Public procurement: sorting out the mess

The first part of the article that appeared on April 10 explained the shortcomings in public procurement norms and suggested alternatives. The second and final part elaborates on how bidding procedures can be improved, so that the best contractors are rewarded and corruption is kept in check.

Rate contracts (also known as ‘framework agreements’) provide a very useful service by allowing purchase by government departments of various things at a fixed price without tenders. The rate is fixed through a competitive process by the Director General of Supplies & Disposals.

Currently, they are confined to goods; they need to be extended in a big way to services. ‘Classes’ of labour could be notified along with monthly / hourly / daily rates for each kind. Even higher level consultancy services can be incorporated in this by creating a separate classification.

The setting of the rates should be done by an independent agency after following a proper competitive process similar to that used for goods. Departments can determine the quantity of labour involved and buy services from the accredited companies who are under the rate contract.

A system should be introduced where service providers can register with one central authority, get classified by size or technical area, present / update their credentials, pay an annual Earnest Money Deposit (EMD) and then be eligible to submit bids to any government tender (state or central) without requirements of either verification of their qualifications or payment of EMD.

Software companies, consultancy firms and so on could be divided into, say, three or four classes for each category based on their size, turnover and so on, and pay a lumpsum EMD reflecting a reasonable security amount for the number of bids they may be simultaneously involved in at one time.

Registered contractors should be automatically qualified for all contracts of the relevant class / size without any separate verification of their qualification criteria or payment of EMD.

Only unregistered tenderers would go through detailed scrutiny of eligibility and payment of EMD. The main benefits of such a system would be speed (shortens or eliminates verification stage), reduced overheads for government (less work to verify credentials, preserve and refund EMDs) reduced costs for bidders (who need not lock up money in EMDs in multiple projects all over the country and chase refunds), and consequent lower prices for the government.

Currently, there is a natural predilection to delay or defer a procurement at the first sign of a problem. The implicit and sometimes unrecognised assumption is that doing a fresh procurement will improve matters. The reality is that delay incurs costs, too: first, the service is not provided and the public suffers thereby; second, there may be lost revenues; and third the cost of the project is likely to increase with inflation (though this may not apply to equipment where prices often fall).

The General Financial Rules (GFR) should provide that if the decision-takers decide to re-tender despite procedurally compliant bids being available, they should explicitly consider the costs of non-acceptance and resulting delay and record reasons.

The purpose of this is to force decision-makers to consider the costs of inaction rather than focus only on the risks of action, and increase for them the danger that the deferral may have to be defended. This will also require external agencies such as audit and vigilance to factor in this consideration. Overall, the aim is to reduce the tendency of being indecisive where only decisions are questioned but indecision is not.

A defect arising from the ‘procedural fairness’ obligations of government, is perverse incentives. If contractors do a good job, they cannot be given more work except through a tender which they may not win. If they do bad work, they continue to be eligible for new work unless formally blacklisted — which is very difficult to do and generally stayed by courts for years even if done properly.

Thus the government as a buyer is handicapped vis-à-vis private sector buyers. This can be mitigated by introducing a vendor rating system. After each contract, the organisation for which the work was done should provide an evaluation in a standard format to a central agency. The ratings should be confidential at the individual level but aggregated and disclosed. A problem faced by officers in procuring services is that it is not clear who is suitable and competent to advise on evaluating technical questions or awarding technical marks. Judging technical merit inevitably involves some value judgements.

Leaving ambiguity on this aspect allows the corrupt to pack committees with poorly chosen nominees while it leaves the honest with the risk that their discretionary evaluation may be questioned. Prescribing a standard committee composition for different kinds of services and then empanelling a sufficient number of technical experts who can serve on those committees will reduce this problem.

The writer, who is a director at the World Bank, is an IAS officer. The views are personal

COMMENT NOW