Caught in the power maze: India’s discoms battle solar’s growing costs

Part 3 of CarbonCopy’s three-part series on India’s solar sector finds that discoms are struggling to manage the rising share of renewable energy. Between financial strain, outdated incentives and the challenges of balancing power intermittency, they need a more equitable distribution of renewable power’s risks

Read Part 1, Part 2

Do you remember the day the lights went out?

On 5 April, 2020, Prime Minister Narendra Modi told Indians to switch off lights for 9 minutes at 9 pm as a symbolic act against “the darkness spread by the corona crisis”. 

During those minutes, India’s grid saw power demand plummet steeply — by 32 GW to 117 GW — before climbing, just as steeply, back up again.

It was an instructive moment. Grids have an optimum frequency. They cannot store power. Ergo, power supply has to be calibrated to power demand. 

On the night of April 5, India’s grid operators responded by replacing thermal power generation with hydel power, which can be dialled down or up rapidly. As people started switching their lights off, power supply from dams was reduced, and then, at the end of the nine minutes, it was ramped up again, keeping the grid stable till thermal generation rose again.

What India saw on that night was an isolated instance, handled seamlessly thanks to extensive preparation and coordination. 

Now imagine similar mismatches between demand and supply every single day.

Is India ready for clean power?

As the first part of this series said, India’s solar sector is settling into ebullience. Recent tenders for round-the-clock (RTC) renewable power have netted prices comparable to that of coal, suggesting that the sector has bested its old bugbear of intermittency. 

The country is also seeing a massive expansion in solar manufacturing — from 10 GW in 2021 to 60 GW now. With more capacity addition in the works, the country’s solar manufacturing capacity is expected to touch 110 GW by 2025-26.

While a part of this production might be shipped to countries like the USA, much will be deployed within India. As mentioned in the first story in this series, a senior advisor to the Indian government told CarbonCopy that if India improves its processes for land acquisition and transmission, the country could more than double its current installed solar power capacity from the current 81 GW by 2030.

The question that needs to be asked here is—Is India ready for 160 GW of renewable power?
The answer to that question extends beyond land acquisition and expansion of transmission lines. Land can be acquired, as projects like Karnataka’s Pavagada Solar Park have shown, through leasing. Additional transmission lines can be laid as well. 

Instead, the primary challenges to India’s solar expansion come from two other fronts. One, already discussed in our report yesterday—the high cost of Indian modules. Two, the unequal distribution of risks between developers and discoms. 

Navigating a maze

Broad numbers are enough to grasp the central issue here.

India’s power demand ranges between 170 GW to 240 GW, even 250GW, in a day. 

The country’s base demand stays around 170 GW throughout the day with one supply spike in the day and one demand spike at night. The supply spike comes from solar, which accounts for 84GW of India’s total installed generation capacity of 444GW. The demand spike comes at night between 7-11PM as households switch their lights on. Hardwired into this pattern are two problems. 

First, India has to find takers every day for the 84 GW of solar power pumped by developers into the grid. Then, as solar generation wanes with the sun, the country has to replace that 84 GW with something else — be it thermal, hydel or something else.

About two months ago, trying to understand how India’s grid balances these mismatches between demand and supply, CarbonCopy met a former executive director at India’s central grid controller, Power System Operation Corporation, or POSOCO.

Till now, discoms have managed intermittency by their load management, increasing internal generation or by drawing power from the national grid, he said. “It’s POSOCO which has had to prepare every evening.” 

This approach, he said, is running out of steam. The share of renewables is rising beyond what the national grid controller can manage. According to solar sector executives, India’s solar generation will soon spike to 160 GW. At that point, given solar’s must-run status, grid controllers will have to find takers for that 160 GW in the day and then find an equivalent amount of replacement power each day.

Even if the supply spike can be managed — lower tariffs in the day might move farm-, EV-charging-demand, etc, from the night to the day — can India find 160 GW of replacement power every evening?

Hydel is the best way to balance power supply and demand, but India’s hydropower capacity stands at just 42 GW. Given the biggest reservoirs in the country belong to irrigation and drinking water projects — and their need to hold water for longer — the actual number would be lower yet. Pumped storage is nascent right now. Further, even if India meets its pumped storage installed capacity target of 96 GW by 2030, that will still leave 64 GW unaddressed.

Discoms, said the former executive director, will have to create more resources for managing intermittency on their own. “This can be greater forecasting tools’ usage, their own storage/grid backup,” he said. The alternative, he added, will be penal rates for POSOCO power.

A few days later, trying to understand the options before discoms, CarbonCopy also met Abhimanyu Gartia, a senior advisor to Odisha’s Gridco, and yet earlier  an executive director at POSOCO. Gartia listed four options before grid controllers — load-shifting; changing thermal generation minimums; running thermal in two shifts; and storage.

“Load which is not time-dependent needs to be shifted to solar hours by charging lower rates,” he said. “Say around ₹2.5 to ₹3 per unit as compared to the rest of the day.” 

As for thermal power generation minimums, as solar generation rises, grid controllers will have to proportionately suppress power production by other generators, mainly thermal power plants to keep the grid parameters within limits. They might have to lower the technical minimum by running two shifts a day or through other technically possible methods that let the plant operate at lower output. In addition to these, there is storage — be it potential for kinetic (pumped storage), molecular (hydrogen saved as ammonia), or chemical (batteries).

That is the framework. The question is: Are India’s discoms up to such a task?

The hidden costs of decarbonised power

Of these options, load-shifting might be the easiest for a discom.

Storage, too, is in the hands of state discoms. Changing generation minimums for non-renewable plants, however, will require action from the Ministry of Power.

All states, however, do not have equal options for storage. Pumped storage, for instance, has more potential in Maharashtra and Karnataka than in Jharkhand. The latter could try alternatives — like electrolysers or batteries — but for one niggle. 

The financial state of most Indian discoms is dire. By 2022-23, as a NIPFP report noted, “State-owned, public electricity distribution companies… had collectively accumulated losses of around ₹6.77 lakh crores.” Amongst the loss-leaders are some of India’s largest discoms, added the report. “By March 2023, Tamil Nadu reported losses over ₹1.6 lakh crores… (discoms) in Uttar Pradesh and Rajasthan reported losses over ₹90,000 crores each,” it said. 

The reason extends beyond old explanations like power theft, sluggish tariff revisions, and wanton capital expenditure. Discoms also subsidise the country’s renewable power producers.

“Back in 2011-12, when the Jawaharlal Nehru Solar Mission was launched, the thinking was that ‘Green is good and should be supported’. It’s like saying a baby is growing and needs support. The catch is: the baby has grown up but we are still supporting.”

Kapil Mohan, a former additional chief secretary, who has worked at both centre- and state-levels in the electricity sector.

Unlike other generators in the country, solar developers get a transmission loss waiver. Given the electrical principle of resistance, a part of the power a generator puts into the grid is lost by the time it reaches its customer. An injection of 5 MW will dip to 4.6 MW or so in central transmission alone. RE, however, enjoys a transmission loss waiver, said Gartia. “If you put in 5 MW at source/injection point, you can take out 5 MW at sink/load point.” There is also transmission charge, levied by the central transmission utility (CTU) and state transmission utilities (STU). “Central transmission charges vary every month as per the monthly calculation by CTU (presently around 50 paise/unit) and STU charges as per appropriate state regulation or a default value of  ₹80/MWh,” Gartia said. This, too, has been exempted for many states in the case of renewables.

Apart from these, there is the cost of intermittency. Grid frequency should be between 48-52 hertz. If demand is higher, frequency will drop, resulting in brownouts and blackouts. If supply is more, frequency rises, the grid gets overwhelmed, and something — the transformer or the substation — will collapse. And so, given the unpredictability of renewable power, discoms have to make backup arrangements.

So far, that response has been to keep thermal power on standby, paying generators the fixed cost component of the agreed tariff of their coal-based power plants. In other words, a discom paying ₹2.5 for solar is also paying extra for standby thermal power. “This subsidy may add up to as much as ₹45,000 crore for Karnataka alone,” said Kapil Mohan, a former additional chief secretary, who has worked at both centre- and state-levels in the electricity sector.

Renewables’s must-run status is another cost. “I can tell hydel and coal to stop but not solar because it has been granted ‘must run’ status,” he said. “The argument is that hydel can bank its power. So can coal. But solar cannot. In that sense, solar generates without control. But once they buy that power, if it is surplus to requirement, discoms have to sell at any rate — even 50 paisa a unit — but their PPA might have them paying the generator ₹2.50.”

At the same time, with spikes in solar generation resulting in lower thermal and hydel generation, discoms end up with fixed cost recovery charges.

The cost of all these subsidies falls on the wider electricity system. Thermal power producers supplant renewables’ transmission losses. Discoms either absorb transmission charges and the cost of standby power and slip into loss, at which point tax-payers bail them out. Or levy higher tariffs on customers. Indians, as this report says, pay the highest power tariff in South Asia. 

It’s a suboptimal state of affairs. Given these hidden costs, the effective price Indians pay for decarbonised power goes up. Given their straitened economics, discoms don’t buy all the renewable power tendered out by central bodies like SECI. India’s adoption of renewables, ergo, slows.

The generators, however, make supernormal profits.

How did we get here?

Discoms in crisis

Blame it on outdated policy. “Back in 2011-12, when the Jawaharlal Nehru Solar Mission was launched, the thinking was that ‘‘Green is good and should be supported’,” said Mohan. “It’s like saying a baby is growing and needs support. The catch is: the baby has grown up but we are still supporting.”

As things stand, discoms’ financial health is set to worsen further. Until now, India has followed a model of cross-subsidisation where the biggest power customers pay the highest rates. “Some users get free power while others pay a lot,” as Gartia said.

This has created a space for private gencos, which have begun weaning commercial and industrial users away from discoms. “The high end consumers like industry/commercial units in many states where energy rate is around ₹12/unit have moved out of DISCOM to become Open Access customers for availing lower tariff power,” said Gartia. “And so, they have begun buying from private generators at lower rates, after paying the additional transmission charge and transmission loss.” With their exit, discoms have begun losing crores per day.

What this means for discoms is easy enough to understand. They will slip deeper into losses. Their capacity to prepare for a future with a preponderance of RE — 450 GW by 2030; 50% of electricity from non-fossil fuels by 2050; net-zero by 2070 — will be further impaired. In addition, as books like California Burning show, weak discoms might cut back on maintenance, exposing people to all the ancillary costs that follow.

Privatisation is not the answer here. Even a private discom will have similar economics. 

“Generation companies merely sell the power they produce,” Mohan said. “It’s the discom which has to balance commercial interests and equity.” 

For this reason, he said, the financial state of the discom is the financial state of a country’s electricity system. A recast is needed. 

The need for a recast

RTC tenders are one way to move the risk of intermittency to generators. And yet, that is not the complete solution. RTC tenders, for one, turn generation into a game where conglomerates will best standalone players. Also, other forms of subsidy like transmission loss waivers continue unhindered.

A deeper recast is needed. “Tariffs have to be controlled for C&I [commercial and industrial] users ,” said a former Union power secretary on the condition of anonymity. “At the same time, you cannot give free power to everyone. We see some of this change. States have been redefining the amount of free power — that’s come down from 300 units to 120 units etc.”

In addition, the calculus of risk and rewards has to be better balanced. When CarbonCopy asked the former power secretary about the REMCL tenders mentioned in the first story in this series, he said “Too much is being made of those bids. We cannot see the point of generation in isolation. We have to also look at the larger impact of solar on the power system.”

He spoke about the recast that is needed. “The PPAs being signed are not sustainable,” he said. “The industry can no longer have a ‘must run’ status. That is not sustainable for GW scale utilities. No discom can handle so much RE related risk. We now need to move to market-based dispatches.”

What India has right now are inflexible contracts for coal and solar, he said. “There is no incentive to put the system first. The country has to move towards a future where developers have to adapt to demand. Right now, they are dumping power during the day. They should be made to store energy. We now need market reforms for contracting where the generator is exposed to some of the market risk as well. Where they have to do some planning as well.”

In the absence of such measures, India’s discoms are taking more dramatic steps.
Aiming to shore up its own storage capacities, Tamil Nadu will now charge ₹50 lakh per MW for wind projects that connect to the central transmission infrastructure.

Electricity generators themselves are not acting in a manner that suggests solar has gained ascendancy over coal.“Competition to pick up stranded thermal power plants from bankruptcy courts has intensified,” said Ashish Gupta of Axis AMC. Indeed, recent resolutions— like Lanco Amarkantak, KSK Mahanadi and Coastal Energen — have seen stranded assets change hands with high recoveries for banks — a marked change from the past.

Hardwired into all these processes is a snapshot of this ongoing moment in India’s solar sector. Even as sector executives turn increasingly bullish — on the back of events like the REMCL tenders — the larger picture is still developing. It remains to be seen how import tariffs and domestic module manufacturing will affect module prices. It is also unclear if discoms can make adequate provisions to boost their solar energy intake. 

As the cliche goes, wait and watch. 

Read Part 1, Part 2

About The Author

M Rajshekhar

See author's posts