Scams, logic and the CAG

Having known Vinod Rai, the CAG, on and off since 1967 when as a third-year student he ragged me with civilised disdain and genuine languor, both accompanied by an open contempt for the bad language used routinely by the others, I can say at least this much: he means well.

It is also not the purpose of this article to put up a defence for the government in any of the things pointed out by Rai. For all we know, regardless of what Rai says, ministers have been up to no good. They usually aren’t.

Also, as so many other civil servants have pointed out after the 2G affair, whether the policy was appropriate or not is a matter of opinion; but whether it was implemented correctly is a matter of fact.

But having entered all the important caveats, it does seem to me that the time has come to apply the rules of formal logic to the method adopted in the CAG’s reports.

Oftentimes, when you apply these rules to reports prepared by serving and former government employees, they appear less convincing than at first sight.

CAG’s method

The method used by the CAG is known as ‘statement calculus’ in logic. It seeks to establish causal relationships between two statements. Here are some examples.

1. If Ram is a thief, I will eat my hat.

2. Clear air turbulence leads aircraft to wobble in flight.

3. If there had been an auction, the exchequer would have gained.

Implicational relationships of this type between two statements are required by logicians to fulfil two conditions in order for the implication to be true: necessity and sufficiency. This rule is also used by good economists a lot.

In a nutshell, the rule says that for the second statement to be true, the first must be true as well.

Thus, anyone can assert that A must happen before B can happen. Or that whenever A is true, B is also true.

But more often than we like, the assumption can be a false one.

Specifically in the case of auctions and losses to the exchequer, Rai is assuming that auctions always lead to superior results than discretionary allocations.

This is not borne out by the evidence. A policy of auctions is not inevitably and inherently superior to a policy of discretionary allocations. Both can and do lead to sub-optimal outcomes.

In fact, the CAG’s method suffers from another flaw: it assumes that an auction is both necessary and sufficient for the best outcome to be achieved.

This is most certainly not true. It may be necessary but it is certainly not sufficient because collusion between bidders is always a danger.

Design Vs discretion

The key to a successful auction, as is well established now, is very precise design and zero communication between the different players, including the auctioneer. It is also well established that there is no ‘one size fits all’ design. Each auction has to be separately designed by experts.

As for discretionary allocations, the problem in India is not so much the discretion itself as the practice of demanding a bribe for using that discretion correctly.

This has been happening since the early 1980s when, strangely enough, it started in the coal sector. Bofors was another good an example of this.

In sum, the CAG therefore may not be quite right in presenting the case in the manner it has chosen. It means well but is perhaps using the wrong method.

COMMENT NOW