Renewable and Sustainable Energy Reviews 144 (2021) 111015

Contents lists available at ScienceDirect

Renewable and Sustainable Energy Reviews

journal homepage: http://www.elsevier.com/locate/rser

Reduced renewable energy stability in India following COVID-19: Insights

and key policy recommendations
Jai Shekhar a, 1, *, Dhruv Suri b, 1, Priyanshi Somani c, Stephen J. Lee d, Mahika Arora e
a
Department of Chemical Engineering, Manipal Institute of Technology, India
b
Candela Energy, India
c
Department of Computer Science and Engineering, Manipal Institute of Technology, India
d
MIT Energy Initiative, Massachusetts Institute of Technology, USA
e
Symbiosis School for Liberal Arts, India

A R T I C L E I N F O A B S T R A C T

Keywords: The COVID-19 pandemic has dramatically altered global energy consumption, particularly affecting investment
COVID-19 in renewable energy projects. In India, strict shelter-in-place orders enforced during March 2020 have since led
India energy demand to a considerable change in public and private sector investments in planned renewable energy installations.
Renewable energy
In this paper, we attempt to highlight trends in energy consumption and installed renewable energy capacity
Gaussian processes
noted in India during a period concurrent with the shelter-in-place orders. We discuss recent policy measures and
Machine learning
Electric power sector additions to installed renewable energy capacity, and propose key policy recommendations that may help the
Stressed assets sector adopt a growth trajectory similar to one noted pre-pandemic.
Renewable energy policy This paper is organized into four main parts. In the first section, we draw focus to India’s renewable energy
Climate change policies and pay special emphasis on recent interventions and campaigns targeted towards achieving high growth
rates in the sector. We briefly discuss the need for effective public-private partnerships in order to meet these
targets. In the second part, we quantitatively characterise the growth of renewables in India. We present an
overview of several mechanisms and missions the government has launched in line with their policy to mitigate
the environmental impact of India’s energy mix. In the third part, we analyse the decrease in electricity demand
in India from 24 March to 30 June 2020, a period concurrent with shelter-at-home orders issued by the Gov­
ernment. We also characterise changes in installed renewable energy capacity between March to December
2016–2020 to provide causal evidence of the effect of the pandemic on the growth of renewables. In this section,
we also compile and analyse data on state-wise stressed assets across renewable energy generators in the country.
Lastly, in the fourth and final portion of this paper, we highlight policy recommendations that may help the
sector overcome logistical and financial bottlenecks in the short-term. We do this with the hope of outlining key
measures that decision makers may employ to achieve pre-COVID sectoral growth in the long term. Our rec­
ommendations cover three different policy instruments: investment subsidies, operational subsidies, and rec­
ommendations for DISCOMs.

1. Introduction the pandemic and analyse the effect of COVID-19 induced foreclosures
on electricity demand and installed capacity. We conclude with key
The COVID-19 pandemic has had a severe impact on multiple in­ policy recommendations that may help re-align the sector in order to
dustries across the world. More recently, several studies have docu­ meet short-term pledges and commitments set by individual states and
mented changes in industry-specific drivers such as growth rate, the central government.
employment, and supply chains [1–3]. In this article, we discuss the India’s renewable energy sector encapsulates the world’s fourth
impact of the pandemic on the Indian renewable energy sector. We largest wind energy market and the sixth largest solar-PV market in
highlight trends in renewable energy installed capacity noted prior to terms of installed capacity [4]. More recently, the renewable energy

* Corresponding author.
E-mail address: jai.shekhar@learner.manipal.edu (J. Shekhar).
1
These authors contributed equally to this work.

https://doi.org/10.1016/j.rser.2021.111015
Received 12 August 2020; Received in revised form 12 January 2021; Accepted 21 March 2021
Available online 12 April 2021
1364-0321/© 2021 Elsevier Ltd. All rights reserved.
J. Shekhar et al. Renewable and Sustainable Energy Reviews 144 (2021) 111015

market in India has become a focal point for investors owing to re-duced provides the first step towards understanding the deficit in funding for
tariffs, low market volatility and an increased impetus towards sus­ new capacity installations.
tainable, low-carbon sources for energy generation. The following three sections are organized as follows. Section 2
Prior to the outbreak of the pandemic, India was well in line to characterises the growth of renewables in India, while providing an
achieve its short-term renewable energy targets set by the government in overview of mechanisms and missions that the government has
2015 (175 GW of elec-tricity generation by 2022, which comprises of launched in line with its policy to mitigate the environmental impact of
100 GW of solar-PV generation, 60 GW of wind energy, 10 GW of small India’s energy mix. Section 3 outlines the effect of COVID-19 on elec­
hydro power and 5 GW of bio-powered energy. For more details, refer to tricity demand, in particular, renewable energy capacity installations.
Appendix 1) [5]. Between 2014 and 2019, installed renewable capacity Lastly, Section 4, comprises of policy recommendations that may help
in India increased five-fold to 86 GW [6]. the sector overcome logistical and financial bottlenecks in the short-
Due to a high reliance of the renewable energy sector on private term. We do this in hope of outlining key measures that decision
investment and stimulus, the onset of the COVID-19 pandemic has given makers may employ to achieve pre-COVID sectoral growth in the long
rise to several financial obstacles for both the government and private term.
energy service companies and investors. Shelter-at-home orders pre­
scribed by the government have led to reduced workforce on-site and 2. India’s renewable energy sector - installed Capacity and
slowed down project development activities across the country, thereby government subsidies
devaluing the status of ongoing projects as well as those in the design
pipeline. In addition, electricity demand across the country reduced 2.1. Installed capacity
substantially, thereby leading to a decrease in revenue and lower mar­
ginal costs. In the last decade, the average share of renewable energy generation
In many cases, manufacturing, transportation and distribution of in India rose to 16% of total energy generated per year. Prior to the
materials for projects has been delayed due to shelter-at-home orders - COVID-19 outbreak, with the previously projected growth rate, elec­
either because manufacturing itself has been impacted, or a lack of ac­ tricity generated from renewables would go on to account for a third of
cess to transportation facilities was experienced. Small and medium the total electricity generated in 2030 [16]. In January 2020,
enterprises that previously had an increasing involvement with the grid-connected electricity output from renewables reached 84 GW, with
country’s green supply chain management system have also been 32 GW of electricity generation from solar-PV, 37 GW from onshore
adversely affected, damaging logistical support for the industry which wind and the remainder from small hydroelectric power [16]. The large
has resulted in widespread unemployment within the energy sector [7]. share of wind and solar-PV in the generation mix stems from recent
In April 2020, India’s proposed ‘Green Window’ campaign [8] - technological and business model innovation. Wind energy installed
which was based on a globally successful bank model - lost its key in­ capacity in India has grown 14% per year from 2007 to 2016 [4]. In
vestors as most financial blocks began to divert cash flow towards sta­ tandem, the biggest leaps in growth have been made by solar-PV in-­
bilizing losses created due to a reduction in electricity demand [9–11]. stallations (64% per year between 2013 and 2017) [4].
The Green Window campaign’s value proposition relied heavily on Since the roll-out of the aggressive 2022 target of 175 GW of
three key factors. The first was expanding low carbon energy in­ renewable energy installed capacity, the Indian renewable energy sector
vestments for the renewable energy sector. This involved raising funds was set to receive USD 5 billion from the public sector to establish
from the private sector which is commensurate with funding from state adequate infrastructure [17]. To accelerate the development of
governments. Following the outbreak of the pandemic, ongoing invest­ non-performing renewable energy assets, the government introduced
ment is now insufficient to transition to a business-as-usual scenario [12, the Renewable Purchase Obligation (RPO) scheme. RPOs made it
13]. Second, the model hoped to provide logistical and financial support compulsory for distribution companies (DISCOMs)2 in India to buy 22%
to distributed energy resource segments including rooftop solar in­ of consumed electricity from renewable sources. This would comprise of
stallations and off-grid energy service providers. The onset of 10.5% from solar-PV and 10.5% from non-solar sources.
shelter-at-home orders has delayed this process and put certain energy Decrease in solar and wind energy tariffs have made it easier for the
service companies in substantial debt-traps due to under-utilization of government to expand renewable energy installations in the country
capacity in most areas of the country [14]. Third, the model relied [18]. The Government of India is simultaneously attempting to increase
heavily on existing non-banking finance companies (NBFCs) that have solar-PV output, which has led them to commission 42 solar parks that
been significant lenders to clean energy projects in the country. In recent hold 25.5 GW of Indian solar capacity [19]. Furthermore, the National
years, NBFCs in India have seen a increase in non-performing assets from Solar Mission that was launched in 2009 aims to install 100 MW of
5.8% to 6.6% [15]. This has led to investors being susceptible to poor solar-PV capacity by 2022 [20]. The funds required to meet these targets
asset quality and an asset-liability mismatch, thereby proving to have a primarily stem from various bidding rounds under the National Solar
negative impact on the sector. NBFCs have further curtailed in-vestment Mission. These auctions were conducted in 2015 and involved two
in unreliable renewable energy infrastructure since the onset of the phases of bidding that took place within a time span of 6 months be­
pandemic. tween February and July [18]. Since January 2015, the installed wind
In this paper, we trace trends in electricity demand and new generation capacity has increased by 16471.3 MW as compared to the
renewable energy capacity installations in India during a period con­ 31403.73 MW in-crease in solar generation capacity.
current with the COVID-19 pandemic. We argue that although the na­
ture of COVID-19 is unexpected and punctuated, the central government 2.2. Government subsidies
is responsible for re-aligning the renewable energy sector in order for
India to continue to meet its installed capacity and generation targets in Most of the growth in renewables across the country has been
2022. We show that since the onset of shelter-in-place orders in March spurred by missions established by the central government, supported by
2020, the growth of renewable energy capacity in India has been sty­ an array of subsidies and funding mechanisms enabling an ecosystem
mied due to logistical and financial bottlenecks. The sector may not be
able to overcome these hurdles if not given proper direction and stim­
ulus by the government. We intend for this paper to provide a high-level 2
DISCOMs are state-led organizations responsible for the distribution of
overview of the effect of the COVID-19 pandemic on the renewable electricity to an array of customers within the state. These organizations do not
energy sector in India. Although our discussion on subsidies and finan­ generate electricity but purchase it from electricity generators and sell it to
cial stimulus is not as granular as the situation warrants, we feel that it customers within the state, hence acting as an electricity distributing entity.

2
J. Shekhar et al. Renewable and Sustainable Energy Reviews 144 (2021) 111015

conducive to low risk and high growth. In a recent study, Garg et al., energy from low-carbon sources. From January through February 2020,
documented subsidies across fossil fuels, renewables and electric vehi­ India’s electricity demand increased by 8.7% as compared to January
cles in India [21]. Their analysis classified subsidies into four categories: and February 2019, and new renewable energy capacity installations
1) direct and indirect transfer of funds and liabilities, 2) government revenue declined by 55%. During the enforcement of shelter-at-home orders
foregone, 3) provision of government goods or services below market value, from March through July 2020, electricity demand declined by 284 GW
and 4) income or price support [21]. Capital subsidies, below-market in­ (15.9% decline) relative to 2019, while installed renewable energy ca­
terest on loans and government insurance are encapsulated in direct and pacity also declined by 1.7 GW (9.21% decline). Lastly, from July
indirect transfer of funds and liabilities, while tax exemptions and mar­ through December 2020, electricity demand increased by 713 GW (11%
ginal tax rates are covered under government revenue foregone. Fig. 1 increase) relative to 2019, while installed renewable energy capacity
shows the correlation between new installed renewable energy capacity declined by 0.3 GW (5.8% decline). The onset of the pandemic slowed
and the cumulative subsidy for that year. The data for cumulative sub­ down growth across demand and new capacity installations during the
sidies was obtained from Garg et al. [21] and the installed capacity data shelter-at-home period. That said, following the gradual re-opening of
was obtained from CEA [22]. the Indian economy post July 2020, electricity demand recovered (which
Clearly, new installed renewable energy capacity is commensurate is evidenced by observed historical monthly values lying within the
with the total subsidy disbursed by the central government across confidence interval bounds in Fig. 4). New renewable energy capacity
different technologies. The government lowered the subsidy for both installations are yet to achieve pre-pandemic growth rates.
wind and solar in 2019, hoping it would act as incentive for private In Section 4, we highlight measures that may accelerate the recovery
players and encourage the sector to be independent of government of the Indian renewable energy sector. Ensuring new renewable energy
stimulus. However, the impact was adverse, leading to lower capacity installed capacity increments commensurate with increase in electricity
installations during 2019. The outbreak of the pandemic has further demand is crucial in order to meet the central government’s 2022 and
undermined growth trends, as can be seen by state of new capacity in­ 2030 targets.
stallations in 2020.

3. Impact of COVID-19 on India’s electric power sector 3.1. Is there a decrease in electricity demand from March to November
2020, and is it statistically significant?
In this section, we analyse changes in India’s electricity demand from
March to December 2020, a period concurrent with shelter-at-home Several studies report a causal link between the onset of COVID-19
orders issues by the government followed by a gradual reopening of induced foreclosures and decrease in electricity demand [11,13,
the Indian economy. We employ a counterfactual framework wherein 23–26]. In this study, we employ a simple and interpretable probabilistic
electricity demand from March 2013 through February 2020 is used to model to evaluate the statistical significance of electricity demand be­
train a Gaussian process regression model that predicts India’s elec­ tween March and December 2020. The framework employed was
tricity demand from March to December 2020. This is a counterfactual adopted from that used by Luke et al. [27]. We elaborate on the data and
scenario that models electricity demand in the absence of COVID-19. We methods employed in the modeling exercise, and discuss the implica­
then compare observed historical demand from March to December tions of the results.
2020 to the counterfactual projections to demonstrate the impact of
COVID-19 on the electricity sector. 3.1.1. Data
Looking at India’s long-term commitments to installed renewable Power System Operation Corporation (POSOCO) Limited, a Gov­
energy capacity in accordance to the 2015 Paris Agreement, we note ernment of India enterprise, compiles electricity demand and source-
that the increment in electricity demand in India should be commensu- wise electricity generation data, daily [28]. This data is available at
rate with new renewable energy capacity installations. This does not three levels of geographical resolution: state, region, and country-level.
imply that central and state governments should resist investing in We aggregate the total electricity generation data at the country-level
conventional energy generators, however, it helps us establish a metric for India in monthly intervals and utilize this for our analysis. POS­
wherein India fulfils its commitment to generate 40% of the country’s OCO also lists monthly renewable energy capacity installation by source
(solar- PV, bio-energy, wind, geothermal, nuclear and hydro) using

Fig. 1. New installed renewable energy capacity in India from 2015 through 2020. The solid black line indicates cumulative subsidies for renewables in million USD.
The graph has been compiled by the authors using data for cumulative subsidies obtained from Garg et al. [21] and installed capacity data obtained from CEA [22].

3
J. Shekhar et al. Renewable and Sustainable Energy Reviews 144 (2021) 111015

which we compute the percentage change in capacity installations be­ was minimal or contained. We hypothesize that these relaxations, with
tween March and May as well as March and November, for 2016–2020. the subsequent announcement of Unlock 1.03 may be responsible for the
gradual increase in electricity demand in the months of May and June
3.1.2. Methods 2020. That said, the observed electricity demand in these two months is
We use Gaussian process (GP) regression, a probabilistic modeling still 17.65% and 11.32% below the mean predicted by the GP regression
technique used in machine learning, to compute the mean trend in model in Fig. 2 in a counterfactual scenario for May and June 2020,
electricity demand from monthly POSOCO data. GP regression is a non- respectively.
parametric, Bayesian approach to regression that helps characterise Table 1 summarizes the statistical significance of the actual elec­
uncertainty. The GP regression model computes empirical confidence tricity demand from March through December, 2020. We measure the
intervals that allow us to analyse the statistical significance of projected deviation from the mean projected by the GP regression model.
versus observed historical data [27]. Historical demand from March
2013 to February 2020 is used to train the GP model, which is then 3.2. Delay in the installation of new capacity
projected into Q2 and Q3, 2020, to forecast electricity demand in a
counterfactual scenario in the absence of COVID-19. In order to better The renewable energy sector in India relies heavily on imports from
characterise the input data, the GP regression model uses a series of other Asian countries such as China and Taiwan [31]. The solar power
kernels to help fit the input dataset and compute projected data. In this industry, with 88% of modules imported from China, is expected to
setup, we use three kernels that fit different properties of the input data: undergo minimal increments in installed capacity in the coming months
the long-term smooth rising trend is implemented using the RBF kernel, [32]. In a study conducted by Wood Mackenzie in April 2020, the au­
with the length scale and amplitude as hyperparameters. The seasonal thors forecasted a delay in commissioning of 400 MW of renewable
trend in the data is implemented with the ExpSineSquared kernel with a capacity installation in 2020, an 11% reduction compared to renewable
frequency of 1 year to represent electricity demand annually. Short-term energy capacity installations in 2019 [33]. The study also analysed that
and medium-term irregularities of the order of a few months are rep­ Q1 2020 will see a 60% downgrade in installed capacity - only 1.2 GW of
resented using the RationalQuadratic kernel. The model takes into ac­ new installed capacity from January to April 2020 as compared to
count seasonal patterns in electricity demand (such as the effect of January to April 2019 [33]. Further, our analysis of installed capacity
temperature and rainfall) via the ExpSineSquared kernel. The entirety of data from POSOCO’s monthly reports showed that there was an incre­
the GP regression model and its associated framework is designed using ment of 0.75% in installed capacity in India between February through
the GPy library in Python [29]. May 2020, as compared to 4.62% in 2019, and 10.38% in 2018. This
year-over-year comparison is shown in Table 2.
3.1.3. Results and discussion Similarly, a large drop was noted in increased installed capacity
We show in Fig. 2, deviations from the GP regression model’s pre­ between February through November 2020, with an increment of just
dicted electricity consumption. The solid blue line indicates the mean 9.21% as compared to 17.28% in 2019 and 21.33% in 2018 as shown in
computed by the GP, while the shaded blue region represents the 95% Table 3.
confidence interval. The counterfactual data is used to project monthly In a recent announcement by the Department for Promotion of In­
electricity demand in a business-as-usual scenario without COVID-19. dustry and Internal Trade (DPIIT) in March 2020, the organization
The red points represent actual monthly electricity demand from stated that its proposed launch of 50 solar parks with a cumulative
March to December 2020, a period concurrent with shelter-at- home funding of USD 1.3 billion has been pushed back to an undetermined
orders in India. date due to logistical concerns in light of the pandemic [34].
82 out of 83 historical monthly electricity demand points shown in Across the renewable energy sector in India, the outbreak of COVID-
green fall within the 95% confidence region (or 95% credible interval) 19 has considerably slowed progress, created logistical bottlenecks,
represented by the shaded blue region. We observe that none of the delayed auctions and stymied construction of new projects. In Section 4,
points representing electricity demand from March through June 2020 we present recommendations that may fast-track the stabilisation of the
lie in the 95% confidence interval. The average deviation from the mean industry and help establish a robust framework resilient to COVID-19
for the months during which shelter-at-home orders were enforced is induced disruptions.
17.5%.
Although this statistical evidence may not be sufficient for us to draw 3.3. COVID-19 induced foreclosures and loss of employment
a causal relationship between historical monthly electricity demand and
decreases in demand during the shelter-at-home period, it does help us 3.3.1. Data
draw a preliminary conclusion about the correlation between the two. DPIIT publishes state-wise stressed assets in India across 26 sectors
Our claim of a causal relationship between electricity demand and the [35]. We filtered this dataset to include only assets of the Electricity
onset of COVID-19 shelter-at-home orders may be further validated by generation (renewable) category across all states. For the purpose of this
analysing year-over-year monthly electricity demand from January to study, only asset listings where the entry was made on or after March 22,
June from 2016 to 2020, shown in Fig. 3. 2020 have been included. This is concurrent with the announcement of
Looking at Fig. 3, we see that the monthly electricity demand for shelter-in-place orders imposed in India.
January and February 2020 is higher than the historical electricity de­
mand for these two months between 2016 and 2019. Following the 3.3.2. Stressed assets
enforcement of strict lockdown measures in India around mid-March DPIIT has pooled 55 stressed renewable energy assets since the onset
2020, the electricity demand drops by 7% in March 2020 and 23% in of the pandemic. These assets include 21 solar energy assets, 28 trans­
April 2020 compared to total electricity demand in February 2020. On mission and distribution assets, 4 bio-power energy assets, 1 nuclear
April 14, 2020, the Government extended the nationwide lock-down energy asset and 2 hydro power energy assets. The state-wise distribu­
until May 03, 2020, with conditional relaxations following April 20, tion of these assets are shown in Fig. 5.
2020, in regions where the outbreak of cases associated with COVID-19 Financial setbacks caused due to incomplete projects and stressed

3
The Government of India carried out a structured re-opening of the econ­
omy in one-month phases. As of November 20, 2020, the country is in phase six
of its Unlock campaign.

4
J. Shekhar et al. Renewable and Sustainable Energy Reviews 144 (2021) 111015

Fig. 2. Gaussian process (GP) regression

applied on historic monthly energy demand
of India. The green points indicate actual
historical values from April 2013 to
February 2020 used to fit the GP. The solid
blue line indicates the mean computed by
the GP, while the shaded blue region repre­
sents the 95% confidence interval. The
counterfactual data is used to project
monthly electricity demand in a business-as-
usual scenario without COVID-19. The red
points represent actual monthly electricity
demand from March to December 2020, a
period concurrent with shelter-at-home or­
ders in India.

Fig. 3. India, year-over-year monthly electricity demand trends for India for January, February, March, April, May and June 2016–2020.

assets have affected 2253 contractors and 248 project developers 3.3.4. Effect on DISCOMs
working in the energy sector [34]. Reduced revenue for firms due to DISCOMs and electric utilities act as branches of the electric power
lower investment trends in addition to delayed auctions in the first half network in India. Historically, utilities have been a source of major debt
of 2020 have stymied growth and threatened the stability of renewable and slow market growth due to lack of oversight and sub-par imple­
energy contractors. mentation of billing and collection. In addition to this heavy debt,
DISCOMs in India encounter nonviable aggregate technical and com­
3.3.3. Loss of employment and pulling of funding mercial (ATC) losses, which usually occur due to poor transmission
Shelter-at-home orders have put 300,000 personnel in the renewable infrastructure, unregulated maintenance, and theft of electricity in rural
energy sector at the risk of losing their jobs [36]. This situation arises and peri-urban areas [40]. Since the onset of the pandemic, DISCOMs
due to a forced trend of layoffs by organizations running in loss [37]. face many additional challenges [41].
Heavy financial diminution has led current private players and organi­ Since March 2020, DISCOMs have suffered major financial draw­
zations who were already investing in the sector, or those who had backs, which have now lead to a cumulative loss of USD 3.98 billion
shown serious interests in investing, to pull their funding. [42]. This figure does not include their pending dues of USD 12.21
Financial setbacks such as this have resulted in the auction of non billion prior to February 2020. The Ujwal DISCOM Assurance Yojana
performing assets (NPA), including coal plants and coal mines, to be (UDAY) 2.0 scheme (also referred to as the ‘ADITYA’ scheme) was set to
pushed back to a later date [38]. The auction was open to foreign direct include reforms for the setup of smart meters and incentives for tariff
investments (FDI), however, due to major losses suffered by private rationalization to decrease financial burdens on DISCOMs. This has since
organizations both in India and around the world, the bidding was been delayed in its roll out, initially planned for 2020 [43]. Further­
deemed to be non-conducive. These assets include 3.1 GW of thermal more, due to depleting customer electricity demand, these DISCOMs
capacity that were initially allocated under major Indian public sector continue to demonstrate a decreasing performance when measured in
units such as the National Thermal Power Cooperation (NTPC) and Coal terms of units of electricity sold compared with their sales performance
India Limited (CIL). From a financial standpoint, these assets would in previous years.
have attracted USD 330 billion in investments from global mining firms
set to bid for them during the highly anticipated coal auctions towards
the end of 2020 [39].

5
J. Shekhar et al. Renewable and Sustainable Energy Reviews 144 (2021) 111015

Table 3
Percentage increase in installed renewable en-ergy capacity, India, March to
November 2016–2020. Data compiled from POSOCO [30].
Year Percentage Increase Capacity Increase

2016 24.47% 9029.71 MW

2017 25.22% 12116.9 MW
2018 21.33% 12660.56 MW
2019 17.28% 12286.21 MW
2020 9.21% 7629.14 MW

4. Key policy recommendations

4.1. Investment subsidies

4.1.1. Reducing upfront project costs

Direct capital investments have been provided as a share of overall
project investment costs, or per-kWh of rated power of the installation.
The Indian Renewable Energy Agency (IREDA), India’s financial insti­
tution dedicated to clean energy, also provides low-cost financing to
renewable energy projects. In addition to this funding mechanism, the
Ministry of New and Renewable Energy recently inaugurated the Central
Public Sector Undertaking (CPSU) Scheme Phase-II for setting up 12000
MW of grid-connected PV projects with viability gap funding (VGF)
Fig. 4. State-wise percentage decrease in electric-ity demand from January to support. This funding is issued in the form of grants to projects under­
June 2020 as compared to 2019. The color-scale represents a decline in demand taken through public-private partnerships with the intent to make them
from 0% to 32% for states all over India, with the darker regions representing a commercially viable. The maximum funding cap for VGF support is fixed
higher decline in demand. Graph generated by the authors using data compiled at 20% of the total project cost. The mechanism also allows for the
from POSOCO [30]. government or public entity that owns the project to provide additional
grants out of its own budget for financing a further 20% of the total
project cost.
Table 1
This provision gives impetus for public-private partnerships to invest
Percent deviation between observed values and counterfactual estimates
and 95% credible interval bounds for monthly electricity generation in
in new renewable energy capacity installations in the country. Not only
India for March through December 2020. should VGF support be promoted for solar installations, but it should
cover all renewable energy generation technologies, including those
Month Percent Deviation
that have a high capital cost. Since the onset of the pandemic, govern­
March 2020 − 12.68% ment agencies such as the Solar Energy Corporation of India (SECI) have
(95% credible interval bounds) (+/-7.55%)
reported under-utilization of funds, and hence may allocate the un-
April 2020 − 27.09%
(95% credible interval bounds) (+/-7.36%) utilized budget towards new capacity installations in partnership with
May 2020 − 17.79% private organizations. Thapar et al. [18] found that VGF support for
(95% credible interval bounds) (+/-7.13%) solar projects reduced the upfront cost of equity investment by the
June 2020 − 11.01%
developer down to 15% (instead of 30%). Such capital subsidies applied
(95% credible interval bounds) (+/-7.29%)
July 2020 − 5.09%
to domestic solar modules and other renewable energy technologies,
(95% credible interval bounds) (+/-7.19%) such as offshore wind, may help increase the rate of new capacity in­
August 2020 − 7.25% stallations leading up to 2022.
(95% credible interval bounds) (+/-7.07%)
September 2020 − 2.95%
4.1.2. Accelerated depreciation
(95% credible interval bounds) (+/-7.16%)
October 2020 − 5.11% Through the accelerated depreciation (AD) benefit, renewable en­
(95% credible interval bounds) (+/-7.08%) ergy projects depreciate by a specified amount in the first year, allowing
November 2020 − 10.71% investors and project developers to avail tax benefits in the short-term.
(95% credible interval bounds) (+/-7.38%) The benefit was first introduced in 1994, with a depreciation rate of
December 2020 − 2.83%
(95% credible interval bounds) (+/-7.16%)
100% and subsequently lowered to 80% in 2002. In 2012, the AD benefit
Average deviation − 9.81% was withdrawn, although, it was reinstated in 2014 with the deprecia­
tion rate set at 80%. Following a review by the Income Tax Department
of India, the depreciation rate was lowered to 40% in 2018. Prior to this
Table 2
revision, the AD benefit was recognized as one of the key factors un­
Percentage increase in installed renewable energy capacity, India, March to derlying an increase in renewable energy capacity installations as it
May, 2016–2020. Data compiled from POSOCO [30]. offered investors a low-risk instrument to offset profits. Furthermore, as
the benefit did not require direct fund disbursal from the Government, it
Year Percentage Increase Capacity Increase
has been seen as an effective policy instrument to supplant capacity
2016 10.83% 3998.28 MW installations in the country.
2017 15.07% 7242.23 MW
2018 10.38% 6164.28 MW
With the sharp decrease in renewable energy capacity installations
2019 4.62% 3285.02 MW during the lockdown, there are two vital alterations that will have to be
2020 0.75% 624.83 MW made in order to attract investors and promote new projects:

6
J. Shekhar et al. Renewable and Sustainable Energy Reviews 144 (2021) 111015

Fig. 5. Distribution of stressed assets that have emerged in 2020 due to the COVID-19 pandemic, grouped by state. The distribution reflects the setback caused to the
Indian renewable energy sector in terms of projects that may encounter pandemic-induced foreclosures. This figure has been compiled by the authors using data
published by DPIIT [34].

• Revising the depreciation rate to 80%: although the decision to made thermal power cheaper and solar-PV infrastructure 6% more
revise the depreciation rate from 80% to 40% in 2018 was made to expensive. Furthermore, the Government proposed an import duty of
reduce the tax revenue foregone by the central government, a low 20% on solar cells and modules to promote local manufacturing,
depreciation rate may not be enough to promote new capacity in­ although this has been put on hold [44].
stallations in the short term. Revising the depreciation rate to 80% The high price of individual components may prevent project de­
offers financial support with a low risk to investors. Furthermore, as velopers from planning new capacity installations in the short term.
organizations are increasingly concerned about cash-flows following Regular cash flows have been disrupted and developers may still need to
the outbreak of the pandemic, a benefit that does not require any pay installments on capital expenditure loans for ongoing projects.
direct disbursal of funds from the government to the organization Keeping this in mind, import duty concessions may help achieve two
increases investor confidence. targets: 1) a lower interim import duty may help developers import
• Redirecting tax incentives for foreign investors: at present, the available components from other countries without increasing the pro­
AD benefit can only be availed by investors and entities that draw a jected capital expenditure for ongoing projects. Which then leads to 2)
profit within the Indian subcontinent. For organizations not eligible developers may redirect funds to complete existing projects halted due
to draw a tax benefit in India, an additional subsidy in the form of a to the unavailability of domestic components.
generation based incentive (GBI), over-and-above the feed-in-tariffs
provided by SERCs, can push foreign investors to invest in capacity
4.2. Operational subsidies
installations. One of the draw-backs of the AD benefit was that it
rewarded capacity installations and not performance or generation.
4.2.1. Generation based incentives
With the AD benefit redirected to a GBI in the first year (with the
Under the GBI scheme launched by MNRE in 2008, new wind power
option to carry forward to subsequent years), the central and state
and solar-PV projects with at least 5 MW of grid-connected capacity are
governments can guarantee the proliferation of functional renewable
eligible to receive INR 0.5/kWh for ten years, which is in addition to the
energy projects in the long term, and of course, promote domestic
tariff set by state regulators and utilities. The maximum benefit that can
and foreign investors to develop new renewable energy projects in
be availed by a project is capped at INR 10 million per MW. The GBI
the short term.
scheme is targeted at independent power producers (IPPs). Since the
announcement of the scheme in 2008, it is estimated that one-third of
4.1.3. Import relaxations
new wind capacity in India has been set up via IPPs.
With the pandemic imposing stiff constraints on both domestic and
Power purchase agreements (PPAs) in the short-term can include a
international supply chains, the central government may provide
direct cash incentive for every kWh of energy generated from renew­
interim import duty concessions for project developers importing com­
ables, regardless of the installed capacity. This method has been suc­
ponents for new capacity installations. These interim concessions may
cessfully employed in Germany and the U.S [45,46]. In Germany, the
be in place till domestic manufacturers can reorganize supply chains and
process involves a direct cash payment of 0.03 Euros/kWh of renewable
reinstate their workforce. The problem of taxation on sub-components is
energy generated by IPPs. Additionally, a feed-in law enforced by Ger­
further exacerbated since the roll-out of the Goods and Services (GST)
many requires distribution utilities to purchase electricity generated by
tax in 2018. Soman et al. [44] found that the newly introduced GST
renewables at a price equivalent to 90% of the final tariff paid by the

7
J. Shekhar et al. Renewable and Sustainable Energy Reviews 144 (2021) 111015

consumer. Such a mechanism ensures a constant purchase of energy purchasing electricity from the grid owing to long-term affordability.
generated from renewables even if the cost per kWh is higher than that Further, transitioning to sustainable low-carbon sources of energy can
of thermal plants. In the long run, such measures may help ensure that also aid in achieving organizational renewable targets via renewable
project developers planning new capacity installations leading up to energy certificates.
2022 are confident about their return on investment and project
financing.
4.3. Bailing out Lumbering electricity DISCOMs

4.2.2. Three-tier approach

One of the key issues within India’s renewable energy sector has
Existing renewable energy producers should also be beneficiaries of
been poor monetary fitness of DISCOMs [49]. A recent study conducted
implemented operational subsidies. To encourage individual power
by Veluchamy et al. concluded that high monetary rates could only be
producers to perform better post-pandemic, they may be divided into a
stabilized if the Government of India creates conducive tariff policies
three tier system. The system proposed here draws on insights followed by
and infrastructure requirements for existing DISCOMs [50]. A disjointed
the California Energy Commission [47].
and unstable energy ecosystem during the enforcement of
The first tier accommodates the least cost-effective utilities that have
shelter-at-home orders only compounded the problems faced by DIS­
yearly turnovers 10% greater than the total cost of producing energy for
COMs in India.
the same period of time. The second accommodates moderately cost-
Post-pandemic, various DISCOMs will resume their billing and
effective utilities that have a yearly turnover 20% greater than the
collection operations. The Government of India released a USD 11.94
total cost, and the third accommodates highly cost-effective utilities that
billion package to support DISCOMs and streamline their operations
have yearly turnovers of 30% or more than the total cost of energy
[42]. In addition to using this stimulus towards covering debt, DISCOMs
production.
should invest in strategic planning and debt-recovery mechanisms that
Individual power producers in each tier should be provided with
may help prevent a similar situation from re-occurring in the long-term.
direct cash payments per-kWh of energy produced. The direct cash
Policies such as the UDAY 2.0 aim to enforce a rigorous system of
payment to individual power producers in tier three would be higher
energy accounting. UDAY 2.0 ensures the payment of subsidies every
than those in tier one and tier two. This would create a substantial
month while mandating the use of smart-meters to ensure timely billing
incentive for individual power producers to create greater profit margins
and collection. DISCOMs need to improve their management structure,
which can only be driven by providing better quality of customer ser­
keep track of inaccurate bills, recruit more human capital, and provide
vices [48]. The system can be put into effect from the second quarter of
frontline personnel with performance-based incentives. Additionally,
2021 and could include an additional investment subsidy that in­
state regulators should account for DISCOMs to spend more on metering,
centivizes expansion or repowering of solar or wind utilities in the
billing and collection via state-sponsored infrastructure investments.
country. The fiscal package for this could be included through agreed
While this may cause further financial expenditure in the short-term, it
PPAs that would include a must-run or deemed generation clause for the
will help DISCOMs resurface post-pandemic.
expanded section of any power plant. Cumulatively, this would promote
expansion and generation via existing renewable energy generators in
5. Conclusion
the nation. Although its timely implementation must be supported by
the implementation of strict, transparent anti-curtailment and must-run
The COVID-19 pandemic has exposed the fragility of the Indian
guidelines to increase willingness to invest among national and inter­
renewable energy sector. Large-scale losses, project delays and overly-
national stakeholders.
ambitious plans now face unavoidable setbacks. A clear indication of
Since every state in India does not have a well-established network of
the impact of the pandemic is reflected in the sharp reduction in elec­
renewable energy utilities [31], a large number of utilities in a
tricity demand all over the country. The Government of India should
non-performing state might find it difficult to navigate into a tier with
work towards creating flexible polices that could aid the renewable
better cash incentives. Therefore, the criteria for a utility to be assigned
energy sector to grow with minimal hindrances post-pandemic. It is now
to a particular tier could vary from state to state. A disparate tier system
time for the country to work towards creating a stable ecosystem by
must also be created for utilities depending on their respective renew­
following well-planned short term goals that can be worked towards,
able energy source. This would prevent utilities working on relatively
using the resources in hand.
untapped sources such as wind or geothermal from not gathering sub­
The pandemic may have caused unforeseen logistical and financial
stantially greater cash incentives only because the demanded positive
complications, but enough backing by the Government of India through
turnover would be too high to achieve.
stimulus packages and policy implementations can create a renewable
energy friendly market post pandemic. Shareholders seem highly
4.2.3. Corporate power purchase agreements
interested in the Indian renewable energy market, as demonstrated by
Investor confidence in newly established renewable energy projects
patterns in renewable energy investment in India. Key policy in­
may be instilled if state governments take the lead in forming alliances
terventions will reduce general uncertainty in the renewable energy
between power producers, corporate and industrial users. A corporate
ecosystem. A policy re-framing initiative would also aid in creating a
power purchase agreement (CPPA) covers the framework for energy
sizable stand against the financial implications caused due to the
exchange between a renewable energy plant and a corporate entity.
COVID-19 pandemic.
Ireland’s Draft National Energy and Climate Plan (NCEP) 2021–2030
identifies CPPAs as a subsidy free route to the market for renewable
project developers. Declaration of competing interest
For renewable project developers, CPPAs provide a long-term in­
come stream with a quicker turnaround time for revenue generation as The authors declare that they have no known competing financial
the arrangement does not involve central or state-level government interests or personal relationships that could have appeared to influence
actors. For corporate entities, CPPAs are a good workaround to the work reported in this paper.

8
J. Shekhar et al. Renewable and Sustainable Energy Reviews 144 (2021) 111015

Nomenclature

ACS Average cost of supply

ARR Average revenue released
ATC Aggregate technical and commercial
DISCOM Distribution company
DPIIT Department for Promotion of Industry and In- ternal Trade
FDI Foreign direct investment
GP Gaussian process
GS T Goods and Service Tax
GW Giga watt
KUS UM Pradhan Mantri Kisan Urja Suraksha evam Utthan Mahabhiyan
kW Kilo watt
kWh Kilo watt hour
MW Mega watt
NBFC Non-banking finance company
NPA Non performing asset
POS OCO Power System Operation Corporation
PPA Power Purchase Agreement
RPO Renewable Purchase Obligation
UDAY Ujjwal DISCOM Assurance Yojana

Appendix 1. Source-wise Installed Capacity

Fig. 1.6. Renewable energy installed capacity in India by source. The blue bars represent historical installed capacity from 2014 through 2020 whereas the red bars
represent 2022 targets set by the Government of India. The installed capacity data was obtained from CEA [22].

References [4] NITI Aayog. IEA India 2020 - In-depth Energy Policy Review. 2020.
[5] Goel M. Solar rooftop in India: policies, challenges and out- look, vol. 1. Green
Energy & Environment; 2016. p. 129–37. 2.
[1] Gursoy D, Chi CG. Effects of covid-19 pandemic on hos- pitality industry: review of
[6] Agarwal P. Linkages between the optimum management of nat- ural resources and
the current situations and a research agenda. 2020.
sustainable development. Purakala with ISSN 0971-2143 is an UGC CARE Journal
[2] del Rio-Chanona RM, Mealy P, Pichler A, Lafond F, Farmer D. Supply and demand
2020;31(50):82–91.
shocks in the covid-19 pan- demic: An industry and occupation perspective. 2020.
[7] Rao PH. Green supply chain management: a study based on smes in India. Journal
arXiv preprint arXiv:2004.06759.
of Supply Chain Management Systems 2019;8(1):15.
[3] Kumar S, Maheshwari V, Prabhu J, Prasanna M, Jayalak- shmi P, Suganya P,
[8] Gupta A, Bonds G. Financing india’s renewable energy vi- sion. 2020.
Malar BA, Jothikumar R. Social eco- nomic impact of covid-19 outbreak in India.
[9] Das K. Impact of covid-19 pandemic into solar energy genera- tion sector. 2020.
International Jour- nal of Pervasive Computing and Communications; 2020.
Available at SSRN 3580341.

9
J. Shekhar et al. Renewable and Sustainable Energy Reviews 144 (2021) 111015

[10] Hashemi SSG, Mahmud HB, Ashraf MA. Perfor- mance of green roofs with respect [28] Power System Operation Corporation Limited. Daily Reports. 2020. 8th August
to water quality and reduction of energy consumption in tropics: a review. Renew 2020.
Sustain Energy Rev 2015;52:669–79. [29] GPy, “GPy: a Gaussian process framework in python.” http://github.com/
[11] Le Quéré C, Jackson RB, Jones MW, Smith AJ, Aber- nethy S, Andrew RM, De- SheffieldML/GPy, since 2012.
Gol AJ, Willis DR, Shan Y, Canadell JG, et al. Temporary reduction in daily global [30] Power System Operation Corporation Limited. Monthly Re- ports; 2020. 8th August
co 2 emis- sions during the covid-19 forced confinement. Nat Clim Change 2020: 2020.
1–7. [31] Khare V, Nema S, Baredar P. Status of solar wind re- newable energy in India.
[12] Eroğlu H. Effects of covid-19 outbreak on environment and renewable energy Renew Sustain Energy Rev 2013;27:1–10.
sector. Environment, Development and Sus- tainability; 2020. p. 1–9. [32] Singh AK, Idrisi AH. Evolution of renewable energy in India: wind and solar. J Inst
[13] Gillingham KT, Knittel CR, Li J, Ovaere M, Reguant M. The short-run and long-run Eng: Series C 2019:1–13.
effects of covid-19 on energy and the environment. Joule; 2020. [33] Liew R. Wood Mackenzie Report. 2020. June 26, 2020.
[14] Mehra P. Covid corrections: How the pandemic reveals the failures of india’s [34] Government of India. Department For Promotion Of Industry and Internal Trade
growth model. 2020. (DPIIT). 2020. June 26, 2020.
[15] Shaikh N, Shah N. Role of government and rbi in financial inclusion-an ecosystem [35] Government of India. Department for promotion of industry and internal trade
of development. Studies in Indian Place Names 2020;40(8):261–71. Stressed Assets in India. DPIIT 2020. 26 June 2020.
[16] Majid M, et al. Renewable energy for sustainable develop- ment in India: current [36] I. L. O. (ILO). World employment and social outlook 2018: Greening with jobs.
status, future prospects, challenges, em- ployment, and investment opportunities. 2018.
Energy, Sustainability and Society 2020;10(1):2. [37] Mustakim M, Syafrida S. Covid-19 pandemic as a force majeure for massive layoffs.
[17] Reuters. India seeks bids for USD 5 billion in transmission lines to fuel renewables SALAM: Jurnal Sosial dan Budaya Syar-i 2020;7(8).
growth. 2020. July 04, 2020. [38] IANS. Lockdown delays commercial coal mining auction to July, govt finalises bid
[18] Thapar S, Sharma S, Verma A. Economic and envi- ronmental effectiveness of document. 2020. November 19, 2020.
renewable energy policy instruments: best practices from India. Renew Sustain [39] Government of India. Coal India Limited. 2020. June 26, 2020.
Energy Rev 2016;66:487–98. [40] Das SD, Srikanth R. “Viability of power distribution in India–challenges and way
[19] Fairley P. India grapples with vast solar park-[news]. IEEE Spectrum 2020;57(2): forward. Energy Pol 2020;147:111882.
8–9. [41] K. Balani, B. Sharma, and S. Agrawal, “Improving discoms’ revenue recovery
[20] Behuria P. The politics of late late development in renewable energy sectors: during and after the pandemic,”.
dependency and contradictory tensions in India’s national solar mission. World [42] Ghosh A, Shuva R. Jobs, growth and sustainability. 2020.
Dev 2020;126:104726. [43] Singh K. Credit ratings of power distribution utilities in north- ern region of India.
[21] Garg V, Viswanathan B, Narayanaswamy D, Beaton C, Ganesan K, Sharma S, Asian Journal of Economics, Finance and Management 2020:11–9.
Bridle R. Mapping india’s energy subsidies 2020: Fossil fuels, renewables, and [44] Soman A, Gerasimchuk I, Beaton C, Kaur H, Garg V, Ganesan K. “India’s energy
electric vehicles. 2020. transition. 2018.
[22] Government of India. Central Electricity Authority. 2020. ac- cessed July 04, 2020. [45] Davidson C, Steinberg D, Margolis R. Exploring the market for third-party-owned
[23] Abu-Rayash A, Dincer I. Analysis of the electricity de- mand trends amidst the residential photovoltaic systems: insights from lease and power-purchase
covid-19 coronavirus pandemic. Energy Research & Social Science 2020;68: agreement contract struc- tures and costs in California. Environ Res Lett 2015;10
101682. (2):024006.
[24] Bahmanyar A, Estebsari A, Ernst D. The impact of dif- ferent covid-19 containment [46] Neuhoff K, Amecke H, Stelmakh K, Rosenberg A, Novikova A. Meeting energy
measures on electricity consumption in europe. Energy Research & Social Science concept targets for residential retrofits in Germany: economic viability, financial
2020;68:101683. support, and energy savings. CPI Brief, Climate Policy Initiative 2011.
[25] Kanda W, Kivimaa P. What opportunities could the covid- 19 outbreak offer for [47] C. E. C. I. E. P. R. Committee, Integrated energy policy report. California energy
sustainability transitions research on elec- tricity and mobility? Energy Research & Commission; 2007.
Social Science 2020;68:101666. [48] Goodman J. Strategic customer service: Managing the cus- tomer experience to
[26] Santiago I, Moreno-Munoz A, Quintero-Jiménez P, Garcia- Torres F, Gonzalez- increase positive word of mouth, build loy- alty, and maximize profits. Amacom;
Redondo M. Electricity demand during pandemic times: The case of the covid-19 in 2019.
spain. Energy pol- icy; 2020. p. 111964. [49] Jhunjhunwala A, Lolla A, Kaur P. Solar-dc microgrid for indian homes: a
[27] Luke M, Somani P, Cotterman T, Suri D, Lee SJ. No covid-19 climate” silver lining” transforming power scenario. IEEE Electrifica- tion Magazine 2016;4(2):10–9.
in the us power sector: Co 2 emissions reductions not statistically significant, [50] Veluchamy A, Sunder RG, Tripathi R, Nafi RM. Pow- ering India: the next-
additional risk to coal generators is minimal. 2020. arXiv preprint arXiv: generation reforms in power distribution sector. Int J: lied Engineering Research
2008.06660. 2018;13(21):15016–34.

10