A host of macroeconomic factors affect electricity demand apart from economic growth, development and urbanisation. These include climatic conditions, technological disruptions, consumer preferences, availability of alternative energy sources and state-specific policies. For instance, while policies and schemes such as Make in India, the dedicated freight corridor and Power for All are likely to increase electricity consumption, initiatives such as the rooftop solar programme, the Perform, Achieve and Trade (PAT) scheme, Bachat Lamp Yojana, star labellingprogramme, and solarisation of pumpsets are likely to reduce the demand for grid-connected power. In addition, the introduction of e-mobility and greater dependence on renewable energy will change the way power is generated and consumed. It is, therefore, essential to assess the future electricity requirements in a periodic manner.
The estimation of future electricity demand is also crucial for planning the generation, transmission and distribution infrastructure required for economic growth. The Central Electricity Authority (CEA) conducts a periodic electric power survey (EPS) to assess electricity demand across the country on medium- and long-term basis. So far, 19 such surveys have been conducted. In January 2017, the CEA carried out the 19th EPS for assessing electricity demand from 2016-17 to 2036-37 using the partial end-use method (PEUM). PEUM, a combination of time series analysis and end use method, is a bottom-up approach focused on the end use of different consumer categories.
The CEA carried out demand projections for the period using the econometric method based on parameters such as GDP, real electricity price, weather parameters and historical electricity consumption. The CEA recently released the projections in a report titled, “Long Term Electricity Demand Forecasting”.
Data set and model used
The data set for econometric analysis covers key electricity demand drivers for all the states and union territories (UTs) from 2002-03 to 2015-16. The data set comprises the dependent variable, that is, electricity demand/requirement, and a set of independent variables such as state GDP and weather data for all the states and UTs, except Himachal Pradesh, Arunachal Pradesh, Dadra & Nagar Haveli, Sikkim, Daman & Diu, Andaman & Nicobar Islands and Lakshadweep. For these states/UTs, as weather data (temperature, rainfall, etc.) was not consistently available, it was assumed that the growth rate of electricity demand will converge with the national growth rate in the future. The monthly data of 25 states and three UTs spanning over 168 periods (14 years of monthly data) has been used.
In the current exercise, the CEA has used the partial adjustment model (PAM) and the seemingly unrelated regression (SUR) model for electricity demand forecasting under three scenarios – business as usual (BAU), optimistic and pessimistic. The PAM model estimates electricity demand within the regional panel framework, which assumes that all the states within a region will have the same response to key socio-economic variables included in the model. Thus, to estimate the differential response of each state with respect to change in key drivers, the regional SUR model is used, which takes advantage of the panel data to improve the overall efficiency of state-level parameter estimates.
The BAU case assumes that GDP at the all-India level will continue to grow at an average compound annual growth rate (CAGR) of about 7.3 per cent (as recorded during 2000-01 to 2017-18) and there will be no significant deviations from these past trends. In the optimistic growth scenario, the all-India GDP is assumed to grow at 8 per cent during 2018-19 to 2036-37. In the pessimistic growth scenario, the all-India GDP is assumed to grow at 6.5 per cent during the forecast period.
The all-India electricity requirement based on the regional PAM model is projected to grow from 1,152.4 BUs in 2016-17 to 2,976 BUs in the BAU scenario at a CAGR of 4.86 per cent. The electricity demand is projected to grow to 3,175 BUs in the optimistic scenario, recording a CAGR of 5.2 per cent, while it is likely to grow to 2,691 BUs in the pessimistic scenario at a CAGR of 4.33 per cent (see Table 1). In the three scenarios, the electricity requirement is expected to increase by 2.33 times to 2.75 times between 2016-17 and 2036-37. Meanwhile the peak electricity demand is expected to grow from 158.99 GW in 2016-17 to 359.88 GW (pessimistic scenario), 398 GW (BAU) and 427.5 GW (optimistic scenario).
As per the SUR model, the all-India electricity requirement is projected to increase at a CAGR of 5.58 per cent for the period 2016-17 to 2036-37 from 1,188.2 BUs to 3,517 BUs in the BAU scenario (see Table 2). In the pessimistic scenario, the electricity requirement is expected to grow at a CAGR of 4.86 per cent to 3,066.8 BUs. Meanwhile, in the optimistic scenario, it is expected to grow at a CAGR of 6.09 per cent to reach 3,878 BUs. Under the three scenarios, the electricity requirement is likely to increase by 2.58 times and 3.26 times as compared to the baseline. Regarding peak electricity demand is estimated to grow from 163.15 GW in 2016-17 to 482.95 GW in the BAU scenario, to 532.5 GW in the optimistic scenario, and to 421.08 GW in the pessimistic scenario.
A look at the state-wise electricity requirement as per the SUR model shows that electricity demand in Rajasthan, Haryana, Andhra Pradesh, Telangana, Tamil Nadu, Bihar, West Bengal, Assam, Meghalaya and Tripura will grow at a rate higher than the all-India CAGR of 5.58 per cent in the BAU scenario. In terms of regions, the north-eastern region is expected to witness the highest growth rate during 2016-17 and 2036-37. In absolute terms, the northern region will account for the highest energy requirement at 1,088 BUs or one-third of all-India demand in the BAU scenario.
Energy requirement from 2016-17 to 2036-37 will grow at a CAGR of 5 per cent under the econometric method as well as under PEUM, indicating that the demand derived by the two methods is in a similar range. A periodic assessment and revision of long-term electricity demand estimates is crucial for stakeholders across sectors. The CEA’s latest electricity demand forecasts are expected to play a key role in guiding policymakers, infrastructure planners and industry players to optimally plan capacities in the coming years.