Warning: Creating default object from empty value in /nfs/c03/h07/mnt/56080/domains/broadmind.nationalinterest.in/html/wp-content/themes/canvas/functions/admin-hooks.php on line 160


Vivek H. Dehejia

There is much confusion in the press and the social media about the economics behind the putative loss of the equivalent of US $40 billion to the central government’s exchequer as a result of the way in which 2G spectrum was allocated “first come, first serve” to cronies rather than being auctioned publicly in a transparent manner.

Some people are contending that the loss is not real, only notional. One such, not surprisingly, is the new Telecom Minister, Kapil Sabil, who uses the fancy legal-ese word “presumptive” for the loss. A presumptive loss, presumably, is not real, and therefore shouldn’t worry us. This is bunkum. Economists have a term called “opportunity cost” – it is the cost you incur by making one choice rather than another. It may not show up as an accounting cost, but it is a cost nonetheless. In this case, the $40 billion lost by undercharging telecom providers for access to the 2G spectrum is money that the government didn’t earn, and therefore couldn’t spend on something useful – like infrastructure, public works, social policy, military hardware, whatever. To quote Rupa Subramanya, writing recently in the Wall Street Journal India: “After all, a rupee not earned is a rupee wasted.”

Sonali Ranade, a trader, who is very active on Twitter, makes an interesting argument in favour of the view that selling the spectrum cheaply had economic benefits for the average person. She suggests that giving away spectrum at below the maximal possible cost can be explained as a policy choice, since, presumably, the end result is a lower cost per call to the consumer. In fact, the discredited former Telecom Minister, A. Raja, made a similar argument in his defence. The difficulty with this is that it appears as an ex post facto rationalization of what was evidently done in a shady and opaque manner, not a high-minded policy decision to help the public get cheaper mobile calls. So much for Mr. Raja.

However, Ms. Ranade’s argument needs to be taken seriously. If I understand correctly, her explanation is that the licensing costs have to be amortized by each provider, and, therefore, if the spectrum had been auctioned and the full value extracted by the government, these higher costs would be passed on to consumers as higher costs per call.

This explanation is plausible, but requires some background assumptions on market structure to be considered economically valid. In an oligopolistic industry, in which monopoly rents are shared amongst a small number of sellers, I can see this argument as having some validity. However, as the number of providers increases, and the industry’s structure becomes nearly competitive, the cost per call, which is a marginal cost, should be unrelated to the licensing cost, which is a fixed cost. This is first-year principles of economics, right out of the textbook. Indeed, in the United States, which has a highly competitive industry, the cost per call is nearly zero for many domestic calls.

The right response, in the Indian scenario, would be to open up mobile telephony to much more robust competition, and thereby destroy the value of the monopoly rents being earned by incumbent providers. Costs per call, which now average 25 or 50 paise, should drop down to 5 paise or less in my judgement.

My bottom line: just because a loss to the exchequer doesn’t show up in the account books, that doesn’t make it a fictive or “presumptive” or notional loss. It is as real as the cost of sleeping late, arriving late to a job interview, and so not getting a raise. Your income hasn’t changed. Does that mean there is only a notional loss? No. The loss is the income you didn’t earn but could have if you had shown up on time.

We need to tone down the political rhetoric and hyperventilation in the social media and inject some basic economics into the debate if we are going to dissipate the fog of obfuscation and chicanery surrounding the 2G mess.


After the initial post of this piece, I had an enlightening Twitter exchange with Sonali Ranade, mentioned above. It adds some important nuance to the broad strokes with which I have painted the picture. In her view, the “first come, first serve” policy, which basically gave the spectrum away for free, was the correct policy framework back in 2001 under then-Telecom Minister, Arun Shourie. The argument being, for a nascent industry, this was a way to jump-start it, and that charging the full value of the putative monopoly rents via auction would choke it off. Nominally, in 2008, Mr. Raja just continued the policy of the previous government. But, Ms. Ranade contends, by this juncture the climate had totally changed, and there was no economic rationale to give away the spectrum rather than auctioning it. Through the lens of my analysis, my interpretation would be that the industry was now rather more mature and “contestable” (if not competitive), sufficiently dense in a word, that it could have more nearly approached the competitive rather than the oligopolistic situation, and that an auction would be warranted, without fear of jacking up prices for consumers.

You may follow Vivek H. Dehejia on Twitter @vdehejia.

11 Responses to Rajanomics

  1. padmashta Roy December 28, 2010 at 11:04 am #

    Sir, I find your entire on presumptive loss of revenue very persuasive. Kapil Sibal has used all his lifetime lawyer skills to frame a word which puts the entire loss to exchequer in grey area of accounting.
    I though do not understand the economics part of the loss but the eagerness of various contenders in the DMK to corner this ministry , shows that there was incredible dough to be made.
    Another reason to keep License fee low is to facilitate the kick backs when the real value is realized from serious players.
    But alas i do not see much happening on the scam whether booking the culprits of realization of money lost. I can for see Sibal stretching all his brain muscles to find escape hatches , ways to subvert the investigation processes and putting the entire scam in never ending litigation mode.
    P Roy (@ThoraleSunil)

  2. Amresh December 28, 2010 at 11:25 am #

    what about the companies who ultimately sold equity in their company (Unitech, swan etc.) at much higher valuations, based only on the cheap licences. will this not mean that the entry cost for players remains high and wud ultimately be passed on the consumer?

  3. Prashanth December 28, 2010 at 3:45 pm #

    I think you have rightly pointed out the fact that its indeed a notional loss and not something that can be wished away. The Govt recently sold in a IPO shares of PSU’s. While one of the objectives was to (by way of providing a 5% discount to the retail investor) encourage investments by small investors and traders, the pricing was kept with the view to get the best out of such disinvestment.

    Its no one’s case that telephony (which is also a kind of infrastructure) has to be encouraged, but not by selling for free what is valuable. Of course, as has been proved by subsequent events, even this has not been free but a sort of Quid pro but one outside the purview of the govt.

    I think the entire spectrum scam consists of 2 parts –
    1. The License to be a Cellular company
    2. The Spectrum without which its as good as having a vacant site with no material to build.

    I believe that the License could have been given in First come First basis without guaranteeing any spectrum.

    Next, the Govt should have auctioned spectrum and those companies that bid for the same will be the one’s that actually intended to start and their bids would be reflective of their business strategy since it does not make sense to buy a raw material at a extremely expensive price when the consumer is already been accustomed to low priced services.

    By making spectrum expensive, the Govt will achieve 2 objectives,

    1. Get tons of money :)

    2. Ensure that spectrum to the extent needed is bought rather than getting large spectrum for hoarding and ensuing that others don’t get it (currently all existing players have more spectrum than necessary).

  4. Jiten December 28, 2010 at 6:01 pm #

    Completely disagree with your marginal costing analysis as a financial decision making criteria. What you call as a monopoly rent, has to be recovered too. Just because it was incurred one time does it mean, all companies ignore their investment and not recover. Remember these licence fees are ammortized for a period of 15-20 years and yet we are mo were near completion of tis licence period. Post this period your argument may still not make sense as these companies will again have to pay these ‘ monopolyrents’.

  5. vdehejia December 29, 2010 at 3:55 am #

    Thanks all for your comments and kind words.


    I think you have misunderstood the point. Under oligopoly conditions, as I note, the monopoly rent does have to be recovered or amortized over time, so there is a connection to call cost.

    But as market conditions become nearly competitive, or contestable, the monopoly rent itself goes to zero. There is nothing left to amortize. So the call cost can drop almost to zero. See the case of the US if you don’t believe me. But we are far away from this in India as we still have insufficient density. Hope this clarifies.

  6. Dark Lord December 30, 2010 at 10:52 am #

    The way I understand it, u want the telecom sector is to be an oligopoly in the initial years wherein the government will charge a premium for entering the sector.

    At a later stage, you want the sector to be nearly in the state of perfect competition, wherein call prices are determined by marginal cost considerations.

    But frankly, no one will pay a significant premium to enter into a field which will be driven by perfect competition (in 5 -10 years?). Telecos are paying a premium because of an underlying assumption that supply being scarce (hence end result cannot be perfect competition) will automatically result in oligopoly conditions. In fact, government will be able to charge a premium in a near perfect-competition market if at a later stage, the market will turn into an oligopoly or monopoly.

  7. why January 1, 2011 at 6:28 pm #

    “Indeed, in the United States, which has a highly competitive industry, the cost per call is nearly zero for many domestic calls.” US has two major mobile operators AT&T and Verizon. T-mobile has less than half their subscriber base and all the rest have less than 1% of AT&T’s and Verizon’s subscriber base. http://en.wikipedia.org/wiki/List_of_United_States_wireless_communications_service_providers

    The more I read your blog the more I feel that you fall for all the disinformation dished out for you proles by the US. Engrave the following on your brains. US does not have a free market. It is only a euphemism for monopolies and oligarchy. The sad thing is Indians are blindly emulating it in the name of liberalization and privatization and creating oligarchies.

    We need reforms. But reforms don’t necessarily have to be privatization and liberalization. Monopolies and oligarchies would be a greater threat.

    • vdehejia January 5, 2011 at 10:57 am #


      Your interesting argument is blunted by your intellectual arrogance, or insecurity. Is it necessary to label someone with whom you disagree a “prole”? Does that make you a patrician? Learn some manners before you make comments on these blogs – else you reveal your own ill-breeding and bad taste.

  8. @india_review June 30, 2011 at 4:49 pm #

    You are lost Idiot trying to sound intelligent

  9. abhinav kumar January 24, 2012 at 1:12 pm #

    this comment is particularly related to the assumption put by sonali ranade that selling spectrum at low cost at last really benefit the customer. I am posting this comment very late still i want that author should take a view.

    (1) we have to really study that what is the actual cost of a sms or call really cost to the service provider and at what price it is sold.

    (2) Selling a spectrum at low cost so that it will incur a low cost to customer also is in reality very cunning statement from Raja. Any company who is buying these spectrum are not going to get there return in the one or two years. It is a very long term investment and the basic reason for price reduction is
    (1) competition
    (2) Government initiative.

    what is the business model of these companies and how they are making money. If we will carefully study that there is no connection between the prices of spectrum at which government sold it.
    consider an sms costs 1 paisa to a customer(in reality it is not) then 1 paisa sms is even sold at 50 times price it costed to a service provider. the costing and pricing strategy will tell us exactly why A Raja decisions where a scam.

    If the government is not earning money by selling such a big opportunity to these companies then it is really a fraud which has been committed in this country.


  1. Tweets that mention My new blogpost "Rajanomics" at INI on the economics behind the 2G scam and why the loss isn't just "presumptive": -- Topsy.com - December 28, 2010

    […] This post was mentioned on Twitter by Indian Moron, Vivek Dehejia, Vivek Dehejia and others. Vivek Dehejia said: @sonaliranade Please check this out I cite you, I hope correctly, on reason why selling spectrum cheaply may be good: http://is.gd/jCUjB […]

Leave a Reply