Auctions are everywhere these days— from the 2G (second-generation) spectrum auction that never was, to the recent televised Indian Premier League auction. Received wisdom has it that an auction is the most optimal way to allocate scarce resources among a group of bidders. Besides, astronomical sums, intricate rules and adrenalin-pumping speed make for a good show. But are auctions indeed the best way? Should they be employed for “selling" everything —from the right to use of land and mining areas, to collectables, talented cricketers, water and spectrum? Is price the best metric for sale of precious resources?

Also See Narayan Ramachandran’s previous columns

Imagine a public good such as the Lodhi Gardens in Delhi, the Juhu Beach in Mumbai or the Cubbon Park in Bangalore. Let’s set up an auction of the use of time in these spaces. The winning bidder is obligated to maintain the place and has the right to set an “entry fee"—say 50 per entry. The criterion for economic efficiency has been met. But would you pay?

Auctions have been used for millennia and are a simple way of price determination for multilateral trading. There are many types of auctions. In a unit auction, an item is sold to the bidder with the highest bid, and at a price equal to it. In a share auction, bidders receive fractional shares at a sales price that equates the demand and supply of the share. On another dimension, auctions are classified as either static—with a one-time price bid, often sealed—or dynamic—where multiple attempts can be made with full knowledge of competing bids. The Federal Communications Commission (FCC) in the US has been in the business of allocating radio spectrum for over 80 years now. For much of that time, comparative hearings, petitions and a lottery system have dominated spectrum allocation. The first modern spectrum allocation took place in New Zealand in 1990. It is now the most commonly used method around the world.

There are also many types of assets. Some assets such as spectrum are indivisible, intrinsically renewable and do not require much in the way of maintenance. Mining resources, in contrast, are discrete, depleting and non-renewable. Talented cricketers are individual and have expiry dates. What method is good for one type of resource may not be appropriate for another.

The auction solution to spectrum allocation has become accepted so quickly that it masks an underlying problem— spectrum is misconceived as a form of property that requires individualized allocation. The spectrum auction regime retains the government in its traditional role of providing centralized allocation. In practice, this discourages innovative competition, particularly from small players, and establishes an oligopoly of sorts. In addition, by requiring all spectrum users to buy access, the auction regime makes it more expensive for rural and poorer users from fully participating in new media such as wireless Internet. The winning oligopolists also have to contend with the winner’s curse, which is that the bidder who most “overestimates" the value of the asset is also the most likely to win. In India, we have an additional problem—the defence department “hogs" spectrum without disclosure and without real accountability to any other entity.

The speed of change of technology allows spectrum to be treated as an indivisible common good—the so-called “commons" made famous by Nobel laureate Elinor Ostrom. Technology now allows for multiple users to operate in the same area of the spectrum without interference. A commons in the spectrum could offer several benefits, including freedom to experiment with spectrum usage and a greater incentive to develop technologies for spectrum sharing. It may facilitate efficient transactions among competing users and render viable cross-boundary uses.

The critique against this approach is what Garrett Hardin called the “tragedy of the commons" in his eponymous 1968 article. Hardin wrote that a group of individuals, acting out of individual self-interest, would ultimately deplete a shared resource even if it were clearly not in their long-term collective best interest. He contended that there was no technological solution to the misuse of common property. There is an entire body of work that refutes this observation by distinguishing between “common property" and “open access". This view allows for a certain framework and regulation around common property, which maximize the long-term potential of the shared resource for a group.

Regulating the usage of spectrum as a commons may be premature. The requisite institutional and regulatory apparatus are not yet in place. In a 700 MHz spectrum allocation concluded by FCC in 2008, Google Inc. requested four open platforms as part of the licence conditions—open applications, devices, services and networks. While not fully a “commons" requirement, it is a suggestion of things to come. With rapid advances in technology and with other incidental failures of spectrum auctions, it is incumbent upon us to seriously investigate alternatives to “divide and auction" for spectrum allocation.

P.S.: Confucius said, “They must often change, who would be constant in happiness or wisdom."

Narayan Ramachandran is an investor and entrepreneur based in Bangalore. He writes on the interaction between society, government and markets. Comments are welcome at