Moving beyond L1/H1

1994 was the year that changed the way governments sell or buy resources. In that year, the United States Federal Communications Commission introduced a new way to sell radio spectrum to private telecom operators, based on the advice of, among others, economists Paul Milgrom and Robert Wilson. The duo won the Nobel Prize for economics this year for the kind of work that led to that landmark 1994 auction. Since 1994, over $200 billion has been raised by governments through such auctions. The research by Dr Milgrom and Dr Wilson have also transformed the ways in which governments across the world procure goods and services from private parties, and not just how they sell them.

Yet much of this path-breaking research has passed this country by. Much of the procurement by the government in India, in both infrastructure and non-infrastructure contexts, has tended to follow the L1/H1 bidding process where the lowest/highest bidder wins the right to fulfill the contract, such as the right to supply power. Yet anyone with even a passing familiarity with the infrastructure and related sectors, as it has functioned over the last three decades, knows that even though competitive bidding has brought substantial benefits, it is still beset by problems.

There are a number of institutional reasons why our bidding and tendering processes have problems, but at least part of the reason goes back to the fundamental research done by Dr Milgrom and Dr Wilson on how auctions should be conducted and the rules which an auction process must follow.

For instance, are bids open or closed? What price does the winner pay — their own bid price, or the second highest bid? Even the question of who actually should be the winner of an auction (not necessarily the highest or the lowest bidder) is an aspect explored in the research.

The core of the problem goes back to the information available to each bidder about the final value of the contract or project, and how they (individually) value that contract. For instance, a government may invite bids on a road construction project. The actual cash flows (from toll revenues, or even from the final sale) over the life of the project are likely to be the same, irrespective of who wins. But every bidder’s assessment of that net value of the project, including costs, will be different at the time of bidding — and which could change if competing companies could learn each other’s estimates of that “common value”. Research has shown that in such circumstances, and especially given an L1/H1 auction, it is highly likely that the winner will be the company that overestimates the value of the project and bids too aggressively, only to regret it later. This is what we have seen again and again across infra projects.

Why should such aggressive bidding be a problem for the “seller” — in this case, the government? After all, if a private party is too optimistic about the value of a project, say, a power procurement contract, that would be their lookout — the public will benefit from tariffs that are as low as possible.

But as we have often seen in reality, over-aggressive bids lead to poor quality work on the projects, difficulties in financial closure, delayed execution, or worse, early attempts to renegotiate the contract. Ultimately, the government, and the public lose out. How do you design an auction that ensures that bidders don’t overplay their hand?

How do you create rules that ensure that all bidders have access to as much accurate information about the project as possible?

Research on auction design predates Dr Milgrom and Dr Wilson — most famously in the work of William Vickrey, who also won the Nobel Prize. But Dr Milgrom and Dr Wilson extended that work to cover more complex and realistic scenarios described above, and adverse outcomes such as the winner’s curse (that is the tendency to bid too much). For instance, Dr Wilson argued, in a series of papers as far back as the 1970s, that in auction formats similar to L1/H1, the possibility of a winner’s curse leads to serious problems.

Others, following on their work have deepened our understanding of the best practices for government procurement, which can vary according to specific circumstances. This has led to reforms in a number of countries in public procurement processes — for instance, some public authorities across the world follow the so-called Average Bid auctions to eliminate the winner’s curse altogether —the winner is not the “best” bidder, but the one who bids closest to the average of all the bids, the idea being that the “average” bidder has made a better estimate of the value of the project.

The bigger takeaway from this year’s Nobel winners is that an overhaul of public procurement processes in India is long overdue. It’s time that economic practice in India, in the field of public procurement, caught up with the state of the art in theory, as well as practice.

The writer is the chairman of Feedback Infra