Interested in further auction analysis and wireless strategy? Contact us at or 703-506-0041.

Friday, February 15, 2008: Bidding at the end of Round 81

License MHz Aggregate Total PWBs Aggregate Reserve Delta % of Reserve Total Pops Current Cost per MHz-pop
A 12 $3,937,588,100 $1,807,380,000 $2,130,208,100 218% 285,620,445 $1.15
B 12 $9,112,246,800 $1,374,426,000 $7,737,820,800 663% 284,720,605 $2.67
C 22 $4,748,268,000 $4,637,854,000 $110,414,000 102% 285,620,445 $0.76
D 10 $472,042,000 $1,330,000,000 $(857,958,000) 35% 285,620,445 $0.17
E 6 $1,180,244,200 $903,690,000 $285,620,445 131% 285,620,445 $0.69
Total 62 $19,450,389,100 $10,053,350,000
Graph of FCC Auction 73 New Bids at the end of Round 81

(Click to enlarge)

Graph of FCC Auction 73 Proceeds vs. AWS Auction Proceeds at the end of Round 81

(Click to enlarge)

Once again, bidding was very consistent throughout the day, although daily net proceeds are down slightly to $26.7M. The E block was the big winner today netting over $12.4M with the A block a distant second at $7.9M. The B block rounded out the day's activity at $6.4M. Today's bidding brought the total for the auction to $19.45B.

Yesterday's graph showing the overall value of bids placed by round has not changed much, with total bidding units down slightly but still hovering around 11M bidding units per round. However, today's first graph shows the raw number of bids placed per round, and indicates that although the number of bids placed in A and E has been slowly declining, the number of bids placed in B started to rise yesterday afternoon and continued this morning. This indicates a shift of focus toward even smaller licenses, which is further confirmed by the fact that two of the licenses held by the FCC were picked up today, one in South Carolina and one in Virginia. Hopefully this trend toward smaller licenses is also a sign that we are near the end.

The second graph at right shows how the cost per MHz-pop compares between Auction 73 (700 MHz) and Auction 66 (AWS), the latter of which was held in September 2006 and represents the last major auction of spectrum to be used for mobile wireless communications. As one might expect, the 700 MHz spectrum netted more per MHz-pop in every category, although the 700 MHz C block barely holds an advantage over the similarly sized AWS F block, presumably due to its open access requirement. As most people realize, 700 MHz is inherently more valuable than AWS because of its better propagation characteristics, which results in fewer cell sites to cover the same area. For those who may not know, AWS uses a 2.1 GHz downlink paired with a 1.7 GHz uplink, which can require 3X to 4X more sites than 700 MHz. But 700 MHz also has an advantage in that it requires no spectrum clearing. AWS winners had to incur additional expenses to relocate services that are currently operating in the band, whereas 700 MHz will be free and clear of its current UHF TV signals by federal mandate after the Super Bowl in 2009. The width of the bars on the graph is proportional to the bandwidth of each block, and whereas the cost per MHz-pop is a normalized value, one would expect larger chunks of spectrum to command a higher premium, since they are better suited for next generation wideband technologies, and in most cases this is true. Although Auction 73 is technically not over yet, we don't expect these values to change much.

There is no bidding on Monday due to the holiday, so our next update will be on Tuesday, February 19. We are expecting the FCC to announce an increase to eight rounds per day soon, but bidding on Tuesday will use the current six-round format.

Thursday, February 14, 2008: Bidding at the end of Round 75

License MHz Aggregate Total PWBs Aggregate Reserve Delta % of Reserve Total Pops Current Cost per MHz-pop
A 12 $3,929,695,600 $1,807,380,000 $2,122,315,600 217% 285,620,445 $1.15
B 12 $9,105,888,900 $1,374,426,000 $7,731,462,900 663% 284,636,891 $2.67
C 22 $4,748,268,000 $4,637,854,000 $110,414,000 102% 285,620,445 $0.76
D 10 $472,042,000 $1,330,000,000 $(857,958,000) 35% 285,620,445 $0.17
E 6 $1,167,826,000 $903,690,000 $264,136,000 128% 285,620,445 $0.68
Total 62 $19,423,720,500 $10,053,350,000
Graph of FCC Auction 73 New Bidding Units at the end of Round 75

(Click to enlarge)

Bidding was extremely consistent throughout the day, with the overall bidding levels remaining very constant at about 13M bidding units per round. In other words, the auction did not get much closer to completion due to consistent bidding on smaller licenses. The auction netted nearly $28M today, about $1M less than yesterday, and once again the A block netted the most at $13.3M, with E block close behind at $12.3M, and the B block a distant third at just over $2M. Bids on the Pacific package of Guam and American Samoa increased the C block total by $38,000.

Overall, it was a pretty boring day, so it may be more interesting to talk about the action in the C block back in Rounds 27 to 30, when the sum of the bids on the individual licenses finally exceeded the package bid placed in Round 17. We believe that Verizon pulled off this amazing coup, but as a result has not been able to place a bid since Round 30 because the sum of their PWBs has exceeded their total eligibility since Round 30. This occurred because the six individual C block licenses covering the continental US were bid on two and three at a time, which requires less eligibility than the total package, but ultimately results in converting old bids that were not provisionally winning to bids that are. This was first suggested in the February 5th post below, but after further investigation we are convinced that this is the case, although the details of the analysis are beyond the scope of this post. Please e-mail us if you would like more information on this theory.

If we are correct that Verizon has been rendered ineligible to place any further bids, then this is significant for the D block. That is, if Verizon holds the Round 1 bid on the D block, but is currently over-extended due to their brilliant play on the C block, then there is absolutely no possibility that D will be bid up to the reserve. It is now mathematically impossible for anyone to outbid Verizon on any of the big licenses in the C block and reduce their PWB total to below their eligibility.

So why would Verizon have gotten involved with the D block at all? Maybe at one time they were genuinely interested in winning it and were willing to compete for it (perhaps), but when no competitor showed up, Verizon's Round 1 bid on the D block greatly helped them achieve their goal in the C block. It did this by giving them much more flexibility than other bidders with regard to the 80% activity rule. For example, let's say that Verizon made an upfront payment sufficiently large to cover both the entire C and D blocks, i.e., $410M ($282M + $128M). In each round they must have been active on 80% of this, or 328M bidding units worth of licenses. But since the D block was a provisionally winning bid in Round 1 worth 128M bidding units and since this counted toward Verizon's eligibility in every round, they actually only needed to bid on an additional 200M bidding units in each round. This effectively reduced their eligibility requirement on licenses outside the D block to 71% (=200/282). If they had let their eligibility decrease to 300M at any point, the same math shows an effective activity requirement of just 65%. This may have proven to be a critical advantage for Verizon.

Wednesday, February 13, 2008: Bidding at the end of Round 69

License MHz Aggregate Total PWBs Aggregate Reserve Delta % of Reserve Total Pops Current Cost per MHz-pop
A 12 $3,916,383,000 $1,807,380,000 $2,109,003,000 217% 285,620,445 $1.14
B 12 $9,103,823,300 $1,374,426,000 $7,729,397,300 662% 284,579,600 $2.67
C 22 $4,748,230,000 $4,637,854,000 $110,376,000 102% 285,620,445 $0.76
D 10 $472,042,000 $1,330,000,000 $(857,958,000) 35% 285,620,445 $0.17
E 6 $1,155,497,600 $903,690,000 $251,807,600 128% 285,620,445 $0.67
Total 62 $19,395,975,900 $10,053,350,000
Graph of FCC Auction 73 Daily Proceeds at the end of Round 69

(Click to enlarge)

Graph of FCC Auction 73 New Bidding Units at the end of Round 69

(Click to enlarge)

The auction slowed down even more today and netted a mere $29M. For the first time in over a week, the E block was not responsible for the lion's share of the FCC's daily proceeds. Today, that distinction went to the A block, which netted $12.3M to the E block's $11.1M. The graph at right shows the net proceeds to the FCC by block since bidding slowed down on February 4. However, it's worth noting that this week's proceeds are miniscule compared to the first seven days of the auction. In fact, each day during the week of January 28 to February 1 netted the FCC at least $2.4B, with Thursday January 31st netting the biggest one-day total of over $4B. $2.6B of that day's total was due to bidding activity in the B block. The week of January 28 was by far the most lucrative week of the auction, as total bids increased nearly $15B over the course of the five days of bidding, and more than half of the weekly total was attributable to the B block.

In addition to a reduction in the FCC's daily proceeds, the second graph at the right shows very clearly that bidding activity has slowed down dramatically in all the spectrum blocks, including the E block. Again, this graph shows the value of new bids in terms of bidding units (i.e., not the price paid for the license, but rather the "value" the FCC assigned to each license). Therefore the graph is an excellent gauge of bidding activity and represents the remaining eligibility of the active bidders. It may take several more days for things to stop completely, but it's safe to say that most of the fun is now over.

It appears that the bidder who mounted the E block attack in Round 37 is currently holding LA, Philly, and DC, and with less certainty, New York and Miami. As the total new bids in the E block are still around 5M bidding units per round, and around 13M in the entire auction, it's possible that one or more of these losses may be regained by the victim, but we feel that this is unlikely. Smaller licenses such as San Antonio, Richmond VA, Toledo, Honolulu, and Omaha are still being hotly contested, but for the most part, the war for the largest licenses appears to be over.

One of the A block bargains I mentioned in a previous post was involved in a small bidding war today that started in Round 66, as the Pittsburgh license received its first bid since Round 26. The license was then counter-bid in Round 69, but still stands at just $0.24 per MHz-pop, up from the bargain basement price of $0.23.

Tuesday, February 12, 2008: Bidding at the end of Round 63

License MHz Aggregate Total PWBs Aggregate Reserve Delta % of Reserve Total Pops Current Cost per MHz-pop
A 12 $3,904,074,000 $1,807,380,000 $2,096,694,000 216% 285,620,445 $1.14
B 12 $9,098,022,500 $1,374,426,000 $7,723,596,500 662% 284,579,600 $2.66
C 22 $4,748,170,000 $4,637,854,000 $110,316,000 102% 285,620,445 $0.76
D 10 $472,042,000 $1,330,000,000 $(857,958,000) 35% 285,620,445 $0.17
E 6 $1,144,415,600 $903,690,000 $240,725,600 127% 285,620,445 $0.67
Total 62 $19,366,724,100 $10,053,350,000
Graph of FCC Auction 73 Cost per MHz-Pop at the end of Round 63

(Click to enlarge)

Graph of FCC Auction 73 New Bidding Units at the end of Round 63

(Click to enlarge)

The auction netted another $65M today, as bidding progressed in the A, B, and E blocks, and once again there was no significant action in C or D. This brought the total proceeds to $19.366B and counting, with the E block once again accounting for the majority of the day's increase. Bidding continued to generally subside, however a late-day decrease in B coincided with an increase in activity in A, as can be seen in the second chart to the right, indicating that one or more bidders has decided to switch their focus.

The most notable downturn in activity was in the E block where the bidding war for some of the bigger licenses continued, but seems to be declining. This may be a sign that more than two bidders have been involved, especially in the smaller licenses, and that the smaller bidders are starting to refrain from placing counter-bids when they are outbid by the larger players. Rather than placing bids, smaller bidders are choosing to let their eligibility decrease or take waivers. Most of today's rounds featured one or more bidders exercising proactive waivers, another indication that the auction is approaching its conclusion.

The average number of new bids on the A block licenses was 18 per round today, down from 23 yesterday and 29 on Friday. The B block averaged 32 bids per round today, down from 41 yesterday and 54 on Friday. The E block was less consistent with an average 71 bids per round today, 80 yesterday, and just 64 on Friday. However a quick look at the graph shows that the value of the bids placed on Friday was generally the highest of the three days, indicating that focus of new bids is slowly shifting away from the larger licenses.

Only ten licenses are still held by the FCC, all of which are in the B block. Of these, three are MSAs in North Dakota (Bismark, Fargo, and Grand Forks) and six of the other seven are RSAs in Louisiana, the Carolinas, Tennessee, and Virginia. The FCC got the tenth license back today, as a bid placed in Round 1 on the American Samoa license was withdrawn in Round 61.

Monday, February 11, 2008: Bidding at the end of Round 57

License MHz Aggregate Total PWBs Aggregate Reserve Delta % of Reserve Total Pops Current Cost per MHz-pop
A 12 $3,893,990,900 $1,807,380,000 $2,086,610,900 215% 285,620,445 $1.14
B 12 $9,093,813,400 $1,374,426,000 $7,719,387,400 662% 284,636,891 $2.66
C 22 $4,748,169,000 $4,637,854,000 $110,315,000 102% 285,620,445 $0.76
D 10 $472,042,000 $1,330,000,000 $(857,958,000) 35% 285,620,445 $0.17
E 6 $1,093,623,400 $903,690,000 $189,933,400 121% 285,620,445 $0.64
Total 62 $19,301,638,700 $10,053,350,000

The bidding war for the E block continued today, which made the E block the most active block for the fourth day in a row, adding nearly $57M to the FCC's bottom line. The total take for the day was about $72.5M, which brought the total to date to just over $19.3B. Bidding in the A and B blocks continued their very slow decline, and the only bids in C today were on American Samoa and Guam. And as expected, the D block remained untouched and is still well below its reserve.

The chart below shows the bidding activity in the Top 30 E block licenses since Round 34 and has been color-coded in an attempt to identify the attacker and the victim. We have made two simplifying assumptions: that there are only two parties involved in the war for these licenses and that in each case the bids alternate between the attacker (in red) and the victim (in grey). This is reasonable if you assume that neither party would have a reason to increase his own bid on any given license. In addition, we believe the attacker used a proactive waiver in Round 36 and therefore did not bid in that round (see the next bullet). As a result, the Round 36 bids on Chicago and Tampa were very likely the bidder we call the "victim" regaining the PWB on these licenses. We are also assuming that the victim was well established with PWBs on most of the top E block licenses prior to Round 37. Clearly the examples in Chicago and Tampa show that this is not necessarily the case, but our assumption is based on the premise that Qualcomm is the victim and has had an active interest in the E block since the start of the auction, whereas the attacker's interest is less established. We maintain that the chart is most definitely not 100% correct, but it helps to illustrate what may be happening, and gives the reader some insight as to how the bidding has progressed. The other 146 licenses in the E block complete the story, and there are clearly more than two parties still vying for many of these, which creates an even more complex story that is very diffcult to predict. Regardless, the chart shows our best guess at the current state of the major activity in E, and indicates that it will be the attacker's turn to bid tomorrow morning in Round 58.

It appears that the attacker has taken at least two of his three waivers now, having taken them in Rounds 36 and 43. Round 36 was the first round of Stage 2, in which the activity requirements were increased from 80% to 95%, and it is likely that the bidder planning the attack in E took a waiver to allow other bidders to outbid some of his PWBs, counting on the fact that the 95% activity rule would drive more bids. This likely helped the attacker maximize his new bid eligibility for Round 37, so he could bid on a higher number of large E block licenses, although it's not clear why he may have had this objective, other than the element of surprise. With few exceptions, the bidding pattern is very regular up to the second waiver taken in Round 43, but the purpose of this waiver is not as clear. It's quite apparent that it was the attacker that took it, as it was his turn to bid, and perhaps it was taken to allow time for a strategy session, as the bidding patterns become much less regular afterward, the New York license enters the war in Round 46, and many other smaller licenses get involved in the action. Someone also took a proactive waiver in Round 56, but it's not clear whether this was either of the two parties assumed to be involved in the major bidding, and if the chart is correct it could not have been the attacker. It did cause the value of new bids to drop by about 10M bidding units, or about 40%, so there could still be a very significant third player involved.

Other than the waivers taken, the E block war is showing few signs of stopping. At $0.64 per MHz-pop, the spectrum is still relatively cheap, and clearly has more value to some of the bidders still involved. We'll have more developments tomorrow.

Graph of FCC Auction 73 E Block Bidding War at the end of Round 57

(Click to enlarge)

Next Week | Previous Week