ArticlePDF Available

Indian Telecom: Regulation, Spectrum Allocation and Dispute Management



Legal disputes have stymied the progress of the Indian telecom industry. This paper uses a contest theory model, where legal battles are a form of contest, to show that there can be equilibrium without legal battle. But this needs the policy makers to have dispute management as a prime area of focus. In that context the paper suggests suitable policy design against the backdrop of major telecom disputes.
IIMB Management Review, December 2009 287
Indian Telecom:
Regulation, Spectrum
Allocation and Dispute
Legal disputes have stymied the progress
of the Indian telecom industry. This paper
uses a contest theory model, where legal
battles are a form of contest, to show that
there can be equilibrium without legal
battle. But this needs the policy makers to
have dispute management as a prime area
of focus. In that context the paper suggests
suitable policy design against the backdrop
of major telecom disputes.
Subhasish M Chowdhury
University of East Anglia
Debabrata Datta
Institute of Management Technology, Ghaziabad
The privatisation of the telecom industry, the advent of cellular
telephony and the concomitant scarcity of spectrum have
burdened governments worldwide with the task of allocating
the scarce spectrum among incumbent firms and potential entrants.
These developments also throw open a challenge to economic
researchers—that of modelling a market where firms compete, subject
to availability of a scarce natural resource allocated by the government.
Given that spectrum is in short supply and acquiring a larger share of
the spectrum means more profit, it can be concluded that there will be
conflict over spectrum acquisition and firms will engage in rent seeking
activities. Unless the government (or the appointed regulating body)
manages this task of spectrum distribution and regulation efficiently,
disputes will surface and firms are likely to take recourse to legal redressal
procedures. What the nature of equilibrium will be in this kind of market
is an open question. In this paper we address this issue, using the Indian
telecom industry as a case study.
Before 1984, the telecom sector in India was operated exclusively by
the government. In 1985 the Department of Telecommunications (DoT)
was created as a separate department to serve as the primary policy-
maker and regulator for the telecom industry. The telecom sector was
opened to private firms in 1991 to bring in value added teleservices
other than fixed telephony. The Government of India released the National
288 Indian Telecom: Regulation, Spectrum Allocation and Dispute Management
Telecom Policy in 1994, and established the Telecom
Regulatory Authority of India (TRAI) in 1997. This period
of market driven reforms culminated in the incorporation
of Bharat Sanchar Nigam Limited (BSNL) in 2001 as a
public limited company. Controlled entirely by the
government, BSNL and Mahanagar Telephone Nigam
Limited (MTNL) remain at present the representatives of
public sector presence in the telecom industry in India.
Early Licensing Procedure
Initially in India, private entry was viewed as necessary
only for premium services (mobile telephony, paging etc)
as a supplement to public provision of basic telephony.
Therefore in 1992, the government invited competitive
sealed bids for two non-exclusive digital mobile licences
for a 10-year period, extendible by five years, for the
four metropolitan cities of Mumbai, Delhi, Kolkata and
Chennai. The licence specified the use of Global System
for Mobile Communications (GSM) standards for offering
cellular services. Once interim licences were awarded to
the private players, the unsuccessful bidders sought legal
intervention and after clearance from the Supreme Court
of India, eight licences were issued in 1994. DoT invited
tenders for two non-exclusive licences for each of the 20
circles (usually coterminous with states); selection from
among the technically qualified bidders was on the basis
of highest levy (later converted to licence fee) which was
measured over a 10-year licence period.
The issuing of cellular licences called for an initial allocation
of spectrum. In the metros each licensee was allocated
4.5 MHz in the 900 MHz spectrum, and in the circles the
allocation was 4.4 MHz each. The licence fee based cellular
service however could not take off as smoothly as had
been anticipated because operators reported huge losses.
The new telecom policy 1999 (NTP 99) provided for a
new policy framework for cellular mobile service
providers, and replaced licence fees with an arrangement
for revenue sharing. The initial licence period was
increased to 20 years, further extendible by ten years
NTP 99 for the first time explicitly recognised that the
number of cellular operators in a given geographical area
would necessarily be limited due to the scarcity of
spectrum. It announced that apart from the two private
operators already licensed, DoT/MTNL would be licensed
as the third operator in each service area. In order to
ensure a level playing field for the different service
providers in similar situations, the licence fee would be
payable by DoT also. However, as DoT was the national
service provider and had immense rural and social
obligations, the government would reimburse the full
licence fee to DoT.
NTP 99 also proposed that the spectrum utilisation should
be reviewed from time to time, keeping in view the
possibility of increased spectrum availability, the
requirements of the market, and other interests of the
general public. The entry of additional operators in a
service area was to be based on the recommendation of
TRAI. Cellular service providers would be required to
pay a one-time entry fee and also a licence fee on revenue
sharing basis, to be determined as per the
recommendations of TRAI. The growth of the Indian
cellular industry picked up in this new revenue sharing
regime. But this phase also witnessed the dispute between
GSM and Wireless Local Loop (WLL) operators (which
will be discussed later). This controversy was
subsequently resolved, but all along spectrum allocation
remained a bone of contention. Even TRAI in making its
recommendations on the entry of the fourth cellular
operator in June 2000 pointed out that it was constrained
in its ability to make recommendations because of the
lack of information about the availability of spectrum,
although it emphasised the critical role of spectrum
planning in sustaining competition and ensuring service
In short, the policy makers and the regulating body in
India as in any other country found it very difficult to set
up the regulatory mechanism and to choose the right mode
of spectrum allocation. As a result many loose ends
remained, paving the way for legal disputes.
Spectrum Planning and Dispute: A
Theoretical Overview
Mobile service operators are provided with the use of
radio spectrum. This spectrum is limited in supply in view
of its large number of uses—radio broadcasting, mobile
telephony, mobile satellite services, meteorological uses,
disaster management services, defence etc. The available
spectrum of frequencies ranges from 300 Hz (cycles per
second) to 300 GHz. However, given the present
technological standards, much of it is not available for
deployment. As such, radio spectrum is rationed out
internationally. Although the radio spectrum allotted to a
nation is national property and does not belong to the
IIMB Management Review, December 2009 289
government, it becomes the responsibility of the national
government to ensure that the scarce spectrum is optimally
The issue of spectrum allocation has occupied the centre
stage of policy debate in every country with the advent
and rapid expansion of mobile telephony. As the number
of subscribers keeps increasing, the demand for additional
spectrum is also rising. Further, the advent of new
technology like 3G also increases the need for additional
spectrum allocation. After the dust generated by the first
phase of spectrum allocation settled in Europe and North
America, there is now another controversy over the
optimum mode of spectrum allocation with regard to 3G
technology1. India is also facing the same problem now,
and is finding it difficult to come up
with a quick solution.
The allocation of any scarce
resource is always a problem.
Theoretically, there are five solutions
to this problem— muscle power,
luck, time, political power, and
market power. Of these methods,
the first one will never receive the
approval of a civilised society. The
second one involves the allocation
of the resource based on luck—by
flipping a coin, in a manner of
speaking. Its randomness makes it
undesirable. The third is based on
the principle of first come, first
serve, which in the context of
spectrum allocation gives undue
advantage to the first players in the
market or the existing players, and fosters monopoly, and
is therefore not an optimal solution. The fourth method
relies on the government to decide who best deserves the
scarce resource. In spectrum allocation literature this is
known as a ‘beauty contest’. The main argument against
a ‘beauty contest’ (in this context) is that it encourages
rent seeking and corruption. Moreover it is often arbitrary
and lacks transparency, and is prone to controversies.
This method can lead to litigation and eventual delay.
Market power is another much applied method of spectrum
allocation, and the instrument here is auction. The literature
on spectrum auction in the telecom industry is quite vast.
Vickrey2, and Wilson3 pioneered the attempts to develop
auction theory. Milgrom and Weber4, and McMillan5 have
elaborated on many ideas of spectrum auctions.
Economides6 gives an introduction to the
telecommunication regulation. Klemperer7, and Krishna8
are comprehensive discussions on strategic auction theory.
Many economists favour auctions over the ‘beauty
contest’ because it is efficient in theory—only an efficient
firm can be the highest bidder and eventual claimant of
the scarce resource9. Secondly, auctions generate revenue
for the government. Thirdly, they are transparent and
therefore subject to less controversy. However, experience
has shown that auctions can lead to problems especially
the winner’s curse10, where the winner bids an exorbitant
amount but the deal finally fails to pay off11. Another
problem is post contract opportunism where after winning
the spectrum, the winner may refuse to pay the committed
amount. No method is foolproof
and coming up with a good solution
to the problem of spectrum
allocation is no easy task for the
The main problem of allocating
spectrum on a discretionary ‘beauty
contest’ basis is in determining the
optimal spectrum requirement.
There is no empirical formula
which can calculate the exact figure
in this regard. Additional spectrum
requirements depend on various
parameters like the number of
subscribers, the density of
subscribers, and their geographic
morphology. Other determinants
are the types of applications (voice,
data, multimedia applications such
as video/audio, mobile TV etc) and the technology itself—
2G Code Division Multiple Access (CDMA IS95A, GSM),
3G Wide Band Code Division Multiple Access (WCDMA/
CDMA2000), WiFi, WiMax etc. In a multi-operator
scenario it is difficult to assess the value of these
parameters. Even when past data is available, a few
parameters such as current subscriber base, current
growth rate, etc might be known, while there would be
several indeterminate factors which could substantially
alter the spectrum requirement estimates. When major
technologies like GSM and CDMA are involved, the pattern
of spectrum use is distinctly different. Hence it becomes
doubly difficult to determine the efficient distribution of
spectrum between the two. The market makes a choice
between the two technologies on the basis of the cost of
The issue of spectrum allocation
has occupied the centre stage
of policy debate in every
country with the advent and
rapid expansion of mobile
telephony. As the number of
subscribers keeps increasing,
the demand for additional
spectrum is also rising. Further,
the advent of new technology
like 3G also increases the need
for additional spectrum
290 Indian Telecom: Regulation, Spectrum Allocation and Dispute Management
production and the quality of services provided. But in
this case both cost and quality are dependent on the
quantum of spectrum allocated, and therefore no
independent basis for cost comparison is available.
This scenario quite often leads to disputes, and the problem
is exacerbated by some additional factors. The
telecommunication sector is one of the fastest growing
sectors in the world—over 9 million users are added every
month—involving large sums of money and big players12.
The industry is in a state of flux, having witnessed the
transition from a state of natural monopoly to open
competition among the incumbent firms as well as the
new entrants. Liberalisation and privatisation have also
led to disputes, and both the government and the service
providers are finding it difficult to adjust to the regulatory
regime. In countries like India, where public sector firms
and private service providers coexist, the latter often allege
that the government’s decisions favour the former.
Additionally, the rapid changes in technology and business
practices render the process of decision making a
harrowing task for all those involved.
In such a scenario disputes can crop up frequently, leading
to constant and prolonged litigation, an atmosphere of
uncertainty, and a delay in decision making. Given the
short life of telecom technology, timely technological
upgradation gets affected. Thus litigation tends to become
a hindrance to the growth of this industry. The model in
the next section illustrates the scenario of a firm resorting
to legal action, which can result in a ‘bad’ equilibrium.
The legal battles among firms and their implications on
market equilibrium have received attention in the literature.
Robledo has discussed asymmetric litigation costs as entry
deterrence instruments13. Bruce and Macmillan14
summarise dispute resolution in the telecom sector. Our
paper attempts to introduce legal battles as a form of
contest, which is new to the literature.
Rent Seeking Contest Model
We consider a rent seeking contest model where two risk
neutral firms—say firm 1 and firm 2—are involved in a
Cournot type oligopoly game in the telecom industry.
They may engage in fighting between themselves over
spectrum allocation, and spend resources (xi) to seek
redressal if they feel aggrieved. So 0.
xIf firm 1 wins
the fight, its costs may or may not fall, but firm 2’s costs
will definitely go up. Thus the prize of the fight is some
expected relative cost advantage over the rival firm.
We posit a two-stage game between two players. In the
first stage the firms involved decide whether to engage in
a legal battle or not. In the second stage the firms produce
goods for the market, and play a quantity setting game.
The marginal cost of production of a firm depends on the
outcome of the legal battle, if it does happen.
For simplicity, we assume a linear demand function of
the form P(qi + qj) = a - b(qi + qj), where a (vertical
intercept of the demand curve), b (slope of the demand
curve) >0, and the effort spent on litigation by firm i is xi.
P is the price in the market that depends on quantity (q)
produced by firms i and j. The marginal cost of the
winning firm weakly decreases by
∈[0, 1]
and the cost of the rival firm rises by .15
We use the revised Tullock contest success function16
and represent the probability of success in the litigation
as pi = xi / (xi + xj + T), where i, j = 1, 2 and T is the effort
of the regulator to reduce the possibility of litigation, and
the chance of success in it. Specifically, T / (xi + xj + T)
is the probability that nobody wins in the legal combat.
T can be transparent policy, negotiation skills, general
acceptability of the regulator etc. c is the constant part of
marginal cost.
Superscript s represents success, and superscript f
represents failure.
The Pay off function is:
is the fixed price of the resource x.
We apply backward induction and first find out the
solution of the second stage Cournot game. Solving for
first order condition and assuming global concavity of
the pay-off function
Hence we get
IIMB Management Review, December 2009 291
the global concavity condition, we get the slope and
curvature of implicit best response function of firm i:
So the reaction curve of firm 1 is convex from below
when it rises upwards, and concave from below when it
slopes downwards. The reaction curve of firm 1 will turn
backward below the 450 line. Similarly the reaction curve
of firm 2 will turn backward above the 450 line. One of
the equilibria takes place at the intersection point and the
other one takes place at the origin. The two equilibria are
given in Exhibit 1.
Implications of the Model
The redeeming feature of this model is that the fight-fight
equilibrium is not the only equilibrium. There is also a
‘good’ equilibrium where there is no legal battle. Thus, at
The firms decide whether to engage in litigation or not
after taking the above profits into consideration.
Proposition: There are two equilibria in this litigation
game. In one, the equilibrium is xi
** = xj
** = 0 provided
T >
_ ki
1 /
, where k is defined below. In the other
equilibrium we get xi
* = xj
* > 0.
(i) Solving for equilibria
Calculating the profit from the Cournot game we get
It is observed that if
then .
If we get one Pure Strategy Nash
Equilibrium (PSNE) at
Again solving for first stage first order condition
for xi > 0, we get
similarly .
Solving both Best response Functions, we get the PSNE
with the following positive litigation effort.
In case of symmetric costs we get .
(ii) Sufficient conditions for global concavity of profit
i e
i e
As we look at the first order condition of the firms, under
Where R1(x2) represents reaction curve of firm 1 as a best response
to x2 and R2(x1) represents reaction curve of firm 2 as a best
response to x1; O and E are two equilibrium points.
Resources of firm 2
Resources of firm 1
Exhibit 1 Two Equilibria in a Rent Seeking
Contest Model
292 Indian Telecom: Regulation, Spectrum Allocation and Dispute Management
least theoretically, attempts can be made to reach the good
The implications of the model are clear. If in a country
like India, the government cannot manage the dispute in
the telecom industry—i e, it cannot maintain a high T
(regulatory effort)—the country may end up with the bad
equilibrium xi
** = xj
** > 0 where firms waste resources in
fighting. The good equilibrium is xi
** = xj
** = 0 where
there are no legal battles, but the government and the
regulator have to play an effective role to ensure this
equilibrium is maintained.
Spectrum Allocation in India: Case Studies
of Various Legal Disputes
We can examine spectrum allocation in India in the light
of the results of the above model. As was mentioned
earlier, the Indian telecom industry has been beset with
legal disputes and wrangling since the introduction of
market-driven reforms. We will discuss two major
Case 1: Transition from Fixed Fee to Revenue
Sharing Model
The original auction process for spectrum allocation
divided the country into 21 circles. DoT stipulated that
there could be only two operators per circle for cellular
services, and only one additional operator for fixed phone
services. The use of auctions appears to have been
motivated by the desire to raise revenues, and to ensure
transparency in the allotment of licences. As part of the
auction process, the bidders would first be evaluated on
the basis of their experience, and other criteria, and only
qualified bidders would be allowed to bid. The design of
the auction process was faulty as it not only retained the
licensing fee—determined by bids, and fixed for the first
three years—but also fixed call charges at higher levels
than fixed telephony rates (provided by DoT), leaving the
monthly rental element as the only adjustable instrument
for a bidding firm to attract customers and compete with
others17. This was brought about by DoT playing a
prominent role in this auction system both as a player and
as the referee. In the role of a player, DoT and MTNL got
cellular licences sans licensing fee in all circles. As the
referee, DoT set per-minute cellular charges at extra-high
levels in relation to its own fixed line charges. Moreover,
it required all interconnections to go through its fixed lines
and charged a high fee for this service. Private cellular
operators were also denied the right to provide the services
of international calls18.
Despite these restrictions, the first licensing attempt
through auctioning in 1995 evoked a good response. But
the bid pattern was lopsided. One firm with a turnover of
US$ 0.06 billion quoted the highest bid in nine circles,
incurring a payment liability of $15 billion. This suggests
that this firm and some other bidders were unduly
enthusiastic and susceptible to the winners’ curse, leading
to the ‘hold-up’ problem or a spot of post contract
opportunism. Just after the completion of licensing, the
firms started pleading with the government that with this
kind of bid, call charges would be too high for the industry
to break even. There were also some lawsuits involving
DoT. Ultimately in 1999, several powerful private firms
pressurised the government to change the system and
introduce revenue sharing, absolving the firms of their
auction commitment.
This debacle can be attributed to several factors. A
fundamental flaw in this auction was that the government
played the role of a player as well as the referee. This
made the entire scenario murky and led to squabbling.
There were too many institutions governing the situation,
who sometimes worked at cross purposes. For the
formulation of the telecom policy, in addition to DoT, the
Department of Space, and TRAI, the government had set
up ad hoc bodies, such as the Information Technology
Task Force and the Group on Telecommunications.
The growth scenario of cellular business was not properly
projected. The mobile service was assumed to be a luxury
whereas it turned out to be widely used by people. There
was a failure to project future demand, and gauge the
extent of future decline in the prices of handsets and
telephony services.
Moreover, the possibility of post contract opportunism
was not taken care of. TRAI had both regulatory
responsibilities as well as the task of adjudication, and its
role was not unambiguously specified. One of TRAI’s
rulings against the granting of a licence to MTNL by DoT
was quashed by the Delhi High Court on the grounds
that the government’s (i e, DoT’s) decision should get
precedence over the regulator’s decision.
Case 2: GSM-WLL Dispute
The single most important legal controversy occurred in
2000 when TRAI proposed that mobile calling using the
IIMB Management Review, December 2009 293
WLL licences should be restricted to relevant short distance
calling areas19. But the basic service operators of WLL
exploited a loophole in the licence and offered the full
range of mobile telephony services by using call forwarding
and multiple number registrations. The GSM based
operators who had paid a significant licence fee contested
this decision, and this legal battle continued for three years.
In January 2003, the cellular operators, based on GSM
system, threatened not to interconnect with WLL operators
and block all the calls unless the issue of access charge
was solved. As blocking of calls violated the cellular service
license, TRAI immediately issued an order on 9 January,
2003 against such blocking of calls originating from WLL
networks. Cellular operators moved the Telecom Disputes
Settlement and Appellate Tribunal
(TDSAT) and the Supreme Court,
seeking redressal against the TRAI
order. Finally, to create a level playing
field between WLL and cellular
operators, TRAI decided to impose
an access charge of 30 paise per
minute on all the calls in metro cities.
With this, the interconnect charges
paid by cellular operators in metro
circles came down from Rs 1.20
for 3 minutes to 30 paise per
The problem was caused by the
failure of the regulatory body to
properly implement the initial
decision to restrict WLL services to
calls within the local circle only.
TRAI had announced that mobile
switching centres (MSC) could not
be used in the handsets for WLL based services. Because
of this restriction, the licence fee for WLL service
providers (Rs 495 crores) was much less than the fee for
GSM cellular service providers (Rs 1633 crores)21. But
WLL firms started providing long distance phone services
(out of the local circle) by innovative use of technology.
TRAI did not take immediate action to prevent this, and
the dispute started snowballing. Meanwhile the major WLL
service providers in India at that time had built up a
substantial subscriber base. TRAI took an ambivalent
stand, and failed to initiate action against the operators.
Allegations surfaced (unsubstantiated) that undue political
intervention prevented TRAI from taking appropriate
action, resulting in favouritism and differential treatment
among the firms.
The problem was finally resolved when the government
introduced unified licensing, making cellular services
technology-neutral, and allowing WLL players to provide
the full range of mobile phone services after payment of
an entry fee equal to what the GSM operators had paid.
But by that time, the two major Indian WLL operators
had already established their base. There are several lessons
to be learnt from these cases. The regulatory body ought
not to make differential treatment on the basis of
technology; it is better to provide technology-neutral
licensing. It is up to the licensee to choose which
technology and equipment will be used to provide the
licensed service. This provides a fair and predictable
regulatory regime flexible enough to embrace the right
technology and market developments.
The merger of political interests and
the interests of particular firms is a
critical issue, and a very difficult
one to tackle. One means of
protection against such a nexus is
transparency, which interestingly is
often ensured by political
competition. The appointment of a
dispute settlement body consisting
of government and private
representatives—as in Japan—or
the use of the services of private
arbitrators—as in Hungary—are
some remedies.
Situation since 2003
An analysis of the second phase of
licensing reveals an improvement
in the Indian government’s ability
to orchestrate an effective bidding and licensing process.
This phase was characterised by much less legal dispute.
There are several reasons for such an improvement. The
introduction of the unified licensing system in 2003
ensured a technology-neutral policy, and eliminated the
distinction between basic and cellular services. The
regulatory and dispute settlement roles of the original TRAI
were bisected, and the TDSAT was constituted to address
dispute settlement in 200022. The scope for court
intervention was curtailed by making the Supreme Court
of India the sole judicial authority with the powers to review
the decisions of TDSAT. DoT’s telephone services were
transferred to a new corporate entity, BSNL. This made
BSNL at par with private firms in the eyes of the law.
In 2000, TRAI proposed that
mobile calling using WLL
licences should be restricted to
short distance calling areas. But
the operators exploited a
loophole and offered the full
range of mobile telephony
services. The GSM based
operators who had paid a
significant licence fee contested
this decision, and this legal
battle continued for three years.
294 Indian Telecom: Regulation, Spectrum Allocation and Dispute Management
In the matter of spectrum allocation, the Indian Spectrum
Management Committee (created in May 1999) has failed
to significantly improve the effectiveness of the auction
itself, resulting in a situation today in which services that
require the same spectrum in neighbouring bands are still
treated differently by the government. Hence the
government has been resorting more frequently to ad hoc
The failure of the initial auction mechanism and the
government’s inability to stick to its position created
trouble in the allocation of spectrum. The problem became
more complicated when CDMA technology came into
being along with GSM. Although this has encouraged
faster telecom development, and has led to an increase in
the number of players in each service area, the GSM-
CDMA row still continues.
In 2001 the government decided to bundle entry level
spectrum (of 4.4 Mhz) for $413 million. Subsequently
the government allocated extra spectrum on the basis of
number of subscribers—a practice not usually followed
elsewhere. This was done on the basis of the report of
the Technical Evaluation Committee (TEC) in October
2007. TEC recommended enhanced subscriber linked
criteria which was even higher than what was
recommended by the Authority23. Since the government
prefers to follow a discretionary policy, there is constant
lobbying by the GSM and CDMA players, and appeals to
court are made off and on. The discretionary policy has
proved to be particularly prone to bargaining and
squabbling on account of the technological differences
between GSM and CDMA with regard to the pattern of
spectrum use.
The GSM lobby argued that CDMA needs less spectrum
since the latter technology is five times more efficient
than GSM in the use of spectrum, provided CDMA
operators invest in proper infrastructure. GSM operators
had invested in setting up cell sites24, and they alleged that
CDMA operators were demanding more spectrum without
making similar investments in order to save investment
costs and to make more profit.
The CDMA lobby argued that if GSM were to evolve to
WCDMA, GSM operators would be able to achieve similar
spectral efficiency and revenue generating services as
CDMA operators. So they demanded a revision of the
policy that allots more spectrum to GSM operators
compared to CDMA operators. They also alleged that
GSM service providers hoarded spectrum and deprived
the nation of the use of precious resources.
The GSM association appealed to TDSAT and then to the
Delhi High Court to grant a stay order against the
government’s decision to allow CDMA operators to
provide GSM services. While this issue is still pending in
TDSAT, the Delhi High Court rejected this petition in
August 2008 citing technology neutrality25. So it is now
up to TDSAT to decide whether CDMA operators should
get an additional allocation of spectrum for GSM based
services. The Department of Telecommunications (DoT),
in July 2009 has sought TRAI’s recommendations on the
comments of the Committee on ‘Allocation of Access
(GSM/CDMA) Spectrum and Pricing’. In addition, TRAI
has also been requested to furnish its recommendations
on the terms and conditions of existing licence for
extending validity of these licences perpetually or otherwise
vis-à-vis 2G spectrum (GSM and/or CDMA) allocated
and/or 3G spectrum owned by existing licensees. DoT
has also sought the Authority’s clarification on auctioning
of all spectrum. On 16th June 2008, the government
constituted another committee consisting of
representatives of DoT, Technical Evaluation Committee
(TEC), defence and educational institutes like IIT, IIM
etc. The second committee submitted its recommendations
on 13th May 2009. Now TRAI has come out with a
consultation paper to settle the policy for spectrum
allocation26. But unfortunately the situation has been vitiated
by allegations against DoT of corruption in spectrum
What transpires from the above analysis is that although
the regulatory system in India has been relatively
streamlined, an effective policy of spectrum allocation is
yet to be devised. Hence in terms of our model, the scenario
tends towards the bad equilibrium. The weaknesses of
the present policy are many. After initial allocation through
auctions, subsequent allocations remain arbitrary and
discretionary. The methods of these allocations do not
match with either auction nor ‘beauty contest’, and keep
the milieu perpetually conducive to litigation. Commenting
on the situation, the Delhi High Court said, ‘prima facie
we find that spectrum has been allocated in the worst
manner and public exchequers have lost thousands of
crores [rupees]’27.
On the basis of the technological differences between
GSM and CDMA in the pattern of use of spectrum, a 2:1
ratio in spectrum allocation has been introduced. The
IIMB Management Review, December 2009 295
arbitrariness of this allocation pattern has made the situation
prone to disputes. The licence fee is charged at differential
rates for different services. There is no charge for pure
internet services, but there is a 6–10% charge for mobile
telephony, leaving scope for one company to allege that
the other is misclassifying its revenue sources.
In view of the shortcomings of the discretionary allocation
of spectrum, and its vulnerability to legal disputes, the
auction method should be adopted to allocate further
spectrum. A properly designed auction mechanism is more
transparent and less prone to legal issues. It augurs well
that DoT has decided to go for auctions for the allotment
of spectrum for 3G and WiMax in the country, and has
announced the guidelines. It is however also necessary
for the government to be aware of
the associated ‘hold up’ problem, and
to insist on some commitment
device, so that the eventual winner
of the auction cannot deny payments
later. It should also ensure that
proper infrastructure is rolled out. A
part of the auction revenue can be
used to subsidise infrastructural
As our model suggests, dispute
settlement should become an
important area of focus for the
seamless growth of an industry like
telecommunications. Regulatory
adjudication is currently the standard
practice of dispute resolution in this
sector. In order to improve this mechanism, policy makers
should particularly focus on the partitioning of judicial
functioning for the area of regulation, antitrust measures
and consumers’ disputes; the technological neutrality of
policy; transparency; promptitude in decision making; and
dispute resolution with the involvement of private bodies
in arbitration and negotiation.
The regulatory mechanism in India is often alleged to have
a politically-driven bias. Political influence cannot be fully
avoided, since the regulating body is appointed by the
government. So it would be helpful to know how different
countries in the world are trying to grapple with this knotty
but commonplace problem.
The participation of non-official bodies in the dispute
settlement mechanism can be a means of reducing political
influence. Such mechanisms include arbitration and
mediation by a team of specialised arbitrators. Many
countries like the USA, Australia, Jordan, Hungary etc
have a system of arbitration in place. When foreign firms
are involved, international arbitration can also be practised.
Indonesia has a system of involving the International
Chamber of Commerce for international mediation. These
mechanisms complement regulatory adjudication, while
maintaining the regulator's role as the prime decision-
maker. It is however important that there should be
official endorsement for the non-official dispute
resolution alternatives. Australia and Canada have a
system of industry-based consensus building
organisations, comprising industry
representatives. In some countries
dispute settlement is considered a
separate area of regulation. The US
Supreme Court opined that the task
of adjudicating regulatory disputes
and anti-trust disputes should be
bifurcated. Japan set up a special
Dispute Resolution Commission
with the powers of mediation and
Telecommunication in India is often
referred to as a success story. The
mobile set has become an essential
possession for many people. It is
heartening to note that the Indian
telecom sector has been attracting
huge investments, and the news of
the 3G auction has evoked keen
interest among big foreign firms.
The industry however needs a well-designed dispute
settlement machinery so that the growth of the Indian
telecom industry is not derailed in legal conflicts.
References and Notes
1 McKnight, L, R Linsenmayer, and W Lehr; ‘Best Effort versus
Spectrum Markets: Wideband and Wi-Fi versus 3G MVNOs?’
Working paper, 28.10.2001, MIT, DSpace@MIT.
2 Vickrey, W, 1961, ‘Counter Speculation, Auctions, and
Competitive Sealed Tenders’, Journal of Finance, Vol 16, No 3,
pp 8–37.
3 Wilson, R B, 1967, ‘Competitive Bidding with Asymmetric
Information’, Management Science, Vol 13, July, pp 816-
820;Wilson, R B, 1992, ‘Strategic Analysis of Auctions’, in R J
Aumann and S Hart, (Eds), Handbook of Game Theory with
In view of the shortcomings of
the discretionary allocation of
spectrum, and its vulnerability
to legal disputes, the auction
method should be adopted. A
properly designed auction
mechanism is more transparent
and less prone to legal issues. It
augurs well that DoT has
decided to go for auctions for
the allotment of spectrum for 3G
and WiMax in the country, and
has announced the guidelines.
296 Indian Telecom: Regulation, Spectrum Allocation and Dispute Management
Economic Applications, Vol 1, North Holland, pp 228–279.
4 Milgrom, P, and R J Weber, 1982, ‘A Theory of Auctions and
Competitive Bidding’, Econometrica, Vol 50, No 5, pp 1089–
5 McMillan, J, 1994, ‘Selling Spectrum Rights’, Journal of
Economic Perspectives, Vol 8, No 3, pp 145–162.
6 Economides, N, 2004, ‘Telecommunications Regulation: An
Introduction’, Working Paper 04-20, NET Institute.
7 Klemperer, P D, 1999, ‘Auction Theory: A Guide to Literature’,
Economic Surveys, Vol 13, No 3, pp 227–286.
8 Krishna, V, 2002, Auction Theory, San Diego: Academic Press.
9 Cave, M, and T Valetti, 2000, ‘Are Spectrum Auctions Ruining
our Grandchildren’s Future?’, Info, Journal of Policy Regulation
and Strategy for Telecommunication Information, Vol 2, No 4,
pp 347–350; Haan, M A, and L A Toolsema, 2003, ‘License
Auctions when Winning Bids are Financed through Debt’,
Working Paper No 200310, University of Groningen, CCSO
Centre for Economic Research; Virmani, A, 2000, ‘A
Communication Policy for the 21st Century’, Economic and
Political Weekly, Vol 35, No 23, pp 1907–1910; Virmani, A, 2004,
‘Competitive Access to Telecom: Spectrum Policy and M & A’,
Economic & Political Weekly, Vol 39, No 7, pp 620–622.
10 Capen, E C, R V Clapp, and W M Campbell , 1971, ‘Competitive
Bidding in High Risk Situations’, Journal of Petroleum Technology,
Vol 23, June, pp 641–653.
11 Zheng, C B, 2001, ‘High Bid, Broke Winners’, Journal of
Economic Theory, Vol 100, No 1, pp 129–171.
12 TRAI Annual Reports 2007–08.
13 Robledo, Julio R, 2005, ‘Strategic Patents and Asymmetric
Litigation Costs as Entry Deterrence Instruments’, Economics
Bulletin, Vol 15, No 2, pp 1–9.
14 Bruce R R, and R Macmillan, 2004, ‘Dispute Resolution in the
Telecommunications Sector: Current Practices and Future
Directions’, Discussion Paper, A study commissioned by
International Telecommunication Union and The World Bank.
15 d’Aspremont, C, and A Jacquemin, 1988, ‘Cooperative and Non
Cooperative R&D in Duopoly with Spillovers’, American
Economic Review, Vol 78, No 5, pp 1133–1137.
16 Tullock, G, 1980, ‘Efficient Rent Seeking’, in J M Buchanan, R
D Tollison and G Tullock, (Eds), Towards a Theory of Rent
Seeking Society, Texas A& M University Press, pp 97–112; Nti,
K O, 1997, ‘Comparative Statics of Contests and Rent Seeking
Games’, International Economic Review, Vol 38, No 1, pp 43–59.
17 Malik, P, 2004, ‘Indian Telecommunications Policy and
Regulation: Impact on Investment and Market Structure’,
Discussion Paper, WDR 0304.
18 Panagaria, A, 2008, India: The Emerging Giant, Oxford
University Press; Jain, R S, 2001, ‘Spectrum Auctions in India:
Lessons from Experience’, Telecommunications Policy, Vol 25,
No 10, pp 671–688.
23 Consulation Paper TRAI, 2009.
24 Cell sites (towers) allow the spectrum to be re-used in order to
carry more traffic.
25 Jain, S, ‘2G spectrum: What’s $10-15 bn between Friends?’,
September 29, 2008.
26 Consultation paper No 6/2009, Oct 2009, Telecom Regulatory
Authority of India.
28 Datta, D, 2006, ‘Spectrum Auction and Investment in Telecom
Industry – A Suggested Policy’, Paper Presented at the Seminar
on Contemporary Issues in Development Economics, Jadavpur
University, Kolkata.
29 For a review of international situation of dispute management in
telecom, see Bruce and Macmillan, ‘Dispute Resolution in the
Telecommunications Sector: Current Practices and Future
Reprint No 09401
Subhasish M Chowdhury is Lecturer, School of Economics,
University of East Anglia, Norwich, UK.
Debabrata Datta is Professor of Economics, Institute of
Management Technology, Ghaziabad.
Copyright of IIMB Management Review is the property of Indian Institute of Management Bangalore and its
content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for individual use.

Supplementary resource (1)

The telecom industry processes are vertically integrated. In the era of voice-dominated landlines, operators were operating on a simple mode of operations though the network system was complex. The industry was characterized by technology with increasing returns to scale and was dominated by monopoly market structure, with little or no competition, and selling of the services was not rigorous.
Globalization opens up possibilities for gains in efficiency through international exchange based on the principle of comparative advantage. These gains are very significantly augmented with the development of communications system that reduces cost of negotiations, monitoring, and coordination. The advent of telegraph as a communication device in 1839 in Britain marked a signal change in this scenario of cost of communication.
Full-text available
This paper asks is whether (i) 3rd generation wireless services, as embodied in the planned and soon to be offered services emerging first in Asia and Europe, or (ii) the unlicensed wireless services such as 802.11 or wi-fi but also including more advanced wideband and ultrawideband (UWB) services which are being experimented with primarily in North America, offer more compelling visions for advanced wireless services. we conclude that secondary spectrum markets are important for the viability of the 3G industry, and not only for reasons of efficiency. One large difference between 2G and 3G networks, observed in our models, was that voice services alone would not generate sufficient revenues for a 3G system. License holders which up to now have concentrated on selling a single product, will need to develop a much larger range of advanced applications, which will have to be marketed and packaged in different ways for different market segments.
If it is true, as common sense tells us, that a lease winner tends to be the bidder who most overestimates reserves potential, it follows that the "successful" bidders may not have been so successful after all. Studies of the industry's rate of return support that conclusion. By simulating the bidding game we can increase our understanding and thus decrease our chance for investment error. Introduction We would like to share with you our thoughts on the theory of competitive bidding. It is a tough business. We are not sure we understand as much as we ought to about the subject. As in most scientific endeavors, we think there is more knowledge to be gained by talking with others than by keeping quiet. Our first attempt at actually using a probability model approach to bidding was in 1962. We borrowed heavily from Lawrence Friedman's fine paper on the subject. But the further our studies went, the more problems we noticed for our particular application. problems we noticed for our particular application. We decided to strike out on our own. By 1965 we had our model just as it is today. But having a model and completely understanding its workings are not the same thing. We are still Teaming. While we refer to the "model" as though it were some inanimate object, it is not. What we want to describe to you is a system for taking the best judgments of people - properly mixed, of course, with historical evidence - and putting those judgments together in a rational way so they may be used to advantage. Lest the reader be too casual thinking that since he is not personally involved in lease sales he need not pay the closest attention, we offer this thought. There is a somewhat subtle interaction between competition and property evaluation, and this phenomenon - this culprit - works quietly within and phenomenon - this culprit - works quietly within and without the specific lease sale environment. We would venture that many times when one purchases property it is because someone else has already looked at it and said, "Nix." The sober man must consider, "Was he right? Or am I right?" The method of analysis we will describe is strictly for sealed bid competitive lease sales, but the phenomenon we will be talking about pervades all competitive situations. Industry's Record in Competitive Bidding In recent years, several major companies have taken a rather careful look at their records and those of the industry in areas where sealed competitive bidding is the method of acquiring leases. The most notable of these areas, and perhaps the most interesting, is the Gulf of Mexico. Most analysts turn up with the rather shocking result that, while there seems to be a lot of oil and gas in the region, the industry probably is not making as much return on its investment there as it intended. In fact, if one ignores the era before 1950, when land was a good deal cheaper, he finds that the Gulf has paid off at something less than the local credit union. Why? Have we been poor estimators of hydrocarbon potential? Have our original cost estimates been too potential? Have our original cost estimates been too conservative? Have we not predicted allowables well? Was our timing off? Or have we just been unlucky? JPT P. 641
We briefly review the rationale behind technological alliances and provide a snapshot of their role in global competition, especially insofar as it is based around intellectual capital. They nicely illustrate the increased importance of horizontal agreements and thus establish the relevance of the topic. We move on to discuss the organisation of industries in a dynamic context and draw out consequences for competition policy. We conclude with an outlook on the underlying tensions between technology alliances, competition policy, and industrial policy.
Spectrum auctions have been used with significant success in many developed countries. From a regulatory and policy perspective, spectrum auctions ensure the efficient use of spectrum by allocating it to those entities that value it most, while also generating revenues for governments. But auctions may lead to unexpected outcomes due to unanticipated problems with their design leading to unexpected bidder behavior such as collusion and over-bidding. The key challenge before regulatory agencies is to design auctions in such a way as to meet the objective of fostering competition while at the same time ensuring that bidders can effectively use the spectrum for their business. While India was one of the early adopters of spectrum auctions, its success in service provision has been low. This paper critically examines issues in auction design that contributed to this delay and reviews the key elements in the design process namely a coherent regulatory framework, choice of service areas, flexibility for service area consolidation, standards and their role, convergence, managing public service regulation and managing defaults. The paper compares the handling of these elements in auctions in the United States (US), United Kingdom (UK) and Australia with the objective of drawing lessons for Indian policy makers.
This paper analyzes the problem of competitive bidding (via sealed tenders) under uncertainty when one of the parties knows the value of the prize with certainty. A real problem of this type has been reported in a case study by Woods. The equilibrium strategies are derived and characterized for computational purposes in terms of the solution to a related differential equation. An example is solved in detail and found to be rather surprising in terms of the relative strength it imputes to the position of the party with better information.
This paper analyzes auctions where budget-constrained bidders have options to declare bankruptcy. It predicts a bidding equilibrium that changes discontinuously in a borrowing rate available to bidders. When the borrowing rate is above a threshold, high-budget bidders win, and the likelihood of bankruptcy is low. When the borrowing rate is below the threshold, the winner is the most budget-constrained bidder and is most likely to declare bankruptcy. This result explains the “high bids and broke winners” anomaly in the C-Block FCC spectrum auction. Based on its equilibrium analysis, the paper proves that a seller can profit from offering to finance the highest bidder at a below-market interest rate, even with default risk. Journal of Economic Literature Classification Numbers: D44, D45, D82, G33, L96.