New research on companies’ bidding strategies in the UK’s 2000 auction of third generation
(3G) mobile spectrum licences suggests that motives other than a comparison of the
benefits from holding a licence and the costs of winning a licence must have been behind
some of the bids.
The speculation in the study by Professors Tilman Börgers and Christian Dustmann,
published in the July 2005 Economic Journal, is confirmed in a comment by Dan Maldoom,
an economics consultant who advised British Telecom (BT) during the auction. He explains
that BT bids were deliberately intended to raise the price that some of their competitors had
to pay.
Börgers and Dustmann focus not on the total size of the bids, but on how companies
switched their bids from large licences (which gave access to 15 MHz of paired spectrum)
to small licences (which gave access to 10 MHz of paired spectrum). They find that several
companies switched between large and small licences in a way that cannot be rationalised
by a fixed estimate of the extra value of more spectrum.
Among the winners of the auction, the two companies with strange bidding strategies were
BT and the Canadian company TIW, which was later sold to Hutchison Whampoa. The
researchers find that BT seemed to revise its estimate of the extra value of more spectrum
from £300 million to £1.8 billion, while TIW seemed to revise their estimate of the extra
value of more spectrum from around £176 million to £745 million.
The researchers’ speculation as to why these valuations seemed to change is confirmed in
a comment by Maldoom. He explains that BT’s strategy in early rounds of the auction was
to conceal their true valuation of licences so that other, less experienced bidders could not
infer anything from BT’s bids.
Later bids for a large licence were aimed at raising the price that Vodafone had to pay. BT
had become convinced after the initial rounds that Vodafone was willing to pay a very high
premium for a large licence. By placing bids on a large licence that were then topped by
Vodafone, BT could raise the price that their competitors had to pay.
The strategies that these papers describe are not consistent with the theory of bidding
behaviour that has led game theorists to the prediction that auctions result in efficient
allocations. This suggests that this theory might be too restrictive in its assumptions.
None of the authors suggests that the UK’s auction actually resulted in an inefficient
allocation. Moreover, the bidding strategies described have most likely raised the revenue
that resulted from the auction.
But the authors argue that the described bidding behaviour has a potential to cause
inefficiencies, and suggest that this needs to be monitored carefully in future auctions.
As the UK government remains committed to auctions as a method for allocating spectrum
licences, this research has potential implications for future spectrum management.
In particular, designs of future spectrum auctions may need to be more careful in the
assumptions they make about bidders’ motives. Spectrum auctions have now become part
of the dynamic by which telecoms companies compete over the long run and need to be
considered in that context.
ENDS
Notes for editors: ‘Strange Bids: Bidding Behaviour in the United Kingdom’s Third
Generation Spectrum Auction’ by Tilman Börgers and Christian Dustmann with a comment
by Dan Maldoom are published in the July 2005 issue of the Economic Journal.
Tilman Börgers is Samuel Zell Professor of the Economics of Risk at the University of
Michigan.
Christian Dustmann is Professor of Economics at University College London.
Dan Maldoom is at DotEcon Ltd in London.
For further information: contact Christian Dustmann on 020-7679-5832 (email:
c.dustmann@ucl.ac.uk); Tilman Börgers on + 1-734-764 8022 (tborgers@umich.edu); Dan
Maldoom on 020-7467-2070 (email: dan.maldoom@dotecon.com); or RES Media
Consultant Romesh Vaitilingam on 0117-983-9770 or 07768-661095 (email:
romesh@compuserve.com).