We initiated on Intelsat this week with a Buy rating and $54 price target (+131%), based on our view that the C-Band auction will see much higher values than most investors realize (see initiation report HERE). One of the most common questions we have received is around what precedent transactions for similar spectrum in Europe tell us about the price the C-Band might fetch in the US. We follow up with some quick thoughts on the most recent European precedent, the 3.5GHz auction in Germany, in this brief note. Bottom line: the German auction value was very supportive of the US seeing $50BN in auction proceeds (the crux of our Intelsat Buy thesis and $54 target).
Baseline: German Auction Comp Suggests $40-67BN In US C-Band Auction Proceeds
Germany auctioned the 3400-3700MHz frequencies earlier this year for a total of $4.8BN (4.2BN EUR), or $0.19 / MHz-POP. Across the most recent comparable auctions700MHz and 900MHz sold in a 2015 German auction for $0.24 / MHz-POP, vs. the US 600MHz which sold for $0.91 / MHz-POP (4x the German comp) in 2017. 1.8GHz sold in the 2015 German auction for $0.33 / MHz-POP, compared to US AWS-3 which sold for $2.66 / MHz-POP (8x the German comp) in 2015. , US spectrum has sold for 4-8x the price of German spectrum on a per MHz-POP basis, implying a C-Band auction prices of at least $0.73 / MHz-POP. With 180-300MHz being made available in the US C-Band auction, this suggests anywhere from $40-67BN for a US C-Band auction (and potentially twice this amount). In addition, we think that there are several reasons why the US should see higher prices than the German auction.
Reason #1: More Bidders In US
The first reason we think the US might see higher prices than the German comp is the sheer number of bidders. The German and US markets both feature 3 MNOs, as well as a fourth entrant who is seeking to become an MNO (1&1/Drillisch in Germany; DISH in the US). Unlike the German market however, the US also has large well-capitalized cable companies with growing wireless ambitionsSee slides 31-32 in our Intelsat initiation HERE; in Germany, cable assets are almost entirely owned by one of the MNOs (Vodafone, following the Liberty transaction). All else equal, this should drive incremental bidding tension in the US.
Reason #2: Fewer Competing Uses Of Capital For US Bidders
The German 3.5GHz auction was conducted along with an auction of 2.1GHz spectrum. If bidders were balance-sheet or budget-constrained, bidding on the 2.1GHz spectrum was a drain on resourcesNotably the 2.1GHz auctioned was licensed nationwide (unlike the 2.5GHz the FCC plans to auction next year) with full-power (unlike the CBRS band the FCC plans to auction next year). . Altogether, the 2.1GHz auction raised $2.4BN while the 3.5GHz auction raised $4.2BN in Germany; had the bidders been able to allocate all of their resources to the 3.5GHz auction, they might have spent as much as $6.6BN for the spectrum. At this level of spend, the German 3.5GHz would have sold for $0.30 / MHz-POP, implying $63-105BN in spend for the US C-Band auction.
Reason #3: Onerous German Regulations Unlikely In US
The 3.5GHz licenses auctioned in Germany came with significant regulations for incumbent MNOs that are unlikely to be present for the US C-Band. The spectrum came attached with aggressive buildout deadlines, requiring 98% HH coverage with 100Mbps in just 3 years’ time, among various other requirements (see more details on slide 10 HERE). In addition, and perhaps more importantly, the incumbents who won spectrum at auction had to offer roaming to new entrants, with the German regulator involved in rate-setting. We don’t think either of these rather onerous conditions is likely to come to fruition in the US, given a long-standing history of free market spectrum rights.