Indian 5G auction is just a few days away. Typically a spectrum auction is laced with many complicated rules, so much so that a layman is sure to get overwhelmed. Hence I thought of taking this opportunity to simplify these rules so that we all can visualize how this auction will run. The reason this is important as it will help us to understand the bidder’s ability to preserve their empowerment (enabled by the earnest money submitted, and eligibility points assigned) to bid in various circles, and their capability to switch between bands/circles during the auction. In the process, we can also analyze what flexibility the operators will have in submitting their bids, and how the decisions made during the course of the auction will impact the final outcome.
The Auction Process
In a conventional auction, a bidder is required to state a price, and the one with the highest value wins the contract/asset. Whereas in a spectrum auction, the price is announced by the auctioneer based on certain predecided rules, and the bidder just has to agree or disagree with that price. If the bidder agrees with the stated price, then it is tantamount to a positive bid, else it will get concluded as a “no-bid”. The bidder agreeing to the announced price will also have to declare concurrently the number of blocks of spectrum (page 58 of NIA) that it is interested in acquiring. Now the bidder will get proclaimed as a PWB (provisional winning bidder) only if it gets identified with all others who are similarly placed and ranked based on some complex rules (page 71 of NIA) with the purpose of mapping the demand (emanating out of these bidders) with the supply of available spectrum for sale. The nonbidders and those falling outside this group are categorized as non — PWBs. These non-PWBs (including the PWBs of the current round) will get an opportunity to bid again in the next round till there are no more bids in any of the circles of all the bands and when the activity factor (which will discuss later in this note) is set by the auctioneer as 100%. The price of the 1st round is mapped to the reserve price set by the government, and those of the subsequent rounds are incrementally raised based on a well-defined rule which is mainly driven by a factor called “Excess Demand” (page 76 of NIA). If the Excess Demand (the quantum by which the demand exceeds the supply in a specific circle) is zero or positive then the price of the next round in that circle is raised else it is kept the same as the previous round.
The Bidder’s Empowerment
The bidder’s empowerment to bid in various circles across bands emanates from the value of the EMD (Earnest Money Deposit) that it has submitted to the DoT and the corresponding EPs (Eligibility Points) that gets assigned to it by the auctioneer mapped to EMD. EPs are extremely crucial as in a competitive auction the bidder's ability to switch circles across and within bands depends upon it. Each circle of a band has an EMD and EPs value clearly mapped and laid out in the NIA. Both EMD and EPs values also linearly scale with the number of slots (minimum quantum of the spectrum that is up for sale) that a bidder decides to bid in a circle. The more the number of slots the bidder bid, the greater the quantum of EMD and EPs will get attached to the account of the bidder for the purpose of calculating his overall empowerment. The EMDs, EPs, and RP (reserve price) are correlated. The following charts map these matrices for various circles across all bands.
Now for a bidder to preserve his empowerment to bid in a subsequent round (at par with the earlier round), he must have a sufficient amount of EPs. Also at no point in time during the auctions, his overall bids in various bands/circles can exceed his original submitted EMD — a possibility, as the EPs are always set higher than the corresponding EMDs in various circles. This is to facilitate proper price discovery by enabling the bidders to switch across bands. Now if the EPs were set correlated with the EMDs/RPs, then the movement across circles would get restricted as the RPs vary significantly even in bands of similar nature (example 800, 900 MHz bands).
Hence, EPs, original submitted EMDs, and band cap rules are limiting factors constraining bidder’s empowerment. But, how a bidder loses his empowerment during the auction process will be discussed in the follow-up sections.
The Activity Factor
The activity factor defines the minimum bidding intensity of the bidder that a bidder needs to bid without diluting his empowerment in the subsequent round. This factor is controlled by the auctioneer and when it is set below 100% it gives a bidder some flexibility to preserve his empowerment (EPs) to bid in a subsequent round even when he has decided not to consume all his EPs in the current round, by bidding to his maximum possible. Note this empowerment is not absolute, but limited and depends on the value of the “Activity Factor” that the auctioneer decides at the start of the auction process. Typically at the start of the auction process, the auctioneer keeps this factor as 80%.
The activity factor is always mapped with the EPs (Eligibility Points) which the bidder needs to ensure during the course of the auction to preserve flexibility. Let's say the bidder is assigned 100 EPs by the auctioneer at the start of the auction process, which is based on his submitted EMD value. While assigning the EPs the auctioneer will assume that the bidder can bid in round 1, without any constraints in all combinations of circles with the highest EPs attached. Note the aggregated EMD values of these chosen circle combinations are at par and mapped with the bidder’s submitted EMD. Now assume that the auctioneer notifies an activity factor of 80% during the start of the auction, and the bidder chooses to bid a different circle/band combinations that the auctioneer originally envisaged, then his calculated EPs can be lower than the originally assigned value, let's say 60. Then the bidder's subsequent empowerment of EPs for the next round (round No 2) of the auction is simply = 60/0.8 = 75 (25 points lower than the original). In other words, the bidder has lost 25 EPs in the first round of the auction and now has to maintain 75 EPs to preserve its empowerment during the course of the auction.
Hence, till the time auctioneer keeps the activity factor as 80%, the bidder can continue to bid for 60 EPs, and still maintain the empowerment of 75 EPs in a subsequent round during the auction. Now assume that the auctioneer raises the activity factor from 80% to 90% in round №5, then the bidder now has to bid (75 x 0.9 = 68 points) in order to preserve his full empowerment of 75 points. And once the activity factor is set as 100%, then the bidder has no choice but to bid the full 75 points, or else the balance number of points will get lost. But why does the auctioneer keep raising the activity factor? He does so to preserve the intensity of the auction, in case he finds intensity is going down a few rounds after a certain activity factor was announced.
Intensity of Auction
The intensity of the auction depends on the demand and supply equation. The first sign of demand can be ascertained by comparing the EMD value submitted by all the operators with the total aggregated value based on the spectrum put out for sale in various circles/bands. Now this EMD value can be calculated as Rs 47802 Cr. But the total EMD of all the operators have submitted to DoT stands just at Rs 21800 Cr — which means the demand is just 45% that too before the start of the auction. Also, note that the demand can even go down even further — as the submission of an EMD value does not guarantee that the bidder will bid for that number. He might choose to bid on a much smaller number.
Therefore this auction can be classified as a non competitive auction.
Non-competitive Auction — Implications
The implication of a non-competitive auction is that the bidder has no flexibility to switch circles/bands once the original bids are placed. As he can only switch when he is no longer a PWB in that circle of a chosen band. That will happen only if there is additional demand in that circle at a minimum that is equal to the supply. Why? As only then price increments will happen (excess demand ≥ 0), and the bidder gets an opportunity to bid either in the same band or in some other bands/circles mapped to his EPs/EMDs. In no case, the aggregated EPs and EMDs of the bidder can be exceeded the authorized values while he makes such choices.
Now, initially, since the activity factor will not be 100%, and hence the bidder will have some points in spare to use in subsequent rounds when the activity factor is raised. But even at that time, it is highly unlikely that the bidder will place bids overlapping with others due to their diverse interests in the various bands. For example, Bharti will be limited to 900, 1800, & 2300 MHz bands (on top of the 5G bands of 3.5 & 26), and RJIO will be limited to 700, 800, and 1800 MHz (on top of 5G bands of 3.5 & 26).
Since a lot of spectrum is put out for sale in these overlapping bands and therefore the chances of tussle are next to impossible.
Hence, we can safely conclude that in this auction the operators will place bids in those bands/circles of their interests and will get no opportunity to make any switch. Also, there will be no price increase in this auction, and all spectrum will get taken at the reserve price. Means — all the above described rules designed by the auctioneer to drive competitive bidding will remain totally unutilised.
Practical Scenarios
Now let's discuss some practical scenarios of the permutation and combination possible based on the EMDs values submitted by bidders.
RJIO Option 1
In this option RJIO bids 700, 800, 1800, 3500, & 26000 MHz band. The RJIO’s demand for the quantum of spectrum in these bands is tweaked such that it maps with its submitted EMD value. The following charts provide us with a snapshot of this option.
A) Spectrum Matrix
B) EMD Matrix
C) EP Matrix
D) Outflow Matrix
One can clearly see that if RJIO bids option 1, then it will lose 24.37% of its EPs at the start of the auction (round 1), and it will end up spending Rs 1.25 Lakh Cr.
RJIO Option 2
In this option RJIO bids on all bands as in option 1 except the 800 MHz. The RJIO’s demand for the quantum of the spectrum has been tweaked such that it maps with its submitted EMD value. The following charts provide us with a snapshot of this option.
A) Spectrum Matrix
B) EMD Matrix
C) EP Matrix
D) Outflow Matrix
In this option also RJIO will lose 24% of its points, but it will be left will a little more EMD of Rs 13853 Cr vs 13996 Cr in option 1, but it will end up spending almost similar amount as in option1.
Bharti
For Bharti, I have assumed only one option, i.e it will bid in 900, 1800, and 2300 MHz bands apart from the 5G bands of 3.5 & 26 GHz. The difference between Bharti and RJIO is that — Bharti has been assumed to not bid for 700, and bid only 80 MHz spectrum in 3.5 vs RJIO’s 100 MHz bid.
A) Spectrum Matrix
B) EMD Matrix
C) EP Matrix
D) Outflow Matrix
Hence Bharti will lose 34.35% of its EPs in the first round of the auction, and it will end up spending Rs 45930 Cr.
VodafoneIdea
For VI, I have assumed only one option, i.e it might bid in 900, 1800, apart from the 5G bands of 3.5 & 26 GHz. The difference between VI and Bharti is that — VI has been assumed to bid only 40 MHz spectrum in 3.5 vs Bharti’s 80 MHz bid, and only 600 MHz in 26 GHz band vs 800 MHz for both Bharti and RJIO.
A) Spectrum Matrix
B) EMD Matrix
C) EP Matrix
D) Outflow Matrix
VI will lose 39.21% of its points in the first round itself and will end up spending Rs 19,379 Cr
Adani Option 1
For Adani, I have assumed that it will bid only on the 26 GHz band. Now in this option, I have assumed that it will bid 100 MHz in all the 22 circles.
A) Spectrum Matrix
B) EMD Matrix
C) EP Matrix
D) Outflow Matrix
Adani will lose 63% of its points in the first round itself and will end up spending Rs 700 Cr if it chooses to go with this option.
Adani Option 2
Now in this option, I have assumed that Adani will bid 200 MHz in limited circles.
A) Spectrum Matrix
B) EMD Matrix
C) EP Matrix
D) Outflow Matrix
In this option Adani will lose 55% of its points in the first round itself and will end up spending Rs 764 Cr.
Conclusion
From the above analysis, one can clearly conclude that this auction will be subdued and will get completed in a few rounds, and may not last more than a day. Actually, the auction should ideally get completed in just one round, provided the auctioneer starts the auction with an “activity factor” of 100%. But since most likely he will start with a factor of 80%, and therefore, the auction will run for a few rounds till it closes when the auctioneer raises the activity factor to 100%. All the complex auction rules designed for deciding price increments will remain unused. Also, the EPs will serve a very limited purpose, in case the bidder decides to bid in all the circles/bands of interest from the beginning. However, there will be some value though as the auction will calculate the ranking of the bidders in various circles/bands, which will decide which frequency slots the operators will end up getting. This is important as some slots (especially in the 3.5 GHz band) are cleaner than others.
By going through the above note a reader will get a feel that designing an auction laced with complex rules does not guarantee competitive bidding. It can only be ensured by managing demand and supply and by fine-tuning the reserve prices in such a manner that the bidder has a motivation to bid responsibly — leading to proper price discovery — the whole objective of the auction. The price emanating out of an auction that has a potential to completed in just one round cannot be branded as “market discovered” as the auctioneer didn’t get the opportunity to trigger all the tools in his arsenal (described above) — leading to that outcome. That looks like a case in this auction.
The charts posted in this note are the output screenshot of the tool in the following link.
(Views expressed are of my own and do not reflect that of my employer)
PS: Find the list of other relevant articles in the embedded link.
