Satellite Spectrum - Auction Models
None of the auction models will work. Why? As it is analogous to administrating antibiotics to a patient who is suffering from a viral infection.
I have written a series of articles on this topic, but this one I will like to focus on why none of the auction models will work when it comes to the assignment of satellite spectrum. First I will go through the fundamental principles, which then will be followed by a detailed discussion on specific models- both TRAI (in CP) and the one proposed by Starlink.
Exclusive Auction
It can be further divided into two types. Option 1 - By slicing the band into small blocks (like a terrestrial auction). Option 2 - Without slicing, i.e. assigning the full band to the winner.
Option 1 (Sliced in Blocks)
This option is pictorially depicted below. Let's say the total spectrum available for auction is 500 MHz. We create 4 blocks, 125 MHz each, and sell them in the auctions as we do for the terrestrial spectrum.
Key Problems
Satellite Internet Peak Speeds will decrease by 75%. Means? If speeds of 100 Mbps were possible with a full block of 500 MHz, it will now get driven down to only 25 Mbps.
>95% of the capacity of the sold block will get wasted, as no one will be able to use it unless it is shared with other players.
Sharing will not work as some will not be interested, as their business model might be opposed to the one seeking to share.
Huge conflicts might happen between players on the type of sharing to be used, as a specific model that favors players with a large number of satellites, might not work for the players with fewer numbers.
Prevent the Introduction of new technologies like 6G even when a lot of capacity will lie unused (Satellites services can’t reuse spectrum with similar efficiency as terrestrial players can).
New players (Startups) will be driven out, even when a lot of capacity will be available to support them in the same block of spectrum.
Option 2 (Full Band)
All available spectrum in a band is sold as a single block.
Key Problems
The market will become uncompetitive as only one player will play in the market and the rest will get excluded.
The winner will become the gatekeeper, as he will have the power to decide commercial terms and rules of sharing.
Prevent the Introduction of new technologies like 6G even when a lot of capacity will lie unused (Satellites services can’t reuse spectrum with similar efficiency as terrestrial players can).
Players with less muscle will be driven OUT as their power to negotiate will be weak.
Even specifying rules of sharing (by TRAI) will not work, as TRAI also has to spell out the commercial terms of sharing in advance, so that same is not used as a tool to exclude competition.
It is a textbook case for the CCI (Competition Commission of India) to intervene and therefore is good for the winner of the spectrum as well.
Non-Exclusive Auction
This is a strange model, as auctions by itself are always exclusive. As the spectrum is used as a tool for acquiring the right of accessing the market - the end objective for all auctions. That is why auctions of licenses (euphemistically can be called Non-Exclusive Auction) do not work, as the value of the license will get diluted in case new players are allowed to enter the market at a later point in time. In other words, the bidder will not know what they are bidding for.
% Revenue Share (Starlink Model)
In this model, a bidder will bid for a percentage share of its revenue to be paid as license fees to the GOI. To prevent non-serious players, the suggestion is to have eligibility criteria, upfront fees (non-refundable), and rollout obligations. This auction design has many fatal flaws. These are enumerated below.
Key Problems
The revenues emanating from this auction will be uncertain. Even if all agree to pay the same rate (% of revenues), the outflow will be different for all - as not all players will have the same revenues. In fact, some will not have any (Zero Revenue Share), as they might use the spectrum for backhauling their existing 4G/5G BTS traffic in remote areas.
The model unlocks a Huge Opportunity for Arbitrage. The players who offer subsidies service with the objective of capturing market share will get rewarded - having to pay less license fees.
The model will allow players to Block Capacity and prevent others from using it. As upfront payment can’t be set at an exorbitant number the same will be used to sit on the spectrum and do nothing as a role out for satellite services can’t be accurately defined (satellites are moving in the sky and not stationary as BTSs on the ground are). Also, history is a testimony to the fact that players after having paid huge sums and did not roll out for a long time.
The model will lead to huge conflict between players sharing the same block of spectrum. As players will NOT have the same preferences for all sharing models (See how FCC is working diligently to facilitate and negotiate between entities). And for the sharing model to work, each of the participating entities to share vital information to each other (some sensitive in nature) and such coordination cannot work unless facilitated by a regulator. However, both TRAI and DOT will be helpless, as they will not be able to intervene after having auctioned the spectrum.
Prevent the Introduction of new technologies like 6G even when a lot of capacity will lie unused (Satellites services can’t reuse spectrum with similar efficiency as terrestrial players can).
TRAI’s Models
Option 1 (Exclusive Assignment with Mandatory Sharing)
This model is similar to Option 1 of the “Exclusive Auction” as described above, attached with all the shortcomings mentioned above in that section with some additional ones. The diagram is reproduced below for ready reference.
Key Problems with Mandatory Sharing
The TRAI has to spell out the rules for sharing in advance, along with pricing. Why pricing? Because unless they do that the winner will have the opportunity to quote unreasonable rates and use it to prevent a desperate player (with a different business model) from sharing the spectrum with the objective of throttling his peak speeds- bad for the consumers. Doing so will also make the overall business unflexible.
TRAI or DOT will have to keep revising these rules on a periodic basis. Why? Not doing so will demotivate players to use the spectrum efficiently - as whatever investments they do (for increasing efficiency) will end up creating more room for others - sharing their block of spectrum.
Players will not be able to bid with clarity as they will not know how many of these new players they will enter in the future, thereby further dividing the pie (market) into smaller fragments. This clause is only applicable if TRAI allows horizontal entry of players (who did not bid in the auction).
DOT and TRAI have to keep intervening for resolving conflicts between players on sharing and pricing - disturbing the overall dynamics of the market.
Prevent the Introduction of new technologies like 6G even when a lot of capacity will lie unused (Satellites services can’t reuse spectrum with similar efficiency as terrestrial players can). As having auctioned the spectrum, the same can’t be opened for new services.
Now if DOT and TRAI have to keep interfering all the time to protect the interest of new players wanting to take advantage of the provision of “mandatory sharing”, then how is this model different from an administrative assignment? And what is the benefit that a player gets by participating and winning the spectrum in the auctions if they have to mandatorily open up their block for others at terms (rules) which might change dynamically?
Option 2 (Exclusive Assignment With Full Access)
The following figure explains the TRAI’s model graphically.
Even this model is marked with the same shortcomings. The few important ones are described below.
Key Problems
The biggest one is the need to create “Artificial Scarcity”. Why? That too when there is none (TRAI acknowledges this fact). It is foolish of us to do this as this will lead to huge capacity being wasted and prevent TRAI from using it to introduce new technologies like 6G which might overlap on the same block of spectrum.
Neither TRAI nor DOT will be able to coordinate sharing - which is a must to prevent bullying by strong players with muscle and to motivate the players to keep investing to improve efficiency. Without such monitoring, innovation will get discouraged as those who invest in improving efficiency will end up creating room for others (who chose not to invest).
Startups and other small players will get excluded (even when capacity will be available to support them), as the market will get artificially restricted to a few players only, else there will be no price discovery (as per TRAI).
Bad for consumers (due to decreased competition) and the country, as a huge quantum of spectrum capacity will lie unused and get blocked for a long time.
Conclusion
Hence, it is clear from the above none of the auction models work in the case of the assignment of spectrum for satellite services. Each of them has significant shortcomings - the biggest one is the regulator’s inability to intervene and facilitate and coordination sharing of spectrum between various players, once the same is auctioned. As if it did, then it will make the auctions futile. And in case they don’t then it will end up wasting spectrum, demotivating the satellite players to innovate, and preventing the introduction of new technologies like 6G (which might end up overlapping the same bands).
Therefore, why auction and lose total control of these bands when India will stand out alone compared to other important nations, who have consciously chosen to retain full control of these bands (by making administrative assignments) for the purpose of enabling efficient usage and keeping the possibility alive of deployment of advanced technologies (6G and beyond)? Let me know what you think.