
Skill India exposed as a non-starter in CAG audit of Rs 10,000-cr spend
Flagship skilling mission found faltering on training, jobs and data integrity; CAG uncovers ghost accounts, dubious certifications, poor oversight across states
The Pradhan Mantri Kaushal Vikas Yojana (PMKVY), the flagship scheme of the Skill India Mission launched in 2015 and relaunched thrice since then, has failed to upskill the Indian workforce.
Telltale signs were visible for long in other official data but, this time, the Comptroller and Auditor General of India (CAG) has nailed its astounding failures in a report tabled in Parliament on December 18.
The CAG report shows the scheme failed for the same reasons that it failed in its previous avatar under the UPA government. It was then called the National Skill Development Mission; to implement it, the National Skill Development Corporation (NSDC) was set up in 2008.
Watch/Read: MGNREGA renaming row: Why dropping Gandhi is more than a name change
Renamed and relaunched as Skill India in 2015, the scheme set out to train 500 million by 2022. However, the CAG report says, it claims to have trained only 13.2 million, of which 11 million were certified. The number of job placements is too sketchy to be counted. But then, even the number of trained and certified is unreliable.
Facts and findings
Before looking at the CAG’s findings, here are a couple of facts to keep in mind:
a) The first two phases of Skill India were run during 2015-2020, the third during 2020-2022. The CAG report covers only these three phases. The fourth phase was launched in March 2022.
b) Skill India provides short-term training (STT), special project (SP) to suit specific needs of industry and recognition of prior learning (RPL). Certified candidates get a reward of Rs 500 through direct cash transfers (DBT).
The CAG reviewed the performances in eight states — Assam, Bihar, Jharkhand, Kerala, Maharashtra, Odisha, Rajasthan and Uttar Pradesh — for the period 2015-22.
In planning areas, it found the following:
1. Absence of micro-level skill-gap information, long-term implementation strategy and a National Skill Development Plan. Hence, sectoral and state-wise training were “not in line” with skill-gap studies.
2. Convergence between the Centre and states was “not effective”, “even after” the completion of three phases in 2022.
Also read: Why Centre urging preference for ex-Agniveers in security jobs has triggered fear in Haryana
3. The apex level framework for skill quality assurance and regulation, to be implemented through the National Council for Vocational Education and Training (NCVET), “was still in the process of establishment, and it performed “only limited regulatory roles.”
4. “Absence of policy” for keeping crucial data/information, like photographic/video evidence of training, records of education/work experience. In addition to that, controls regarding electronic identity, contact and account details of trainers, assessors, and candidates were “not effective”.
As a result, the following implementation issues were noted:
5. Enrolments were done ignoring the prescribed specific criteria like age, education, work experience. Also, the mechanism for onboarding/verifying targeted beneficiaries, that is, unemployed youth and school/college dropouts, was not available either.
6. Of the 56.14 lakh candidates under STT/SP, only 23.18 lakh (41 per cent) were placed. In Kerala, “incorrect placement documents” were produced “as proof of placement of candidates”. On the other hand, in Uttar Pradesh, though the Skill India Portal said 12,616 candidates had found placement, “no record was found with the state agency”.
7. In RPL-BICE (assessment and certification of skills in large and medium industries in their own premises) “several irregularities” were found in the selection of agencies, scrutiny of proposal, implementation and monitoring process. Documents relating to training provided and monitored by the NSDC “were unreliable”.
Also read: TN Health Skills Summit 2025 focuses on digital healthcare future
8. Though 33 Sector Skill Councils (SSCs) participated in certification (RPL-BICE), 34.93 per cent of the certifications were in the media and entertainment and green jobs. The Ministry of Skill Development and Entrepreneurship (MSDE) said media and other SSCs had engaged small employers which were of “non-descriptive nature” and produced “questionable evidence of their employees”. Three out of 10 training centres in Bihar and one of the 17 in Odisha were closed at the time of the survey.
9. Funds were released in violation of Receipt and Payment Rules.
10. In RPL (recognition of prior learning), the Aadhaar-linked attendance system (AEBAS) was "not being followed”. An inspection of Training Centres (TCs) revealed that in 24 TCs, biometric devices were “either not installed or were not working”.
11. Same photographs were presented as evidence “for different trainings”.
12. Preference shifted from training to RPL and excess certifications given in Phase 2 (2016-20) as targets and funding for actual training were cut down because of “financial constraints”. Massive irregularities were found in the payment of Rs 500, through DBT, to the certified candidates.
13. Though providing contact details like mobile number, bank account and email was “mandatory” while registering in the Skill India Portal (SIP), during Phase 2.0 and 3.0, “94.53 per cent” (90.7 lakh of 95.9 lakh) bank accounts “contained zeros, ‘null’, ‘N/A’ or was blank”. Of the rest, 12,122 bank account numbers were repeated for 52,381 participants in two or more instances. In cases of one account for one candidate, “apparent wrong account numbers were noticed” like ‘11111111111…’, ‘123456……’, single-digit account numbers, or just text, names, address, or special characters.
14. When verified through online surveys, 36.5 per cent emails showed “delivery failure”. In cases where emails were delivered, responses came from “only 3.95 per cent”, of which 76.6 per cent “belonged to the same email ID or were from email IDs belonging to TP/TC concerned”.
These findings suggest that the MSDE’s claims about training, certification and placement are entirely fictitious and very few, if any, were actually skilled. How else would the bank accounts of certified candidates to which direct cash transfer payments were made be so high (far higher than “94.53 per cent”)?
Also read | Sabka Bima Bill: Why 100 pc FDI push may not fix India's insurance woes
The CAG says the ministry gave an assurance that from the fourth phase (March 2022), it had set its house in order. But that would be known when the CAG revisits it.
The house should have been put in order long ago. Not after the spending of ₹9,261 (of the ₹14,450 crore outlay, ₹10,194 crore was released). Yet, in February 2025, the Union Cabinet approved a further ₹8,800 crore to restructure the mission, even though the details of the restructuring remain unclear.
Did it skill anyone at all?
In 2017, a Press Information Bureau (PIB) report had said: “So far, only 4.69 per cent of the total work force in India has received any formal skill training.” In 2024, the Economic Survey of 2023-24 said, “Only 4.4 per cent of 15-29-year-olds have formal vocation/technical training”. Since the latter didn’t specify the percentage of the total workforce with formal training, it isn’t possible to compare and know the actual gains.
Perhaps because of Skill India’s utter failure, coherent data on skilling are not released.
These failures that the CAG finds are not very different from those of the UPA era.
In 2017, the Sharada Prasad Committee had reported its findings but those are not in the public domain anymore. It had said, “₹2,500 crore of public funds was spent to benefit the private sector without serving the twin purposes of meeting the exact skill needs of the industry and providing employment to youth at decent wages.”
Also Read: Bharat Taxi bets on Amul-style cooperative model to challenge Uber and Ola
Over the years, media reports have pointed out that employers, even in Delhi-NCR, are unwilling to pay a premium for skilled or certified workers. On the shop floor, there were negligible differences in starting wages across qualification levels, which reduced interest in skilling schemes. Many participants also cited short-duration courses, inappropriate skill training, and a lack of job opportunities as factors that hampered the Skill India Mission.
PPP to Central scheme
On May 17, 2025, NITI Aayog CEO BVR Subrahmanyam suggested that the “government should consider handing over skilling institutions to the private sector.” This is a reversal in the Centre’s policy.
When the NSDC was launched in 2008, it was in PPP mode, with the Centre putting in 49 per cent capital and the private sector (industry bodies), the remaining 51 per cent.
But, from 2016, when Skill India 2.0 was launched, it was turned into (a) a Centrally Sponsored Centrally Managed (CSCM) in which 75 per cent fund and targets were allocated for skilling through the NSDC and (b) a Centrally Sponsored State Managed (CSSM) in which the rest 25 per cent of the scheme fund and targets were allocated through the State Skill Development Missions.

