Retendering Under Scrutiny: CAG Advisories and Global Lessons
Tenth Governing Council meeting of Niti Aayog in New Delhi (Image Niti Aayog)
A vendor performance scorecard that is driven by subjective assessments, unresolved disputes or biased feedback can lock smaller firms out of the market without meaningful appeal.
By P. SESH KUMAR
New Delhi, November 26, 2025 — Two recent advisories (November 2025) emanating from the CAG’s side seek to reshape the way India buys from the market. The first argues for a Unified Central Vendor Registry (UCVR) to end fragmented vendor registration, opaque performance histories and inconsistent blacklisting. The second calls out the rampant misuse of retendering, insisting that cancellation and rebid must be a rare, justified exception rather than a lazy default.
Taken together, they recognize the same malaise from two ends of the pipeline: weak vendor intelligence at the entry gate and undisciplined retendering at the award gate.
This note critically examines the assumptions, strengths and blind spots of both advisories, tests them against international good practices in the UN, World Bank, OECD and EU systems, and argues that India must convert these ideas into a formal doctrine of “intelligent retendering” and “trusted vendor identity” rather than a patchwork of instructions.
The way forward lies in codified triggers, digital audit trails, shared vendor performance data and a culture where both unjustified retendering and unjustified refusal to retender carry consequences.
The first advisory, on a Unified Central Vendor Registry (UCVR), springs from a very real pain point that every procurement practitioner in India recognizes. Today, a vendor who supplies computers or cement or consultancy to government has to prove its existence and credentials afresh in ministry after ministry, PSU after PSU, state after state.
Every portal asks for the same PAN, GST, financials and compliance affidavits, and every procuring entity re-verifies what has already been checked elsewhere. The paper notes that this duplication delays awards, raises transaction costs, and still fails to create a credible record of vendor history, because performance feedback remains trapped inside departmental silos.
The diagnosis is hard to dispute. Fragmented vendor management means no single picture of a supplier’s capability, experience, or default history. It invites fake credentials, because each tendering entity has to trust whatever experience certificates the bidder uploads.
It allows a contractor who botched one project to quietly win another project in a different department that has no visibility of earlier failures. The advisory looks at this landscape and says, in effect: stop playing procurement blindfolded; build a single national registry, give each vendor a unique ID, make basic documents and performance histories available across platforms, and move towards standardised vendor rating.
This direction is very much in line with global practice. The United Nations system has, over time, converged on United Nations Global Marketplace (UNGM) as a shared vendor gateway, where suppliers register once and are visible to multiple UN organizations, with standardized declarations and background checks.
The European Union’s European Single Procurement Document (ESPD) does something similar in spirit by creating a common self-declaration document that public buyers across member states can rely on instead of repeatedly demanding full supporting paperwork up front. Many UN entities and secretariats explicitly link vendor registration with performance evaluation policies, so that repeated delays, quality issues or integrity breaches are captured over time and inform future awards.
Viewed against this global canvas, the UCVR idea is neither radical nor risky in itself. It aligns with the ease-of-doing-business narrative and with the direction India has already taken through Government E-Market Place (GeM) and other digital procurement platforms.
The advisory is also realistic in suggesting a phased approach, beginning with infrastructure and the Union level before going fully pan-India. It wisely proposes deferring the integration of debarment or blacklisting into UCVR until a more harmonised legal framework emerges, acknowledging that today’s patchwork of blacklisting rules could create more confusion than clarity if dumped into a central database overnight.
Yet there are fault lines that a seasoned auditor should not ignore. A central registry can as easily entrench incumbents as it can discipline them, if rating frameworks are poorly designed or opaque.
A vendor performance scorecard that is driven by subjective assessments, unresolved disputes or biased feedback can lock smaller firms out of the market without meaningful appeal. A central system that shows only the “headline” rating but hides the underlying facts can quickly morph into reputational blacklisting by algorithm. And if debarment is kept outside the registry for too long, the worst offenders may still game the system through related entities and proxy firms.
International practice has begun to confront these issues. UN and EU regimes tend to separate vendor eligibility (basic registration) from performance and integrity sanctions, with clear procedural safeguards and due process for adverse actions, often including the possibility of review or appeal. OECD’s recent guidance on bid rigging and competition in procurement also emphasizes that data-driven tools should be designed not to cement cartels or discrimination, but to detect red flags and preserve open competition.
India’s UCVR design will need similar guardrails. A registry that centralizes data without centralizing standards, due process and oversight could become a new arena for vendor harassment rather than a remedy for fragmented vetting.
The second advisory, on retendering, tackles a different but equally chronic problem: the Indian habit of cancelling tenders at the slightest discomfort and treating retendering as a “safe” administrative reflex.
The Audit Watchdog Has Lost Its Bite — The CAG Must Reinvent
The note bluntly states that retendering has become a major bottleneck in Indian public procurement, and that it is used far more domestically than in externally aided projects, where World Bank or other multilateral rules are followed more strictly.
This contrast is telling; the same officers, working under tighter external rules, retender far less often.
The advisory then drives home a crucial point. Retendering should be exceptional, reserved for serious defects in the Notice Inviting Tender or Request for Proposal that cannot be cured within the ongoing process, or for significant post-tender changes in scope.
Problems that arise mid-stream should be resolved mid-stream, not kicked back to square one. Retendering, the note argues, breeds inefficiency, cost and time overruns, and opens doors to collusion and favouritism.
It cites Central Vigilance Commission’s (CVC) guidance and GFR 2017’s insistence that cancellations must not be arbitrary, and reminds readers that the Manual for Procurement of Goods warns against “unwarranted rebidding” on pretexts like high prices or budget changes.
The criticism of current practice is well-founded. Repeated tendering can, in fact, encourage collusive bidding as cartels learn to coordinate over multiple rounds.
It can allow a buyer to keep cancelling until a preferred vendor emerges or a higher price is achieved. It wastes bidders’ time, corrodes their trust and often leaves the government paying more, not less, because the market prices move up while the process loops.
These are not hypothetical risks; they are familiar patterns in sectors ranging from construction to health supplies. OECD’s latest update on fighting bid rigging explicitly cautions against tender designs and behaviours that inadvertently facilitate collusion, including predictable cancellations and repeated rebids that allow cartel members to adjust and rotate.
Where the advisory becomes especially interesting is in its treatment of “resultant single vendor” situations and allegedly high prices. It notes that procuring entities often cancel tenders when fewer than three bids are received or when only one bidder remains technically qualified, assuming that this lack of competition is itself a defect that must be cured by retendering.
The note pushes back hard, citing the Finance Ministry’s October 2021 guidelines that explicitly state that a single bid can be valid if the tender was adequately advertised, criteria were not restrictive and the price is reasonable vis-à-vis the market. This position again aligns with global practice.
The World Bank’s procurement regulations allow contract award on a single responsive bid under similar conditions, treating competition as a matter of process quality and market behaviour, not a mere tally of bid counts.
The advisory also skewers the mechanical use of internal cost estimates as cut-offs. It points out that cancellation is often triggered solely because the lowest evaluated bid exceeds the internal estimate, even though those estimates are meant to be indicative, not sacrosanct, and should be stress-tested against current market prices, regulatory tariffs, last purchase prices and changes in input costs.
Here, too, multilateral practice agrees: estimates are for budgeting and reasonableness checks, not as rigid ceilings, and bid prices must be analyzed, negotiated where possible, and justified rather than simply rejected.
So on paper, one can say that the retendering advisory is well-argued and broadly convergent with international standards. The problem lies in how it may be interpreted and operationalized on the ground. There is a risk that procurement officers and even vendors read this as “retendering is bad; avoid it at all costs”, rather than “retendering is a powerful but dangerous tool; use it only with clear justification and better diagnostics.” In other words, the pendulum could swing from unhealthy overuse to unhealthy underuse.
This is where the connection between the two advisories becomes critical. The UCVR seeks to improve the quality of information about vendors; the retendering advisory seeks to improve the quality of decisions about tenders.
Together, they hint at an integrated future in which procurement behaviour is driven by data and doctrine rather than by instinct and fear. A central registry with standardised vendor performance ratings and basic “one-time” vetting can help procuring entities diagnose whether poor response, high prices or repeated failures are due to market structure, vendor capacity, collusion, or flawed specifications.
That diagnosis can then inform whether retendering is the right remedy, or whether the cure lies instead in revising specifications, re-estimating costs, rethinking packaging, or addressing cartel risks in cooperation with competition authorities.
International systems are moving in exactly this direction. The UN’s procurement manual and UNGM architecture are designed to combine centralized vendor information with decentralized procurement decisions, supported by guidelines on when to cancel and rebid, when to negotiate, and when to rely on performance history rather than formalistic cut-offs.
The EU’s public procurement framework integrates digital tools, central documents like ESPD, and guidance to ensure that procedural rules do not unintentionally stifle competition or ease bid rigging. The OECD’s 2025 update on fighting bid rigging explicitly recommends that governments review tender cancellation practices, data-sharing arrangements and procurement design features to ensure they do not inadvertently support collusive strategies.
In that global mirror, the CAG’s two advisories look less like outliers and more like India catching up with the logic of the times. The challenge is to make them part of a coherent operational guidance rather than isolated notes.
They need to be embedded into a standardized, rule-backed, digitally enabled framework that every ministry, state and PSU can understand and apply consistently.
The way forward, therefore, is not merely to “accept” or “implement” these advisories but to elevate them into a formal doctrine of intelligent procurement. India must enshrine in its manuals and regulations that retendering is a measure of last resort, to be triggered only after structured diagnostics on competition, pricing, specifications and vendor capacity have been performed and recorded.
It must specify that single bids are not to be dismissed out of hand, but to be analysed against quality of advertisement, openness of criteria and market benchmarks. It must make it mandatory that every cancellation decision, like every refusal to cancel, be justified in writing and retained in a tamper-proof digital trail.
It must redesign training and audit expectations so that officers are not punished simply for awarding contracts in single bid or high-volatility markets, but are held to account for the quality of their reasoning.
Simultaneously, the UCVR concept must be sharpened with clear standards for vendor data, transparent performance rating methodologies, safeguard mechanisms for vendors to contest or clarify adverse assessments, and eventually a harmonized approach to debarment that balances deterrence with due process.
The registry should be used not as a blacklist in disguise but as a shared memory of the State’s experience with its suppliers, feeding into better risk assessment, better pre-qualification, and more informed tender design.
If India gets this right, the two advisories could mark the beginning of a transformation in how the Indian State sees its own purchasing power-not as a clerical chore to be navigated nervously between audit and vigilance, but as a strategic instrument that must be wielded with discipline, intelligence and courage.
A future in which officers are no longer terrified of both audit paras and vigilance queries, vendors are no longer exhausted by duplicative registrations and capricious cancellations, and citizens see public works completed on time at fair prices, is not a fantasy. It is the natural outcome of treating retendering and vendor vetting as two sides of the same governance coin, and designing operational guidance that leaves less to chance and more to principle.
(This is an opinion piece, and views expressed are those of the author only)
Follow The Raisina Hills on WhatsApp, Instagram, YouTube, Facebook, and LinkedIn
About The Author
