Blog: From NPAs to PhDs ?

Friday, 30 October 2026

From NPAs to PhDs ?

The news

The Centre is getting ready to ask IITs, NITs and other top technical institutes to start courses in nuclear technology, to power India's dream of 100 GW of nuclear capacity by 2047.

The numbers behind this move :

India will need about 2.5 lakh skilled nuclear professionals over the next 20 years
Every GW of nuclear capacity needs roughly 2,000 skilled workers
The Central Electricity Authority estimates 13,600 new professionals by 2032 alone
An expert committee led by nuclear scientist Dr J B Joshi is designing the curriculum — from diplomas to PhDs — and there is talk of a dedicated Indian Institute of Nuclear Technology
Till now, nuclear professionals have been trained almost entirely in-house by BARC and DAE
The push follows the SHANTI Act, which opens civil nuclear power to private players

Today, India's installed nuclear capacity is about 8.8 GW. To reach 100 GW in 21 years, we must add more than 11 times what we have built in the last 55 years.

Sources : Economic Times (30 Oct 2026), NewsBytes, Whalesbook, Industrial Info Resources

What I wrote 9 years ago

In May 2017, the media hailed the Government's "bold decision" to set up 10 nuclear power plants of 7,000 MW at a cost of Rs 70,000 crore, in "fleet mode".

I asked whether it was a bold decision or a bad one :

Nuclear power would cost consumers Rs 5–7 per unit. How would it compete with solar at Rs 2 per unit ?
A nuclear plant costs about Rs 10 crore per MW, against Rs 5 crore per MW for solar
Nuclear plants take 10 years to build, against 1 year for the same solar capacity
If the economics are so poor, which bank will finance these plants, knowing the Rs 70,000 crore could turn into NPAs even before the plants go on stream ?
And what about the safety fears of people living near the sites ? Have we forgotten Kudankulam and Jaitapur ?

I ended with an appeal to Shri Piyush Goyal (then Power Minister) to take a second look.

Then vs Now

My 2017 concern	What actually happened
Solar will touch Rs 2 per unit	Solar hit Rs 2.44 within days of my blog, and fell to Rs 1.99 per unit in a Gujarat auction in 2020
Nuclear takes 10 years to build	The 10 fleet-mode reactors were approved in June 2017; the first concrete was poured only in 2023, and each needs about 5 years after that
Cost of Rs 70,000 crore	The same 10 reactors were later estimated at Rs 1.05 lakh crore — a 50% jump before a single unit runs
Solar can be built in a year	India installed a record 36.6 GW of solar in 2025 alone — four times its entire nuclear fleet, in one year
Solar cannot give round-the-clock power (the nuclear lobby's reply)	Solar-plus-storage now delivers firm, dispatchable power below Rs 3 per unit (Morena, MP)

I must be fair : nuclear still offers what solar alone cannot — steady, low-carbon base-load power, on very little land. That is a real argument, and it is why the 100 GW goal is not foolish.

But my real point in 2017 was about economics and execution. On both counts, the record of the last 9 years supports caution.

Now the bottleneck is PEOPLE

In 2017 the worry was money. In 2026 the worry is manpower.

We can import reactor designs. We can borrow money. But we cannot import 2.5 lakh trained, security-cleared, safety-conscious nuclear workers.

And a nuclear plant that is built but cannot be staffed is the surest route to an NPA !

So I welcome this move to bring IITs and NITs into nuclear education. But let us not repeat the old mistake of treating this as an elite-only project.

My suggestions

Most of the 2.5 lakh are not PhDs — They are welders, fitters, instrument technicians, radiation-safety supervisors and control-room operators. Start nuclear trades at ITIs and polytechnics near the reactor sites, and let them award Bachelor / Master of Skill degrees (as I have proposed earlier), so that these jobs carry dignity and career growth
Train where the plants are — Tie each new reactor site (Gorakhpur, Kaiga, Chutka, Mahi Banswara and the rest) to a nearby NIT or polytechnic. Local youth trained for local plants will also ease the safety fears and land protests that I mentioned in 2017
Bring back our own — Hundreds of Indian nuclear engineers work abroad. A focused talent-return scheme (on the lines of my MART / STAR Act proposal) can fill senior gaps faster than any new course
Let private players pay for the pipeline — Under the SHANTI Act, private companies will profit from nuclear power. Make every licence-holder sponsor seats and apprenticeships in proportion to the MW they plan to build
No more NPAs — Before any bank lends to a private nuclear project, require an independent, published cost-and-schedule audit, and a comparison with the cost of solar-plus-storage power for the same buyer. Let DISCOMs choose with open eyes
Measure progress honestly — Publish a quarterly "Nuclear Talent Dashboard" : seats sanctioned, students enrolled, graduates placed, plants staffed. What gets measured, gets done

Closing thought

In 2017, I asked : Who will finance these plants ?

In 2026, the question has changed : Who will run them ?

If we answer the second question well — with ITIs as much as IITs — we may yet avoid the first problem. If we don't, the 100 GW dream may give us not power, but NPAs and empty control rooms.

Dear Policy Makers at DAE and NITI Aayog : please make sure the talent pipeline is built before the reactors, not after.

With regards,

Hemen Parekh

www.HemenParekh.ai / www.IndiaAGI.ai

=============================================

Resources :

30 October 2026

Related reading : NPAs in the making ? (20 May 2017)

Translate