTDC informed the regulator that energy demand is projected to stay low for the coming decade.

Contrarily, official publications paint a different picture, indicating an expected increase in demand. Since the fiscal year 2013-14, energy demand has been on a noticeable upward trajectory.

In particular, the domestic sector has been a significant contributor, with its consumption increasing from 33,342 GWh in 2013-14 to 52,408 GWh by 2022-23 (CAGR of 5.82%), consistently forming a substantial portion of the total energy consumption (~50% of total consumption).

The combined total of the agricultural, commercial, and industrial sectors also shows a pronounced growth trend, rising from a collective consumption of 29,471 GWh in 2013-14 to 46,046 GWh in 2022-23 (CAGR of 4.7%).

One may argue that the CAGR of the last nine years may be misleading; contrarily, the CAGR during the previous two years is 7.82%, while the CAGR of agricultural, industrial, and commercial (combined) is 11.43%.

This analysis highlights the urgent need to revise energy demand forecasting methodologies in light of the disparity between sector-specific realities and generalized projections. Policymakers and stakeholders must adopt a more nuanced, data-driven approach, tailoring the IGCEP model to align with the diverse demands of each sector.

This strategic shift in the IGCEP is necessary to reflect real-world data and national ambitions, is critical for economic stability and sustainable growth. Embracing this sector-specific approach is essential not only for bolstering investor confidence but also for guiding country’s energy sector towards a more efficient and robust future.

Copyright Business Recorder, 2023