Enhancing Energy Efficiency of Indian Manufacturing Enterprises: Role of Energy Prices and Firms’ Social and Environmental Consciousness

Abstract

The article examines whether energy consumption and energy intensity of manufacturing enterprises in India are significantly impacted by energy prices, also studying the impact of non-price factors. Data on industrial plants and manufacturing companies are used for two components of the analysis: one addressing the impact of energy prices on energy demand, and the other addressing the impact of non-price factors on energy intensity. Although several studies have reported moderate-to-low price elasticities of industrial power demand in India (ranging from –0.2 to –0.6), we estimate that the price elasticity of energy demand in India’s energy-intensive manufacturing industries is about –1.3. The results indicate that export-oriented manufacturing firms in India have lower energy intensity than non-exporting firms; this finding aligns with similar earlier studies. However, this impact is contingent on firms’ environmental consciousness and commitment.

JEL Classification: D21, F18, L60, M14, Q40, Q41, Q42, Q54

Keywords

Energy demand price elasticity Indian manufacturing exporting corporate social responsibility

Introduction

At COP 26 in 2021, India substantially enhanced its commitments to reduce the carbon intensity of its economy (by 45% by 2030 compared to the 2005 level) and contain the country’s carbon emissions. Indian manufacturing must play a crucial role in meeting India’s commitments to contain the country’s CO₂ emissions by 2030. India’s manufacturing sector accounts for a substantial part of final energy consumption in India. The manufacturing sector accounts for about 25% of India’s aggregate CO₂ emissions and makes a significant indirect contribution through the use of power provided by power generation units (Kant et al., 2022).

Reduction in industry-level energy intensity has been the prime factor behind the decline in the aggregate CO₂ emissions intensity of India’s organised manufacturing in the past (Kant et al., 2022), and such a downward trend in energy intensity of Indian manufacturing needs to be maintained and sped up to realise India’s commitments at COP 26. There are grounds to believe that the energy price policy can play a critical role in persuading industrial firms to decrease their energy intensity, which will, in turn, reduce the CO₂ emissions intensity of Indian manufacturing. Needless to say, a carbon tax or a carbon emissions trading scheme tends to raise the cost of energy vis-à-vis other inputs to an industrial firm, and such an action will therefore tend to reduce energy intensity and carbon intensity (see, in this context, Sen and Sahoo, 2024). Apart from energy prices, non-price factors could play a decisive role in promoting industrial firms’ energy efficiency.

The aim of this article is to identify and evaluate some factors that could significantly improve energy efficiency in Indian manufacturing and thus help contain carbon emissions. First, we consider the role of energy prices. Then, we examine the role that non-price factors, particularly industrial firms’ social and environmental consciousness (S&EC), may play in improving energy efficiency.

The rest of the article is organised as follows. The second section presents a brief review of the literature. The third section discusses the data used in the analysis and the methodology employed to assess the impact of energy prices and non-price factors. The following two sections present the empirical results. The fourth section examines the effect of energy prices on energy use in industrial plants. The fifth section presents and discusses the results of an analysis of the impact of non-price factors on the energy intensity of manufacturing firms, particularly how industrial firms’ environmental consciousness could enhance energy efficiency. The sixth section discusses the study’s main findings and their implications. Finally, the seventh section concludes the article.

Literature Review and Research Questions

Price Elasticity of Energy Demand

A key research question the article addresses is whether energy demand in Indian manufacturing plants is sufficiently responsive to price changes. If the energy demand in manufacturing plants is not price responsive or is only mildly price responsive, an enhanced energy price (or carbon price) for Indian manufacturing plants implemented through a tax on energy use, or a carbon tax, or the operations of a carbon market in India will not successfully reach its desired goal of significantly reducing CO₂ emissions.

Next, let us consider some available estimates of the price elasticity of industrial energy demand in India. Very few studies have been done on this subject. The estimates of the price elasticity of electricity demand in Indian industries presented by Bose and Shukla (1999) range from –0.04 (LT consumers) to –0.45 (HT consumers). For the total industry sector without captive generator sets, the price elasticity of electricity demand is –0.32. From the price elasticity estimates of Bose and Shukla (1999), it appears that industrial demand for power does not respond much to increases in electricity rates. According to Tran et al. (2023), the price elasticity of electricity demand by industry in India is about –0.63, higher than the estimate by Bose and Shukla (1999). However, the price elasticity estimate of Tran and associates is also not high—it may be regarded as moderate.¹

Dasgupta et al. (2017) have estimated the price elasticity of energy demand (considering all energy items such as coal, oil and electricity) in seven Indian industries; these range from –0.3 for iron and steel and cement to –1.17 for non-ferrous metals and –2.3 for chemicals and chemical products excluding fertilisers. The price elasticity estimate for textiles is positive (wrongly signed), signifying low price elasticity. The estimated price elasticity is above 1 in absolute value in three of seven cases. The mean of the seven estimates of price elasticity is –0.95. Thus, for the energy-intensive industries in India taken as a group, it may be said that the price elasticity of energy demand is high at about (–)1, based on the analysis undertaken by Dasgupta et al. (2017).

The results of Dasgupta et al. (2017) indicate that energy demand in Indian manufacturing adequately responds to changes in energy prices. However, several pieces of empirical evidence point towards low price elasticity of aggregate demand for electricity in India. Therefore, doubts arise as to whether it is safe to assume that industrial energy demand responds sufficiently to price changes.

Tiwari and Menegaki (2019) find that the price elasticity of demand for electricity in India is –0.21. The estimate of price elasticity is low. According to a report prepared by the Central Electricity Authority (CEA, 2019, p. 54), a 1% increase in electricity price lowers electricity demand in India by 0.2% in the short run and 0.6% in the long run. These estimates of price elasticity are low to moderate.

Is it possible that the finding of a low price elasticity of aggregate electricity demand in India arises from a very low price elasticity of household (residential) demand for electricity, offsetting a relatively high price elasticity of industrial electricity demand? This is not true. The price elasticity estimate of household demand for electricity in India has been reported as –0.3 to –0.4 by Harish et al. (2014). In the estimates of the price elasticity of electricity demand in India by Bose and Shukla (1999), the estimate for households is –0.65, while that for industry is in the range of –0.04 to –0.32, lower than the estimate for households. In the estimates of Tran et al. (2023) for India, the price elasticity at the aggregate level is –0.66, and those for the residential and industrial sectors are –0.83 and –0.63, respectively.

Let us now turn to international literature and consider some estimates of the price elasticity of energy and electricity demand in other countries. Dussaux (2020) uses data on about 8,000 French manufacturing firms for 2001–2016 to study the effect of energy price hikes on energy use, carbon emissions, output and employment, and finds that the price elasticity of energy demand in French manufacturing is about –0.6. Bjørner et al. (2001) have presented estimates of price elasticity of demand for electricity in Danish industrial companies—the estimated elasticity is about –0.4. Bjorner and Togeby (1999) presented estimates of price elasticity for both electricity demand and aggregate energy demand for Danish industrial companies—the price elasticity of electricity demand is about –0.6, and that for aggregate energy is about –0.55.

Addey et al. (2019) have used a balanced panel data set for 48 contiguous US states, covering the period 1970–2015, and find that the short-run price elasticity of industrial energy demand is –0.39 and the long-run price elasticity is –0.45. These estimates for the US may be regarded as showing moderate price elasticity.

Phoumin and Kimura (2014) have estimated the price and income elasticity of demand for energy in a set of selected ASEAN and East Asian countries using time series data. They have included India in their analysis. The period covered is 1990–2010. The estimated price elasticity of energy demand for the sample of countries covered in the study is generally very low. In most cases, the estimate is statistically insignificant. The estimates of long-run price elasticity for India are (–)0.04 for total primary energy consumption and (–)0.16 for total final oil consumption. The long-run price elasticity estimate for total final energy consumption is positive (incorrectly signed), indicating a low impact of energy prices on final energy demand.

Some studies have found a significant impact of electricity price hikes on electricity demand. For example, Csereklyei (2020) finds that the long-run price elasticity of demand for electricity in industries in the EU is between –0.75 and –1.01. In a study of electricity demand among manufacturing firms in South Korea, Kim et al. (2019) found that the price elasticity of demand is about –0.9. However, several other studies have found a low to moderate price elasticity of electricity demand. Chang et al. (2014) studied electricity demand in South Korea, and their price elasticity estimates are –0.42 for residential and –0.48 for industrial demand. Liddle and Hasanov (2024) have studied electricity demand using panel data for 31 middle-income countries, including India. They found that the price elasticity is, on average, –0.09. The price elasticity estimate for India is positive (i.e., incorrectly signed), indicating low price elasticity.

To summarise the above discussion, the estimate of price elasticity of energy demand in Indian manufacturing presented in Dasgupta et al. (2017) is about (–)1. However, because (a) the estimates of price elasticity of electricity demand in India made in several studies are low to moderate, and (b) the price elasticity estimates for electricity and energy reported in the international literature on this topic are generally low to moderate, doubts arise on whether it is safe to assume that energy demand of Indian manufacturing is adequately price elastic.

The abovementioned issue requires careful consideration to assess the impact of a carbon tax or a carbon-market-generated carbon price in India on the energy use and CO₂ emissions of Indian manufacturing. Micro-evidence on the price responsiveness of industrial plants’ energy use will help clarify the issue raised. The present article makes an attempt in this direction. Estimates of the price elasticity of energy demand in energy-intensive manufacturing industries (EIMIs) in India are presented using a panel data set of manufacturing plants belonging to EIMIs.

Impact of Non-price Factors on Industrial Energy Demand

Besides energy prices, the extent of energy use in industrial firms and their level of energy intensity (and carbon intensity) are impacted by non-price factors, which include technology, international trade (exports and imports) orientation, and government regulation² and firms’ voluntary measures³ to improve environmental performance. Several studies on energy use, energy intensity and carbon intensity of Indian industrial firms have assessed the impact of non-price factors (see, e.g., Azim & Sawhney, 2025; Bagchi et al., 2022; Goldar & Goldar, 2023, 2024; Kumar et al., 2022; Martínez-Zarzoso et al., 2020; Sahu et al., 2021; Sharma & Padhi, 2024; Soni et al., 2017).

There have been a large number of econometric studies on the determinants of energy intensity and CO₂ emissions intensity of industrial firms. Several studies have reported that exporting leads to reduced intensity of energy use and CO₂ emissions (see, e.g., Batrakova & Davies, 2012; Goldar & Goldar, 2023; Holladay, 2016; Jinji & Sakamoto, 2015; Roy & Yasar, 2015; Tran, 2022). Why does exporting lead to a lowering of energy intensity and CO₂ emissions intensity? According to the literature, the explanation lies in firms being induced to upgrade technology, employ better management practices, adopt green technologies and invest in pollution abatement (see Banerjee et al., 2021; Forslid et al., 2018; He & Huang, 2021; Siedschlag & Yan, 2020; Najjar & Cherniwchan, 2021, among others). The fact that international orientation exposes firms to global best practices in environmental management might be another causal factor.

Our contention in the second segment of analysis in the present article is that exporting does not have an unconditional energy-efficiency-enhancing effect for all industrial firms. It depends on the nature of the firm. In particular, the beneficial effect of exporting on energy intensity (and carbon emissions intensity) will be greater among firms that are relatively more socially and environmentally conscious than among other firms. The underlying logic is that both economic and psychological incentives impact energy demand in a firm (see Château, 2022). For a socially and environmentally conscious firm, exporting will have a greater impact on energy intensity, tending to reduce it, because the economic forces that lead to reduced energy consumption and carbon emissions are reinforced by psychological factors working in tandem.

To our knowledge, the international literature on the relationship between trade orientation and energy intensity (or CO₂ emissions intensity) has paid little attention to whether firms’ environmental consciousness conditions the impact. The same holds for such studies undertaken for Indian manufacturing enterprises (e.g., Azim & Sawhney, 2025; Goldar & Goldar, 2023). Research similar to the present study has been conducted by Nguyen and Adomako (2022), who investigated the relationship between international orientation and the environmental performance of small- and medium-sized enterprises in Vietnam engaged in exporting, finding that environmental commitment mediates this relationship.

Data and Econometric Methodology

The analysis of the impact of energy prices in the fourth section is based on plant-level panel data of the Annual Survey of Industries (ASI).⁴ This data source provides detailed data on various energy items consumed by industrial plants, enabling an estimate of energy consumption. A physical measurement of energy consumption is used to compute energy intensity. The panel character of the data set offers some advantages in econometric modelling. This is why several studies have used ASI plant-level data to study energy intensity and CO₂ emissions intensity in Indian manufacturing (e.g., Goldar & Goldar, 2023, 2024; Sharma & Padhi, 2024). The fifth section uses firm-level data drawn from the Prowess database of the Centre for Monitoring Indian Economy (CMIE) to study the impact of non-price factors. There have been several studies on energy intensity among Indian manufacturing firms using the Prowess database (e.g., Sahu et al., 2021; Soni et al., 2017). The Prowess database is used for the analysis in the fifth section rather than ASI unit-level data because we use expenditure on corporate social responsibility (CSR) as a measure of industrial firms’ S&EC, and CSR spending data are not available in ASI unit-level data.

Data, Variables and Econometric Methodology for Plant-level Analysis

Data and Key Variables

Unit-level (i.e., plant level or factory level) panel data of ASI for 2008–2009 to 2018–2019 are used for the analysis. ASI covers India’s organised sector manufacturing plants. Six two-digit industries, according to the National Industrial Classification (NIC), 2008, are included in the study. These industries rank relatively high in energy-output ratio among the various two-digit manufacturing industries. The industries chosen are: (a) textiles, (b) paper and paper products, (c) coke and refined petroleum products, (d) chemicals and chemical products (including fertilisers), (e) other non-metallic mineral products (including cement), and (f) basic metals (including ferrous and non-ferrous metals). Out of the total energy consumption in India’s organised manufacturing in 2018–2019, the shares of these industries are depicted in Figure 1. Industrial plants in these six industries account for most of the energy consumption (and hence CO₂ emissions) in Indian manufacturing.

Figure 1.

Source: Authors’ computations based on Annual Survey of Industries (ASI) data.

The ASI data set contains information on the consumption of different energy items by industrial plants: coal, electricity, petroleum products, natural gas and a miscellaneous category, others, which is treated as fuel wood. The energy consumption across different industrial plants and years, measured in tons of oil equivalent (TOE), has been computed largely as in Goldar and Goldar (2023).⁵

How price indices for energy items have been formed is discussed next. For coal, electricity and natural gas, the unit value (i.e., price or rate) of purchase is available in the ASI data set, which has been used. For petroleum products, the price of fuel oil is taken at the all-India level and applied to all plants.⁶ And the same procedure applies to fuel wood. To form the price index for petroleum products and fuel wood, the base is taken as 2011–2012 (=1.0). It should be noted that in these cases, there is one common price for all plants. To construct the price indices for coal, natural gas and electricity, the average unit value computed at the all-plant level for 2011–2012 has been taken as 1.00. Thus, the price index for coal, natural gas and electricity is 1.00 for 2011–2012 at the aggregate level. However, the index differs from plant to plant for each year.

How prices of energy items are combined to compute an aggregate energy price index for each plant for each year is discussed in the following sub-section.

Model Specification

To analyse the impact of energy prices on energy demand, the following econometric model has been estimated from panel data on industrial plants; the model specification follows Dussaux (2020):

\ln (Y_{i j t}) =_{i j} +_{j t} + β_{1} {l n (U C E}_{i j t}) + + \sum^{k} X_{i j, t - 1}^{k} + u_{i j t}

(1)

{P e}_{i j t} = \sum_{k} w_{i j 0}^{k} P_{j t}^{k}

(2)

The subscripts i, j and t are for plant, industry and time (year), respectively. UCE denotes the unit cost of energy, which is defined as expenditure on fuels (in ₹) divided by the quantum of energy used in TOE. Y is an outcome variable. Two outcome (or performance) variables (Y) are considered: (a) real output (i.e., the deflated value of output) and (b) the quantum of energy use (in TOE). X is a set of controls taken with a 1-year lag.⁷ The plant effects are incorporated through the term α_ij, and industry-by-year effects through the term φ_jt.

The model in Equations (1) and (2) above has been estimated using the instrumental variable fixed-effects model. UCE _t is instrumented by the energy price indices of the two previous years: Pe_t_–1 and Pe_t_–2. The model has been estimated from plant-level panel data for 2011–2012 to 2018–2019. The variable Pe_ijt in Equation (2) is the energy price faced by plant i of industry j in year t. This variable has been constructed as a weighted average of prices of coal, electricity, natural gas, oil and fuel wood (price indices with base 2011–2012). Out-of-sample weights are used for each plant—the weights used for computing the price index in Equation (2) are the average fuel cost shares (for each plant) during the 3 years, 2008–2009 to 2010–2011. The term w in Equation (2) above denotes weights, that is, cost shares during 2008–2010. For each year, the median price paid by all industrial plants is computed for coal, electricity and natural gas, which is then applied to all plants for which information on the applicable rate/price is not available in ASI data. This specification and estimation method help account for unobserved heterogeneity among plants. Also, any potential endogeneity issue with UCE is addressed.⁸

In the equation above, X denotes a set of other plant-level controls in the regression. The controls used are (a) a dummy variable representing ICT use intensity—it takes the value of one if the ratio of ICT equipment to fixed assets is above 1% and zero otherwise, (b) the ratio of deflated fixed capital stock to employment, representing capital intensity of production, (c) two age-related dummy variables: a dummy variable for plants with ages up to 5 years, and another one for ages 6–10 years (intended to capture the fact that energy efficiency is likely higher in plants of a more recent vintage), and (d) a dummy variable representing size of the plant (it is assigned the value of one if the number of persons employed in the plant does not exceed 300 and zero otherwise). The size variable used represents relatively smaller plants in terms of employment, which may be called small and medium plants. Data on these four explanatory variables have been taken from the ASI unit-level data.

Data, Variables and the Econometric Model for Firm-level Analysis

Basis Data Source

The object of the second segment of the analysis is to analyse the determinants of energy intensity of organised sector manufacturing firms in India, particularly whether the impact of exporting on energy intensity is conditioned by the S&EC of the firms, for which a crude measure based on CSR expenditure is used.

The analysis is carried out using a firm-level panel dataset covering 2011–2012 to 2018–2019 (hereafter written as 2011–2018). Company-level data are drawn from the Prowess database. The analysis is confined to firms with a single plant or multiple plants, but all are located in the same state or union territory (UT). Using information on the location of the firms’ plant(s), it is possible to link changes in a state’s power supply to the energy intensity of firms with plants in that state. Firms located in 24 states and two UTs have been chosen for the analysis, based on the share of the states/UTs in the aggregate level organised manufacturing sector gross value added (GVA) in India in 2018–2019 (using ASI data for 2018–2019).

Measuring Energy Intensity of Firms

Energy intensity is measured by the ratio of the deflated value of power and fuel to the deflated value of gross sales (as, for instance, in Sahu et al., 2021). To deflate sales, the wholesale price indices at the two-digit level of NIC-2008 are used. For all firms belonging to a two-digit industry, the corresponding price index is used. The same method is used to deflate the cost of power and fuel; the industry-level deflator is applied to firms within the industry. In this case, an energy price index has been constructed using the wholesale price indices for coal, petroleum products and electricity, and a price index for natural gas,⁹ with the weights for these energy items for each industry taken from the input–output table for 2007–2008.

Explanatory Variables and Model Specification

The explanatory variables used in the model are (a) the logarithm of deflated value of sales, a measure of firm size, (b) capital intensity, measured by total-assets-to-sales ratio, (c) export intensity, that is, exports to sales ratio, (d) imported materials, stores and spares divided by total materials, stores and spares consumed, measuring import intensity, (e) two dummy variables reflecting the age of the firm, (i) up to 5 years and ages and (ii) between 6 and 10 years, age measured based on the year of incorporation, and (f) the extent of surplus/deficit in meeting peak power demand in the state and (g) the renewable energy (RE) sources share in power supply in the state. The last two variables are based on the state in which the firm’s plant(s) are located (discussed further later). A preliminary analysis indicated a lagged effect of export and import intensity, and RE share variable on energy intensity. Hence, these variables have been lagged (this also helps address the issue of possible endogeneity).

The model used for the econometric analysis is specified as:

{l n E}_{i j s t} = +_{i j s} + t + \sum_{r = 0}^{2}_{r} {R E}_{s, t - r} + D_{s t} + \sum_{p = 1}^{2}_{p} X_{i j s, t - p} + \sum_{p = 1}^{2}_{p} M_{i j s, t - p} + \sum_{k}^{k} W_{i j s t}^{k} + u_{i j s t}

(3)

E donates energy intensity, and RE is the share of RE sources in the power supply. The subscript r takes three values, 0, 1 and 2, to allow for lags. D is the extent of the power supply deficit, shortfall in meeting peak demand (surplus is treated as zero deficit). The subscript i is for firm, j is for industry, s is for state, and t is for time. X denotes export intensity (exports-to-sales ratio), M denotes import intensity, and W denotes a set of control variables, which includes firm size. u is the random error term.

The estimation of the equation specified above has been done by the fixed-effects model. In addition to the variables listed above, a time trend variable is included in the estimated model to capture the influence of other factors that have changed with time. Since firm effects (represented by μ_ijs) are incorporated, industry dummy variables or state dummy variables are not needed.

The hypothesis set out in Section ‘Impact of Non-price Factors on Industrial Energy Demand’ regarding the impact of firms’ S&EC is tested using the model described in Equation (3). For the analysis, we use a proxy for firms’ S&EC. For this purpose, we use the level of expenditure incurred by a firm on CSR relative to expenses incurred by other firms (both normalised by the average net profit over the past 3 years). In India, there is a statutory requirement regarding CSR expenditure. A high level of CSR expenditure in a firm beyond the statutory requirement could thus be interpreted as the firm being relatively more socially responsible and environmentally conscious. We hypothesise that such a firm is likely to be strongly motivated to take action on the energy-saving front. If such a firm takes to exporting, this is likely to have a stronger effect on the firm’s energy efficiency.

Accordingly, the model described in Equation (3) above was first estimated by taking data on firms belonging to manufacturing industries. Then, separate estimates were obtained for the relatively high- and relatively low-CSR-spending firms.

CSR Spending

In India, CSR spending is mandatory, with the amendment made to the Companies Act, 2013, in April 2014. Companies with a net worth of ₹5 billion or more, or a turnover of ₹10 billion or more, or a net profit of ₹50 million or more are required to spend 2% of their average net profit in the previous 3 years on CSR. Our regression analysis using firm-level data has been confined to firms found to be eligible for CSR expenditure during any year from 2014–2015 to 2018–2019. The eligible firms are identified based on the following criteria: a net worth of ₹5 billion or more, sales of ₹10 billion or more, or profits of ₹50 million or more in any year from 2014–2015 to 2018–2019. It should be noted that while the regression analysis is conducted using data for 2011–2018, the firms included in the analysis are identified based on whether they were eligible for CSR spending in any year from 2014–2015 to 2018–2019.

Taking data on CSR expenditure and profits, the ratio of CSR to average net profits in the 3 previous years has been computed for each firm and each year during 2015–2016 to 2023–2024, and a median has been taken for each firm. If the average profits in the previous 3 years were negative for a particular year and for a particular firm, then the CSR ratio is not computed. The firms in the sample are then divided into two classes: those whose median CSR expenditure ratio is more than 2%, and those whose median CSR expenditure ratio is 2% (which is the statutory requirement) or less.

The analysis of the data on CSR indicates that a section of firms had spent on CSR below the statutory requirement of 2% of the average net profit of the previous 3 years. A similar finding of CSR spending below statutory requirements has been reported by Bansal and Rai (2014). The fact that there is variation among firms in the CSR ratio makes it possible to use CSR spending as a useful discriminator. Firms that spend more than the statutory requirement may be considered to have a relatively more robust perception of social responsibility and environmental consciousness.

Power Supply Situation

Over the last two decades, there has been a substantial improvement in India’s power supply. There has been a marked improvement in meeting electricity demand in the country. For the utilities, the deficit between requirement and availability was about 7.85% in 2000–2001 and 11.07% in 2008–2009. Since then, there has been a downward trend, and the deficit fell to 0.51% in 2019–2020 and 0.4% in 2020–2021.

Another significant development worth noting is the increase in RE share in power capacity and generation. In March 2010, the capacity of power generation based on RE sources was approximately 15.5 MW, accounting for about 10% of the total capacity. By March 2020, this capacity had increased to 87,000 MW, representing about 24% of the total power generation capacity in utilities.

To econometrically model the energy intensity of Indian industrial firms, two power-related indicators are used. These are (a) the extent of deficit in meeting peak power demand in the state,¹⁰ and (b) the share of renewable energy sources (RES) in power generation in the state.¹¹

A reduction in the extent of the deficit in meeting peak power demand in a state is expected to result in a more efficient operation of the industrial plants located in the state, hence higher productivity. This also leads to a cut in captive power generation with a decisive shift to the grid, thereby avoiding energy losses associated with self-generation. Both factors lead to a reduction in energy intensity.

As regards the impact of rising RE share in power supply in a state on energy use efficiency of industrial firms in the state, following the empirical findings of Goldar and Goldar (2023), an increase in RE share in the power supply is expected to reduce the energy intensity of the firms located in the state.

Price Responsiveness of Energy Use: Regression Results

The estimates of the econometric model described in Section ‘Data, Variables and Econometric Methodology for Plant-level Analysis’ (comprising Equations 1 and 2) are presented in Table 1 (summary statistics are shown in Table A1 in Annexure). The estimated price elasticity is negative and statistically significant. The price elasticity estimate is about (–)1.3, indicating that a 10% increase in the real energy price leads to a 13% fall in energy consumption. The results in Table 1 suggest that a 10% hike in energy price relative to the price of products produced will lower the output of EIMIs by about 5.4%. Thus, an increase in energy price impacts energy demand through two channels: (a) by lowering the energy intensity of production (input substitution) and (b) by lowering the quantity of output produced.

The model estimates indicate that ICT investment reduces energy intensity—a significant negative coefficient is found. Several studies have found that ICT intensity bears a negative relation with energy intensity (see, e.g., Lahouel et al., 2021).

The coefficient on the plant size dummy variable, representing small and medium-sized plants, is negative, as expected (since larger plants have higher production and energy consumption), and statistically significant.

There are indications from the results in Table 1 that an increase in capital intensity of production is associated with higher use of energy. Another finding is that the newly set-up plants are more energy efficient than older plants, especially plants aged more than 10 years. This is probably caused by the fact that newer plants have technology of more recent vintage.

Our finding that the price elasticity of demand for energy in Indian manufacturing is about (–)1.3 indicates that energy demand in Indian manufacturing is amply responsive to changes in energy prices.

As mentioned earlier, Dussaux (2020) has used data on about 8,000 French manufacturing firms, covering the period 2001–2016, to estimate a model similar to that used for the results reported in Table 1. In the estimates of Dussaux, the coefficient of the price variable in the energy use equation is (–)0.6. Our estimate of the price elasticity of energy demand for Indian manufacturing, based on a similar model, is much bigger in numerical value.

Table 1.

Impact of Energy Price on Performance of Plants of Indian Manufacturing, Energy-intensive Industries, Based on Plant-level Panel Data, Instrumental Variable Fixed-effects Estimator.

Explanatory Variables	Outcome = ln(Energy)	Outcome = ln(Real Output)
ln (unit-value of energy)	–1.294 (0.371)**	–0.538 (0.247)*
Ratio of ICT equipment to fixed assets exceeds 1% (dummy variable) (t – 1)	–0.033 (0.013)*	–0.007 (0.012)
Logarithm of capital–labour ratio (t – 1)	0.025 (0.009)**	0.028 (0.008)**
Plant size (plant employing up to 300 persons) (dummy variable) (t – 1)	–0.197 (0.019)**	–0.214 (0.018)**
Plant age up to 5 years	–0.169 (0.029)**	–0.092 (0.025)**
Plant age 6–10 years	–0.070 (0.017)**	–0.004 (0.015)
Plant effects	Yes	Yes
Industry-by-year effects	Yes	Yes
No. of observations	41,338	40,632
No. of plants	8,747	8,735
R ²	0.26	0.12
Wald chi-square and probability in parentheses	821.4 (0.000)	1446.9 (0.000)

Source: Authors’ computations based on Annual Survey of Industries (ASI) unit-level data for 2011–2012 to 2018–2019.

Notes: (a) As explained in the text, the unit value of energy is instrumented by the plant-specific price index of energy (for the previous 2 years). The analysis is confined to energy-intensive manufacturing industries (EIMIs). (b) The models are estimated from data from 2011–2012 to 2018–2019. The out-of-sample (2008–2009 to 2010–2010) cost shares of different fuels are used for constructing the plant-specific energy price. Robust standard errors are in parentheses. * and ** statistically significant at the 5% and 1% level of significance, respectively.

Since the model estimates have been made by the instrumental variable method applied to panel data, some related diagnostic tests are shown in Table A3 in the Annexure. The results in Panel A of the table indicate there is no under-identification. The test statistic for weak instruments crosses the critical limit at the 15%, not at the 10% level. The over-identification test is not satisfied. When a dummy variable representing captive power generation in the plant (lagged by 1 year) is introduced as an additional instrument, the test results reject the hypothesis of weak instruments for both energy consumption and real output models. These test results are shown in Panel B of Table A3. In this case, the estimate of price elasticity of energy demand is –1.56 (statistically significant).

Exporting, S&EC and Energy Intensity: Regression Results

The regression results using Equation (3) are presented in Table 2—a firm-level analysis is undertaken (summary statistics are shown in Table A2 in Annexure). The period covered is 2011–2012 to 2018–2019 (as in the analysis in Section ‘Price Responsiveness of Energy Use: Regression Results’). Regression (1) is for all firms that have been identified as eligible for payment of CSR any year during 2014–2015 to 2018–2019. Regression (2) is for the relatively high CSR spending firms among the eligible firms (over 2% of average net profits, as explained above), and Regression (3) is for the relatively low CSR spending firms (equal to or less than 2%). Robust standard errors are computed, which are clustered at the firm level. t-values are shown in the table.

The results indicate a negative relationship between firm size and energy intensity. Energy intensity falls with increases in firm size. The results suggest that younger firms have lower energy intensity, which seems to be attributable to their having plants and machinery of more recent vintage. A positive relationship between the asset-to-sales ratio and energy intensity is observed, suggesting that capital and energy are complementary. Our results in Regression (1) indicate that the use of imported materials tends to reduce energy intensity, as expected (see Azim & Sawhney, 2025; Imbruno & Ketterer, 2018).¹² This is, however, not found in Regressions (2) and (3).

The regression results indicate that an increase in export intensity lowers the energy intensity of industrial firms. This finding is consistent with the findings of Sahu et al. (2021), Goldar and Goldar (2023) and Azim and Sawhney (2025), as well as with the international literature.

Regression results indicate that lowering the deficit in meeting peak power demand reduces the energy intensity (improving energy use efficiency) of manufacturing firms. Such an effect of improvement in the power supply situation is expected. Another finding is that an increase in the RE share in the power supply tends to reduce the energy intensity of manufacturing firms. The effect seems to occur with a lag.

Table 2.

Determinants of Energy Intensity, Indian Manufacturing Firms, Regression Results, Period: 2011–2018, Fixed-effects Model.

(Dependent Variables: Logarithm of Energy Intensity)
Explanatory Variables	All Firms	Relatively High CSR Spending Firms	Relatively Low CSR Spending Firms
Explanatory Variables	Regression 1	Regression 2	Regression 3
Firm size	–0.289 (–7.10)***	–0.322 (–5.60)***	–0.334 (–6.80)***
Capital intensity: ln(assets/sales)	0.165 (5.14)***	0.161 (3.23)***	0.149 (3.26)***
Age, up to 5 years	–0.199 (–2.48)**	–0.263 (–2.21)**	–0.004 (–0.04)
Age, 6–10 years	–0.032 (–0.90)	–0.033 (–0.87)	–0.007 (–0.11)
Export intensity (t – 1)	–0.084 (–2.65)***	–0.109 (–2.05)**	–0.033 (–0.89)
Export intensity (t – 2)	0.020 (0.48)	0.067 (0.87)	0.010 (0.23)
Share of imported materials (t – 1)	–0.050 (–1.70)*	–0.005 (–0.11)	–0.050 (–1.24)
Share of imported materials (t – 2)	0.038 (1.44)	0.121 (3.21)***	–0.017 (–0.43)
Surplus/deficit in meeting peak power demand (%)	–0.007 (–5.88)***	–0.009 (–5.77)***	–0.009 (–4.26)***
RE share in the power supply	0.102 (0.95)	0.167 (0.86)	–0.006 (–0.05)
RE share in the power supply (t – 1)	–0.402 (–2.94)***	–0.538 (–2.15)**	0.228(–1.76)*
RE share in the power supply (t – 2)	–0.201 (–2.30)**	–0.241 (–1.68)*	–0.077 (–0.73)
Time (year)	0.017 (3.99)***	0.024 (4.38)***	0.020 (3.30)***
R²: within	0.17	0.11	0.15
R²: between	0.03	0.01	0.04
R²: overall	0.04	0.01	0.05
F-value and prob.	21.4 (0.000)	13.1 (0.000)	11.2 (0.000)
No. of observations	11,572	5,604	4,405
No. of firms	1,813	846	701

Source: Authors’ computations based on the Prowess database (CMIE).

Notes: Robust standard errors. t-values in parentheses. *, ** and *** statistically significant at the 10%, 5% and 1% levels, respectively. Regressions (2)–(3) are confined to firms for which data on corporate social responsibility (CSR) could be obtained. Hence, the sum of the number of observations in Regression (2) and (3) is less than that in Regression (1). RE: Renewable energy.

The coefficient of the export intensity variable is negative and statistically significant in Regression (2) but not in Regression (3). This suggests that the energy-efficiency-increasing effect of exports is relatively greater in firms that are relatively more socially and environmentally conscious, as hypothesised. Thus, some support is found for the hypothesis advanced above that the impact of exporting on energy intensity is conditioned by the S&EC of firms as reflected in their CSR spending.

As mentioned above, a large number of studies have concluded that exporting is associated with reduced energy intensity and CO₂ emissions intensity or emissions of other gases (Batrakova & Davies, 2012; Cheng et al., 2021; Cui et al., 2016; Forslid et al., 2018; Goldar & Goldar, 2023; Jinji & Sakamoto, 2015; Roy & Yasar, 2015; Tran, 2021). Understanding why exporting induces firms to reduce energy consumption and CO₂ emissions is vital. It may be argued that, compared to non-exporters, exporters are more likely to adopt ‘greener’ technologies and invest in pollution abatement (see, e.g., Banerjee et al., 2021; He & Huang, 2021, among others). From the results of the analysis presented in this article, it appears that the environmental consciousness and commitment of firms might play a mediating role in this relationship between exports and environmental performance, as indicated by the findings of Nguyen and Adomako (2022).

Tests of serial correlation and heteroscedasticity for Regression (1) in Table 2 are presented in Table A4 in the Annexure. The results indicate the presence of both serial correlation and heteroscedasticity. To address the issue of serial correlation, a dynamic model has been estimated using the system generalised method of moments (GMM) estimator (see Table A5 in Annexure). The results show that the core findings remain intact when the estimation method is changed—a negative and statistically significant coefficient is found for export intensity, import intensity and the share of RE in power supply.

Discussion on the Main Empirical Findings and Their Implications

The manufacturing sector accounts for a dominant part of the energy consumption in India and is responsible for a large part of India’s CO₂ emissions. A study of the energy intensity of organised sector manufacturing firms and what factors influence the energy intensity in such firms, therefore, has significance from the viewpoint of policy formulation for the containment of CO₂ emissions, particularly in the context of India’s commitments at COP 26. We considered in the article the role of price and some non-price factors in containing energy consumption, and hence carbon emissions.

To study the impact of energy price, we estimated the price elasticity of energy demand in India’s EIMIs using plant-level panel data from the ASI for 2008–2009 to 2018–2019. Our econometric results indicate that the energy demand within the EIMIs has a price elasticity of about (–)1.3 (alternate estimate –1.6).

Our estimates of price elasticity demand for energy in India’s EIMIs are relatively higher than the price elasticity estimates for power demand in India reported in several earlier studies. From a perusal of the relevant international literature, one would get the impression that the price elasticity of demand for energy and demand for electricity in the industrial sector is low to moderate. By comparison, our estimates for Indian manufacturing are relatively high.

It is challenging to adequately explain in a limited space why our estimate of the price elasticity of energy demand in industries is significantly higher than the estimates commonly obtained in previous studies. To provide a satisfactory explanation, we need to examine the nature of data used in various studies and the model specification employed. The fact that we have instrumented the price variable could be a possible reason. Also, the fact that in our model, following Dussaux (2020), we allow an indirect effect of energy price on energy consumption through the adverse impact on output is a contributory factor.

The estimates of price elasticity (absolute value) of energy demand in industrial firms based on a fixed effects model applied to panel data are 0.6 in the study of French industrial firms by Dussaux (2020) and 0.55 in the study of Danish industrial firms by Bjorner and Togeby (1999). One possible reason why our estimate of price elasticity for Indian manufacturing firms is higher is that we have confined our analysis to energy-intensive industries. The share of energy in aggregate cost is relatively higher in these industries, and therefore, the price elasticity of demand for energy input is expected to be relatively higher.

To turn now to some general issues, the energy transition issues and carbon pricing in India are receiving much attention. The current thinking that a carbon tax in India or a market-determined carbon price will help in curtailing India’s carbon emissions depends crucially on the assumption that industrial energy demand is sufficiently responsive to energy prices. Our results in this study indicate that the price elasticity of energy demand is sufficiently high.

The analysis of the impact of some non-price factors was based on company-level data. The effects of export orientation, CSR and the proportion of energy derived from renewable sources in various states were assessed. It may be pointed out that India’s clean energy policy sees both sets of policies, that is, RE and energy efficiency, working in tandem to achieve the country’s climate goals. Evidence of this complementarity is found in the results obtained in the article.

The analysis based on company-level data in the article shows that exporting reduces energy intensity among Indian manufacturing firms, echoing the findings of several earlier studies, including such studies done previously for Indian manufacturing. However, from the analysis presented here, an interesting possibility that emerges is that how clean exporters are is conditioned by their environmental consciousness and commitment.

Before concluding this section, some forward-looking observations may be made for the energy scenario in India. Looking at the future, the prevalence of complementarity between RE and energy efficiency becomes uncertain, barring certain cases where high temperatures are required in many heavy industrial processes, which would necessitate combusting fossil fuels. For the rest, it remains to be seen whether, with the influx of relatively cheap and abundant RE-based power, a similar impetus towards industrial energy efficiency would be maintained absent a concerted policy thrust given across the board. Or, will the drive for green energy itself propel industrial firms to energy efficiency through imitation and a sense of citizenship (using the concepts as in Château, 2022)?

With the recently notified Energy Storage Obligations (ESO) Policy aimed at further RE deployment, it is believed that future moves towards industrial energy efficiency would primarily be on the backs of concerted drives for boosting demand for cleaner products with lower emissions/energy embedded. The latter may be done through green public procurement drives by the government and standardisation and labelling of green products, inducing private purchase decisions.

Conclusion

Energy consumption and energy intensity in industrial firms are influenced by both energy prices and non-price factors. We studied both aspects in the context of India’s organised manufacturing using two data sets, a plant-level panel data set and a firm-level panel data set. Econometric models were estimated using data for 2011–2012 to 2018–2019. We found that energy demand in manufacturing plants belonging to six energy-intensive industries is sufficiently price elastic. Our estimate of the price elasticity of energy demand in Indian manufacturing is (–)1.3. In contrast, most of the estimates of price elasticity of demand for electricity and aggregate energy demand in the industrial sector reported in previous studies are about 0.6 or less (in absolute value). We found that export and import orientation tend to lower energy intensity in Indian industrial enterprises, confirming the findings of several earlier studies, including some undertaken for Indian manufacturing. We also found that an increase in the RE share in power generation is associated with a reduction in the energy intensity of Indian industrial firms.

Drawing on the discussion in the studies by Château (2022) and Nguyen and Adomako (2022), we hypothesised in the article that the S&EC of the firms moderates the impact of export intensity on the energy intensity of industrial firms. We found some econometric evidence in support of this hypothesis, using the level of CSR expenditure of industrial firms as an indicator of their S&EC. This issue needs further investigation, perhaps using some other indicators of firms’ S&EC.

We could not adequately explain why our estimate of the price elasticity of energy demand in Indian organised manufacturing is significantly higher than those commonly reported in previous studies, including those on industrial power demand in India. Investigation of this divergence in price elasticity estimates could be a topic for future research. It would be helpful to examine whether the difference is primarily caused by the model specification employed. Since several earlier studies have reported low price elasticity of industrial power use in India, which has important policy implications, new research based on more recent data and alternate model specification could shed light on the question of whether the industrial demand for power is indeed low, as earlier studies suggest, or whether it is actually moderate to high.

Footnotes

Acknowledgement

The authors have greatly benefited from the comments and suggestions of an anonymous referee, helping them significantly improve their article. The views are the authors’ own. The usual disclaimer applies.

Declaration of Conflict of Interests

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

Notes

ORCID iDs

Bishwanath Goldar

Dibyendu Maiti

Annexure

References

Addey

K. A.

, Shaik

, & Sakouvogui

(2019). Comparative sectoral price elasticities of U.S. energy demand. Cogent Economics and Finance , 7(1), 1–17. https://doi.org/10.1080/23322039.2019.1682743

Arimura

T. H.

, Darnall

, Ganguli

, & Katayama

(2016). The effect of ISO 14001 on environmental performance: Resolving equivocal findings. Journal of Environmental Management , 166, 556–566.

Azim

, & Sawhney

(2025). Imported inputs reduce carbon intensity: Evidence from Indian manufacturing firms. Economic Modelling , 150, 107134.

Bagchi

, Sahu

S. K.

, Kumar

, & Tan

K. H.

(2022). Analysis of carbon productivity for firms in the manufacturing sector of India. Technological Forecasting and Social Change , 178, 121606.

Banerjee

S. N.

, Roy

, & Yasar

(2021). Exporting and pollution abatement expenditure: Evidence from firm-level data. Journal of Environmental Economics and Management , 105, 102403. https://doi.org/10.1016/j.jeem.2020.102403

Bansal

, & Rai

(2014). An analysis of corporate social responsibility expenditure in India. Economic & Political Weekly , 49(50).

Batrakova

, & Davies

R. B.

(2012). Is there an environmental benefit to being an exporter? Evidence from firm-level data. Review of World Economics , 148(3), 449–474.

Bjorner

T. B.

, & Togeby

(1999). Industrial companies’ demand for energy based on a micro panel database—Effects of CO₂ taxation and agreements in energy savings . In Energy Efficiency and CO₂ Reduction: The Dimensions of Social Change: 1999 European Council for an Energy-Efficient Economy Summer Study, May 31–June 4, Mandelieu, France.

Bjørner

T. B.

, Togeby

, & Jensen

H. H.

(2001). Industrial companies’ demand for electricity: Evidence from a micropanel. Energy Economics , 23(5), 595–617.

10.

Bose

R. K.

, & Shukla

(1999). Elasticities of electricity demand in India. Energy Policy , 27(3), 137–146.

11.

Central Electricity Authority (CEA). (2019). Long term electricity demand forecasting . Central Electricity Authority, Ministry of Power, Government of India.

12.

Chang

, Kim

C. S.

, Miller

J. I.

, Park

J. Y.

, & Park

(2014). Time-varying long-run income and output elasticities of electricity demand with an application to Korea. Energy Economics , 46, 334–347.

13.

Château

(2022). Energy demand drivers. In Hafner

& Luciani

(Eds), The Palgrave handbook of international energy economics . Palgrave Macmillan.

14.

Cheng

, Yu

& Zhang

(2021). Exporting and electricity consumption: new microeconomic evidence from manufacturing firms in China. Applied Economics Letters , 28(14), 1226-1233.

15.

Csereklyei

(2020). Price and income elasticities of residential and industrial electricity demand in the European Union. Energy Policy , 137, 111079.

16.

Cui

, Lapan

, & Moschini

(2016). Productivity, export, and environmental performance: Air pollutants in the United States. American Journal of Agricultural Economics , 98(2), 447–467.

17.

Dasgupta

, Satija

, & Gauba

(2017). Energy use behaviour of manufacturing industries in India: An analysis using KLEMS framework. Journal of Industrial Statistics , 6(1), 72–88.

18.

Dussaux

(2020). The joint effects of energy prices and carbon taxes on environmental and economic performance: Evidence from the French sector (Environment Working Paper No. 154). OECD.

19.

Forslid

, Okubo

, & Ulltveit-Moe

K. H.

(2018). Why are firms that export cleaner? International trade, abatement, and environmental emissions. Journal of Environmental Economics and Management , 91, 166–183.

20.

Goldar

, & Goldar

(2023). Impact of export intensity on energy intensity in manufacturing plants: Evidence from India. Journal of International Trade and Economic Development , 32(4), 639–664.

21.

Goldar

, & Goldar

(2024). Impact of ISO 14000 on energy intensity of manufacturing plants in India. Journal of International Commerce, Economics and Policy , 15(2), 2450011.

22.

Griffin

J. M.

, & Gregory

P. R.

(1976). An intercountry translog model of energy substitution responses. American Economic Review , 66(5), 845–857.

23.

Harish

S. M.

, Morgan

G. N.

, & Subramanian

(2014). When does unreliable grid supply become unacceptable policy? Costs of power supply and outages in rural India. Energy Policy , 68, 158–169.

24.

L.-Y.

, & Huang

(2021). How can export improve firms’ energy efficiency? The role of innovation investment. Structural Change and Economic Dynamics , 59, 90–97.

25.

Holladay

J. S.

(2016). Exporters and the environment. Canadian Journal of Economics , 49(1), 147–172.

26.

Imbruno

, & Ketterer

T. D.

(2018). Energy efficiency gains from importing intermediate inputs: Firm-level evidence from Indonesia. Journal of Development Economics , 135, 117–141.

27.

Jinji

, & Sakamoto

(2015). Does exporting improve firms’ CO₂ emissions intensity and energy intensity? Evidence from Japanese manufacturing (RIETI Discussion Paper Series, No. 15-E-130). Research Institute of Economy, Trade and Industry.

28.

Kant

, Kumar

, Khan

, & Sharma

(2022). Carbon dioxide emissions from India’s manufacturing sector: A decomposition analysis (Reserve Bank of India Occasional Papers, 43, No. 1).

29.

Kim

H.-J.

, Kim

G.-S.

, & Yoo

S.-H.

(2019). Demand function for industrial electricity: Evidence from South Korean manufacturing sector. Sustainability , 11(18), 5112.

30.

Kumar

, Mittal

, & Pradhan

A. K.

(2022). Magnitude and determinants of energy intensity: Evidence from Indian firms. Environmental Science and Pollution Research , 30, 3270–3281. https://doi.org/10.1007/s11356-022-22346-9

31.

Lahouel

B. B.

, Taleb

, Zaied

Y. B.

, & Managi

(2021). Does ICT change the relationship between total factor productivity and CO₂ emissions? Evidence based on a nonlinear model. Energy Economics , 101, 105406. https://doi.org/10.1016/j.eneco.2021.105406

32.

Liddle

, & Hasanov

(2024). Are the income and price elasticities of economy‑wide electricity demand in middle‑income countries time‑varying? Evidence from panels and individual countries. Environmental and Ecological Statistics , 30, 827–849.

33.

Martínez-Zarzoso

, Roy-Mukherjee

, Semrau

F.-O.

, & Voicu

A. M.

(2020). Pollution reduction by rationalization in Indian firms (Working Paper No. 2020.01). International Network for Economic Research. https://infer-research.eu/

34.

Najjar

, & Cherniwchan

(2021). Environmental regulations and the clean-up of manufacturing: Plant-level evidence. Review of Economics and Statistics , 103(3), 476–491.

35.

Nguyen

N. P.

, & Adomako

(2022). International orientation and environmental performance in Vietnamese exporting small- and medium-sized enterprises. Business Strategy and the Environment , 31(5), 2424–2436. https://doi.org/10.1002/bse.3035

36.

Oak

, & Bansal

(2022). Enhancing energy efficiency of Indian industries: Effectiveness of PAT scheme. Energy Economics , 113, 106220.

37.

Pal

, Mukhopadhyay

J. P.

, & Bhagawan

(2024). Does cap-and-trade scheme impact energy efficiency and firm value? Empirical evidence from India. Energy Economics , 134, 107581.

38.

Phoumin

, & Kimura

(2014). Analysis on price elasticity of energy demand in East Asia: Empirical evidence and policy implications for ASEAN and East Asia (ERIA Discussion Paper No. ERIA-DP-2014-05). Economic Research Institute for ASEAN and East Asia (ERIA).

39.

Pindyck

R. S.

(1979). Interfuel substitution and the industrial demand for energy: An international comparison. Review of Economics and Statistics , 61(2), 169–179.

40.

Roy

, & Yasar

(2015). Energy efficiency and exporting: Evidence from firm-level data. Energy Economics , 52, 127–135.

41.

Sahu

S. K.

, Bagchi

, Kumar

, & Tan

K. H.

(2021). Technology, price instruments and energy intensity: A study of firms in the manufacturing sector of the Indian economy. Annals of Operations Research , 313(1), 319–339. https://doi.org/10.1007/s10479-021-04295-7

42.

Sen

, & Sahoo

(2024). Carbon pricing for sustainable transition in India. World Development Perspectives , 34, 100586.

43.

Sharma

, & Padhi

(2024). Analysis of the carbon emissions trend in the Indian manufacturing sector: A decomposition and decoupling approach. Environmental Science and Pollution Research , 31(35), 1–17.

44.

Siedschlag

, & Yan

(2020). Green investments and firm performance (Working Paper No. 672). Economic and Social Research Institute.

45.

Soni

, Mittal

, & Kapshe

(2017). Energy intensity analysis of Indian manufacturing industries. Resource-Efficient Technologies , 3(3), 353–357.

46.

Tiwari

A. K.

, & Menegaki

A. N.

(2019). A time varying approach on the price elasticity of electricity in India during 1975–2013. Energy , 183, 385–397.

47.

Tran

T. M.

(2022). Environmental benefit gain from exporting: Evidence from Vietnam. World Economy , 45(4), 1081–1111. https://doi.org/10.1111/twec.13177

48.

Tran

N. D.

, Sahu

N. C.

, & Kumar

(2023). Estimation of income and price elasticities of Indian electricity demand. Electricity Journal , 36(5), 107285.