Licences handed out in a day
Published 9 October 2026
The CAG's account of 10 January 2008 makes the compressed procedure measurable. It said 232 applications from 21 applicant companies had been received by the revised cutoff and 121 applications from 16 companies had been found eligible. A press release at 2:45 PM asked applicants to assemble at department headquarters by 3:30 PM to collect responses. The report said acceptance of 120 applications was received that day, compliance with letter-of-intent conditions was completed for 78 applications that day, and the rest followed the next day. It contrasted this speed with the 100-550 days the department had taken to process applications and the 30-day disposal period specified in the procedure. An application received, a letter issued, acceptance of the letter and compliance with its conditions were separate events. The audit's point was that the department moved from prolonged processing to exceptionally compressed applicant action precisely when compliance speed had become decisive for priority. That change was capable of favouring an applicant prepared for the new event over an applicant who had filed earlier but had less immediate access to finance or information. (CAG Report No. 19 of 2010; CAG Report No. 19 of 2010 (NDTV-hosted full copy))
The audit identified 13 applicants with demand drafts drawn before the cutoff notification. It also noted a performance bank guarantee and financial bank guarantee prepared in Mumbai on 10 January and submitted to the ministry that day. From this, it inferred advance information enabling preparation for compliance when the usual 15-day period had effectively become a matter of hours. The criminal trial examined precisely such questions through testimony and documents, rather than adopting the audit's inference as conclusive proof. The High Court's 2024 leave order then identified the treatment of the queue and timing as a matter deserving deeper review. The chronology therefore has three evidentiary levels: the audit's account of preparation and speed, the trial court's assessment of proof against particular accused, and the appellate court's preliminary concerns. A detailed reconstruction can show how the same timing evidence carried different weight at those stages without pretending that an audit inference was a conviction or that an acquittal removed the recorded timetable. (CAG Report No. 19 of 2010 (NDTV-hosted full copy); Special CBI court judgment, 21 December 2017; Delhi High Court leave order, 22 March 2024)
The criminal judgment reproduced the department's 10 January note describing how letters and responses would be distributed through four counters. The note listed companies in application-date order, then placed them into parallel counter queues. By Cell, Swan, Datacom and Shyam were listed at one counter; Tata, HFCL, Loop and Selene at another; Idea, S Tel and Allianz at a third; Spice and Unitech at the fourth. This arrangement meant that application-date seniority and the physical distribution process were not the same simple single-file queue. Judge Saini's concern in paragraph 923 was the absence of a written record establishing when and how the earlier decision about distribution had been taken. He identified the 10 January note as the written record that existed. The exact scope matters: it was not a finding that no distribution note existed anywhere, but a criticism of the record linking the earlier decision to the chosen arrangement. The note itself also distinguished companies receiving responses from those receiving licences, and counted separate applications across service areas. Describing the episode only as a crowd racing for licences loses those operational details. The four-counter arrangement supplied the physical setting in which rapid letter collection and compliance could interact with a changed priority rule. (Special CBI court judgment, 21 December 2017)
The CAG said 85 of 122 licences issued in 2008 went to companies that did not meet basic eligibility requirements. That count is of licences across service areas, not 85 separate companies or 85 criminal convictions. The report's eligibility analysis examined matters including registered changes to company objects, authorised and paid-up capital, and restrictions on cross-holdings. It distinguished an applicant's state on the application date from later documents and changes. The company tables illustrate how one applicant could produce many disputed licences: Datacom's row covered 21 licences, S Tel's six and Swan's 13. A name change or merger also meant that an original applicant could later appear under another commercial identity. An accurate inventory therefore needs an application entity, date, service-area count and specific alleged deficiency rather than only the consumer brand. The audit's concern was the department's diligence in checking its own eligibility conditions. Whether a defect was legally material, curable, knowingly suppressed or criminally ignored was subsequently disputed in the trial. The headline 85 cannot itself answer those questions. (CAG Report No. 19 of 2010; CAG Report No. 19 of 2010 (NDTV-hosted full copy))
Another 85 appears in the same report's rollout discussion, with a different meaning. Paragraph 4.11 said 85 licences had been awarded to six new entrants: Unitech, Swan, Allianz, Shipping Stop Dot Com, Datacom and S Tel, including the later identities mentioned by the report. The audit said these operators had obtained initial spectrum in 81 service areas between April 2008 and January 2009 but had not fulfilled the specified rollout obligations by 31 December 2009. It cited the licence conditions requiring coverage within 12 months and said the department had failed to recover Rs 679 crore in liquidated damages and penalties for delayed rollout. This is not another 85 that can be added to the eligibility 85, nor is Rs 679 crore an additional row in the headline presumptive-loss table. The rollout finding addresses performance after award; the eligibility finding addresses qualification for award. Their equal counts can hide that difference. The report also kept the rollout observation tied to a particular cutoff date. It does not establish that the operators never launched services, that all remained inactive indefinitely, or that the penalty had not later been recovered. Those later propositions require fresh records. The audit supplies a dated assessment of noncompliance and the department's enforcement failure at that stage. (CAG Report No. 19 of 2010; CAG Report No. 19 of 2010 (NDTV-hosted full copy))
The audit's dual-technology chapter addresses operators that already held licences using CDMA and sought permission to use GSM as well. It is not simply another name for the new-entrant licences issued in January 2008. The CAG recorded that the department asked TRAI in April 2007 for recommendations on combinations of technologies under one licence. TRAI's response permitted an existing licensee to request the alternative technology, with a fee matching the amount paid by existing or new licensees using that technology. The audit then criticised the implementation: the department decided on 17 October 2007, gave in-principle approvals to three earlier applicants on 18 October, and publicly announced the policy on 19 October. It contrasted their treatment with Tata's application after the announcement and later permission. The audit regarded the earlier approvals and priority attached to earlier requests as denying equal opportunity to similarly placed firms. These are allegations about access to a newly permitted route and the information and queue governing it. They require a separate chronology from the January letter-of-intent episode, even though both eventually affected the availability of the same scarce GSM spectrum. (CAG Report No. 19 of 2010; CAG Report No. 19 of 2010 (NDTV-hosted full copy))
The report said Reliance Communications deposited Rs 1,645 crore for permission in 20 service areas on the day the policy was announced, and that start-up spectrum was allocated in 14 service areas on 10 and 11 January 2008 after requests for six were withdrawn. It also described the timing of HFCL's and Shyam Telelink's spectrum assignments and delays for other applicants. The figures refer to different units: areas covered by permission, areas in which spectrum was subsequently assigned, and fee paid. They are not interchangeable licence counts. The CAG's larger loss table later used 35 dual-technology approvals, a broader category than the 20-area example in this passage. A technical policy allowing both GSM and CDMA does not by itself resolve the fairness of the dates and priority assigned to applicants. (CAG Report No. 19 of 2010; CAG Report No. 19 of 2010 (NDTV-hosted full copy))
The CAG separately criticised spectrum priority in Punjab and Maharashtra. Its Punjab example said 15 MHz was available in September 2008, enough for the first three applicants in the stated queue: HFCL, Idea and Unitech. Idea's request was not considered because of its proposed merger with Spice, allowing Swan, fourth in the queue, to receive spectrum. In Maharashtra, the audit said Spice was not allocated start-up spectrum because of its proposed merger with Idea, again benefiting Swan. The audit cited the merger guidelines, under which the post-merger licensee could hold the combined spectrum subject to meeting the prevailing allocation criterion within three months of approval. It concluded that refusing the initial allotments on merger grounds was inconsistent with the guidelines. The department replied that it had kept spectrum reserved for the merging firms and later assigned it when the merger did not finish. The audit rejected that explanation by referring to the amount available and the queue. This episode adds a concrete example of administrative discretion after the original licence award. It shows that the story was not exhausted by one January afternoon or one entry-fee decision. (CAG Report No. 19 of 2010; CAG Report No. 19 of 2010 (NDTV-hosted full copy))
In the criminal judgment, the judge examined allocation and eligibility questions through the prosecution's specific conspiracy theory and the evidence of roles attributed to particular accused. The two sources answer different questions about the same distribution system. The merger passage is useful because it explains an administrative mechanism with named applicants, a stated quantity and a particular rule. Its limits are equally useful: the report's date governs the facts it says remained outstanding, its interpretation of the merger rule remains the audit's interpretation, and any claim about subsequent compliance needs a later source. (CAG Report No. 19 of 2010 (NDTV-hosted full copy); Special CBI court judgment, 21 December 2017)
The audit's third main loss category concerned spectrum held above contracted quantities, not the new entrants' initial allotment alone. Paragraph 4.10 listed incumbent operators' additional holdings and criticised assignment without an upfront charge or a market-price determination while new applications remained pending for lack of spectrum. Its estimate of Rs 2,561 crore used the existing price benchmark, while the larger figure depended on charging additional holdings at the proposed later benchmark. The report discussed a May 2009 technical committee recommendation and TRAI's May 2010 recommendation, both of which proposed charging additional spectrum. The resulting Rs 36,993 crore figure was conditional on the recommendation being accepted. It was not a statement that the treasury already had an enforceable, collected receivable of that amount. What can be stated clearly is the category: additional holdings beyond the benchmark, valued separately from new licences and dual technology. This explains why the headline maximum cannot be described entirely as the loss on new licences. (CAG Report No. 19 of 2010; CAG Report No. 19 of 2010 (NDTV-hosted full copy))