Monday, January 25, 2010

3G Auction Design - An Economic Perspective

Economics is built on the basic premises of scarcity, abundance and rational human behavior. As the author of the book, undercover economist puts it, businesses or companies gain bargaining strength through scarcity which they want to exploit by creating pricing strategies and making customers pay more. Businesses would have some scarcity power in their wallet which could let them achieve some sort of monopolistic position within the kind of sector they are or the kind of products or services they are offering to the customers. We take telecommunication sector. We know that radio spectrum is a scarce resource and where there is scarcity there is money to be made. We will have a look at how the rights to use (telecom licenses) this scarce resource are put up for sale by the governments across the world and what are the most efficient methods adopted.

Oligopolistic Market
We see that telecom market is an oligopolistic market, which unlike more competitive markets in which firms are of much smaller size and one firm’s behavior has little or no effect on its competitors, an oligopolist that decides to lower its prices, change its output, expand into a new market, offer new services, or adverstise, will have powerful and consequential effects on the profitability of its competitors. For this reason, firms in oligopolistic markets are always considering the behavior of their competitors when making their own economic decisions.

To understand the behavior of non-collusive oligopolists (non-collusive meaning a few firms that do NOT cooperate on output and price), economists have employed a mathematical tool called Game Theory.
Two great game theorists, John Von Neumann and John Nash (subject of A beautiful mind) both were brilliant mathematicians. A game, according to the author, is an activity in which your prediction of what another person will do affects what you decide to do. Economists try and fit the game theory rules to the real life economics problems such as allocation of radio spectrum.

Spectrum allocations
Governments adopt two most popular methods for spectrum allocation which are beauty contests and auctions. In beauty contests, participants are awarded licences based on certain financial as well as qualitative parameters. The problem here is that the evaluation would be based on subjective criteria and the promise of future performance. Additionally, this is not a transparent process. Auction methods are superior to non-auction methods, as they help in the allocation of spectrum to those who value it the most, give maximum revenue to the government and ensure transparency. However, according to economist peter cramton, the goal for the government should be efficiency, not revenue maximization.

In spectrum auctions, the problems is, each bidder has its own forecasts , expansion plans, technology evolutions and its knowledge of the other bidders plans and insights is limited by the information communicated by other bidders through their bids. i.e .every body knows what is in their wallet but no body knows what is in others wallet.

Auctions could be open bid or sealed bid. In open bidding, the bidding process reveals information about valuations. This information promotes the efficient assignment of licenses, since bidders can condition their bids on more information. Bidders are able to bid more aggressively in an open auction, since they have better information about the item's value. The advantage of a sealed-bid design is that it is less susceptible to collusion. Open bidding allows bidders to signal through their bids and establish tacit agreements. With open bidding, these tacit agreements can be enforced, since a bidder can immediately punish another that has deviated from the collusive agreement. Signaling and punishments are not possible with a single sealed bid.

The US Auction
In the second half of 1990, FCC in US auctioned broadband PCS licenses for bands D, E and F to 153 bidders. Six of the 153 bidders in the DEF auction regularly signaled using code bids or retaliating bids. These bidders won 476 of the 1,479 licenses for sale in the auction, or about 40% of the available spectrum in terms of population covered. Controlling for market characteristics, these signaling bidders paid significantly less for their licenses. This problem occurred because the bids were published without rounding them to nearest few thousand dollars. This provided the bidders an opportunity to signal the other bidders which licenses they would prefer by making bids which contained area codes. This was the basic flaw in the design of the auction which did not take care of this aspect.
Though single-bid auctions are the best form to avoid collusion, the bidding process does not allow information exchange and efficient price discovery. In multi-stage auctions, each block of spectrum in a circle is allocated in separate auctions. This again restricts information availability and hampers price discovery. In a simultaneous ascending auction (SAA), all spectrum blocks are auctioned simultaneously in a single auction. The auction is conducted in multiple rounds. This ensures efficient price discovery. Additionally, the bidder is selected through transparent market mechanisms only. It also allows for superior aggregation when licenses of more than one circle are auctioned simultaneously. It has proved successful in the US and European countries. In India, “4th Telecom Operator” licences in various circles as well as FM Radio licenses were awarded through SAA.

The UK Auction
The UK auction was one of the most talked about auctions till date in the 3G auction history across the globe. The uniqueness of the auctions comes from the fact that the auction designers and Government regulator Ofcom had ensured that bidding needs to gather as many serious bidders as possible and ensure they turn up. By March 2000, thirteen registered bidders were all set and paid 50 million pounds deposits. The bidding process has been explained here. .

Now, around the time when these licenses were to be auctioned, four mobile-phone companies operated in Britain using “second-generation” (2G) technology. The incumbents were Cellnet, One-2- One, Orange, and Vodafone. (British Telecom(BT), the erstwhile state-owned monopolist privatized under Mrs. Thatcher, held a 60% stake in Cellnet which it increased to 100% in 1999.). Earlier, 2G licenses were awarded by UK government using the ‘beauty contests’.

The 3G auction was designed on Anglo-Dutch style. It had the best of both worlds i.e. an advantages of both sealed bid auction (which gives entrants a better chance of winning against strong incumbents such as mentioned above) and simultaneously ascending auction (SAA where the price starts low and competing bidders raise the price until no-one is prepared to bid any higher, at which point the final bidder then wins the prize at the final price he bids).
Earlier engineering considerations provided that only four liceses were available (called them B, C D and E bands). But later engineering advice changed and we were informed that it would be possible to make five licenses available instead of four. An additional band ‘A’ was introduced. Following was the distribution:
Licence A the largest = 2x15MHz of paired spectrum plus 5MHz of unpaired spectrum.
Licence B a little smaller = 2x15MHz of paired spectrum, but no unpaired spectrum.
Licences C, D and E are all roughly the same = each comprising 2x10MHz of paired plus 5MHz of unpaired spectrum.

The trick was to restrict the incumbents to licenses B, C, D and E and license A was reserved for a new entrant. This would ensure that one of the 2 large licenses (A or B) would go to the new entrant. The rules of the auction process were following:
• Auction design involved multiple rounds of simultaneous bids.
• In the first round, each bidder makes a bid on one license of its own choice. To remain in the auction, a bidder must be “active” in every subsequent round.
• An active bidder either currently holds he top bid on a particular licence, or else raises the bid on a licence of the bidder’s choice by at least the minimum bid increment.
• A bidder who is inactive in any round is eliminated from the rest of the auction. A bidder who currently holds the top bid on a license cannot raise or withdraw its bid, nor bid on another license in the current round.
• At the end of every round all bidders’ bids are revealed, the current top bidder for each license is determined, and minimum bid increments are set for the next round.
• The auction concludes when only five bidders remain. They are each then allocated the license on which
they are the current top bidder at the price they have currently bid for that license.

The initial expectations were that the auction might generate revenue of 2-3 billion pounds. But eventually surpassed all the expectations and generated a whopping 22.5 billion pounds and in the process had become the biggest auction ever in the history. Below table shows the 3G licenses issued in European nation, the awarding method and the money rose in the process.


Monday, January 18, 2010

Shift in the profitability zone

India is the second leading market when it comes to mobile phone usage globally. It is ironic when we take into account the fact that in the first 20 months after the introduction of cellular telephony, Indian telcos could not add 200,000 customers. In those days the average rate of cellular airtime was about Rs 16.40 per minute and the mobile to landline rate double that at Rs 32.80. Today, India is adding thousands of customers every day.

The rules of the telecom market are changing. Earlier the greater market share of the Telecom operator meant more customers which translates to greater profitability as there was only few operators and the incumbents like BSNL and MTNL had all the advantage. Also, as focus slowly shifted from wire line to wireless, incumbents could not adjust themselves quickly enough with the market. This gave the private players a leeway and a head start to launch their services in wireless space and slowly build the leadership position while catching up in the wire line space.

If we analyze telecom market, it’s an oligopoly. According to Charlie Munger, who was Warren Buffett's partner at Berkshire Hathaway, there are two kinds of Oligopoly. The constructive Oligopoly which allows all the companies to enjoy their respective share of the market without hurting each others margins. The destructive Oligopoly is the one which we can see now prevailing in the Indian Telecommunication market. In such a market the companies fight against each other for a larger market share but the ultimate winner is the consumer.

Not long ago, the only differentiating factor in this commodity kind of business was the call rates where it was difficult to retain subscribers and cost becomes the major deciding factor. The early mover advantage played a significant advantage but after the start of destructive oligopoly it is difficult to have sustained high return on Investment. Now, with more and more operators coming in (the number has now reached 11 per circle), the market has reached to a stage of hyper competition where players are getting into price wars and voice is rapidly getting commoditized.

We would analyze two key telecom KPIs: Minutes of Usage (MoU) and Average Revenue per minute (ARPM). MOU refers to minutes of usage or talk-time of subscribers. “Minutes of Usage” multiplied by Average Revenue per Minute (ARPM) to arrive at Average Revenue per User (ARPU).
So equations are simple:
No. of calls * call length = MoU
ARPU * Market share * Market = Total revenue

ARPM has shown a declining trend with respect to time but remained resilient through increase in MoU by launching creative campaigns and pricing strategies. But now, MoUs are no longer showing sensitivity to tariffs despite sharp cuts in tariffs and revenue per minute too is falling sharply.

Source: IDFC SSKI; KPIs – Key performance indicators; MOU – minutes of usage


  Source: Businessworld, TRAI

In an analyst’s language we call this a declining MoU elasticity. So we can see the trend:
ARPM Declining, MoU inelastic to price and stabilizing and hence ARPU declining.

In the past, price cuts typically resulted in more traffic on the networks as users increased talk time and hence increased MoU. So despite lower tariffs, telecom companies managed to protect profitability. But now, even that may become difficult as many telecom companies are struggling with spectrum-constrained networks. They may be close to exhausting their limits of taking on additional traffic. As recent TRAI reported that penetration in the urban areas already cross 100% mark, we see a clear trend of telcos to go rural going forward. Also, we can see other factors playing out such as bandwidth limitation in the urban areas leading to crowding up the spectrum by adding too many subscribers/Mhz which would create problems in QoS. So there is also a limit to that. However, China is already using twice of that capacity.
The only way to survive in this market is to identify the profit elements in this business and differentiate in the service offerings. Price is no longer a differentiator as all operators would match up with each other’s pricing strategies sooner or later. The real differentiators in this case would be:
-          QoS (Quality of service)
-          Greater emphasis on CRM and customer life-cycle management
-          Greater focus on alternative revenue models

Friday, January 15, 2010

Mobile Number Portability - To be or not to be..

Will MNP see light of the day in Indian Telecom Scenario? The question still remains debatable as we see another round of dates change from government on implementation of MNP.

As per the original plan set out by DoT, MNP was to be implemented in two phases. Mobile users in metros and category A circles like Tamil Nadu and Maharashtra were slated to avail this provision from January 1, 2010, while the rest of the country would have had access to MNP from April 1, next year. But the government has now decided to implement it in whole of the country in one go, by April 31, 2010.

If we look into the history of MNP across the globe, lot of challenges have been faced by countries like UK, Germany, Hongkong because of inherent complications in its implementation. If we analyze, we see challenges from the cost perspective as:
- Cost involved in upgrading the network infrastructure to support number portability i.e. the set up cost
- Cost involved in maintaining the upgraded infrastructure
- Cost involved in the usage of network resources to route the calls to the ported number

The setup cost can include:
- Central Number portability Database (NPDB) setup cost
- Software development or upgrade ege. Software upgrade in MSC
- Network upgrade
- Customer Relationship Management (CRM) upgrade
- Cost involved in upgrading the billing system

The maintenance cost may include:
- Cost for the agreed procedure involved in the porting process
- Activating the ported number
- Provisioning the routing information
- Informing all the service providers of the ported number

Then we see technical issues that need to be considered are the requirements for Mobile to Mobile, Fixed to Fixed and/or Fixed to Mobile or Mobile to Fixed (intermodal) connectivity. And now VOIP applications like Skype coming into picture, the situation complicates further.