Effectiveness of e-auctions in emerging economies

Our primary research goal is to examine whether governments in emerging economies can increase revenues by selling land through e-auctions (anonymous auctions conducted online over a period of time) rather than through traditional “physical place” auctions (auctions where buyers and sellers come together in physical space and time).

Though auctions are popular in the sale and lease of government assets (e.g. sale of spectrum, mining rights etc.), the lack of price and process transparency along with regulatory capture by private agents in emerging economies make them less effective in maximising revenues. Particularly, auctions that require disclosure of the identity of the bidders make it conducive for implicit collusion among bidders at the expense of the government (for e.g. the spectacular telecom spectrum scam in India). We seek to examine whether governments can mitigate some of this rent-seeking behaviour by conducting auctions in cyberspace where bidders’ identity can be more securely protected.

We use a natural experiment in the Indian land market where one of the larger State Governments used both e-auctions and physical auctions to sell its land to private developers. Though auctions were introduced to sell land only from the early 2000s, it is now common to see large city development authorities use auctions in order to actively increase land supply and finance infrastructure (Acharya, 1998).  Auctions help mitigate lack of price transparency and quantify the marginal value of various uses that the land could be put to use for the government.

Combining government auction data with private land transactions over an 11-year period from 2003-2013, we plan to provide estimates of premiums, both for e-auctions as well as for physical place auctions over private transactions, and to relate any differences to the degree of implicit collusion that is possible between bidders. We are particularly interested in answering the following questions:

1. Has the auction process improved price discovery by comparing prices obtained in government auctions with prices of transactions between private parties, controlling for land characteristics and time?
2. Do e-auctions generate higher revenues than physical place auctions controlling for land characteristics and time?
3. Do physical place auctions facilitate implicit collusion especially when auctions are held for multiple land parcels around the same time?
4. Do e-auctions exhibit lesser implicit collusion compared to physical place auctions?