machineagenyc - Tumblr blog

machineagenyc · 8 years ago

Text

Incentive Auction Concludes. Machine Age Begins.

Well, here it is. The inaugural post of Machine Age.

I begin where I left off, with the U.S. broadcast band incentive auction.

I want to inform you, especially if you are a U.S. citizen, AKA U.S. consumer, that the incentive auction bombed.

Big time.

Or, it succesfully bombed by meeting the final stage rule and then some.

This statement does not make sense if you are new to the incentive auction hedgemaze. For those foiks, let’s briefly get up to speed, and the statement will make sense at some point in this article, I hope.

The auctions, plural, are over, but the incentive auction is not. At least the rules of the foregone auctions still require the arduous repacking of remaining broadcasters out of cleared spectrum, which is part of the “transition” for them and everyone else who uses spectrum above 614 MHz, and the collection and disbursement of funds from participants. The transition began April 13th of this month with the release of the Closing and Channel Reassignment Public Notice. The transition ends July 13, 2020, 39 months from now, and 12 months later on July 13, 2021 for LPTV stations. (If you need a refresher, check out this article I wrote last year, or the FCC’s incentive auction explainer.)

By measures of law and politics and technology, the incentive auctions are an astonishing success. The Commission and in particular the Incentive Auction Task Force used a cold fusion of legal agility, sophisticated software modeling to reckon financial states and future states and all that pesky electromagnetic stuff causing geospatial conflicts between coverage areas, and good old sweet talkin’ to execute Section 6403 in the Spectrum Act as found in The Middle Class Tax Relief and Job Creation Act of 2012 - a Congressional mandate to repurpose UHF broadcast spectrum for 21st century digital services with scant guidelines on how to get it done.

Some compared §6403 to a Rubik’s cube. I don’t think so. More like this:

youtube

§6403 unleashed the giant stone ball from Indiana Jones upon the FCC, amid slings and arrows and petitions and lawsuits from just about everyone with a stake in or to be had in 600 MHz. The FCC delivered the fine print and execution of the first incentive auction with, all things considered, groundbreaking originality of thought, at breakneck speed.

Roadmap and execution are two things. Outcome is another.

When I look at the outcome of this unprecedented event, I see no substantial benefit to the public and public good on the near and middle horizon. Governments, economists, and engineers may have learned a lot from the failure, but the failure was a heavy cost to society. It is much more difficult to say whether what was learned by the failure will compensate society in the long term. Technologists think that it will.

Please don’t confuse this statement with jabs at the Task Force and countless others who brought the concept to bear the fruit of a new kind auction, which as the technologists argue, may find legitimate use by governments and the private sector all over the world. Someday.

Rather, consider it possible that a brilliant idea is not necessarily a good idea.

The modern world is filled with these ideas.

Is the incentive auction of 2016-17 one of them?

My opinion is yes.

It may be unfair of me, who so recently commenced self-directed study of spectrum regulation and the spokes from which it is made, to lay this ‘yes’ down. Especially with the cheap advantage of hindsight. The incentive auction may be a brilliant idea, but once again, unprecedented. No government has ever executed a two-sided incentive auction. Even the wisest and most benevolent authority cannot eliminate ipso facto causality, unintended consequences or unexpected failure of intended benefits. They can only minimize the probability of negative outcomes using informed strategy.

For the United States, in 2017, the incentive auction was a brilliant idea, but known external costs combined with unknown uncertainty led to a disappointing outcome and disappointing benefits for the U.S. The failure may, may, bring benefits someday. But the blade of innovation could swing the other way, too. Incentive auctions might not end up being as useful after all, or something the first incentive auction kicked up may kick back unexpectedly years or decades down the road.

The concept behind the incentive auction is an innovative system of technologies built around a singular aim: re-zoning TV band spectrum to flexible mixed use mobile broadband spectrum - clearing up an old shantytown to make room for a shining city - without the old tenants, broadcasters, getting upset.

“Innovation” is a loaded and multifaceted word, but, if unintended consequences are confined to a laboratory or code repository, designers of innovation rarely be held accountable for good-faith efforts which go awry or fail to deliver expected results.

But when a government gambles (remember the word: “gambles”) the wealth of its taxpayers on an unproven auction technology for a commodity citizens rely on but do not have specialist knowledge of, with undeniable benefits to special interests, and an outcome that falls billions of dollars short of every calculated expectation and uncalculated promise?

That is something entirely different. Something that deserves a closer look so taxpayers who funded the experiment know what went wrong.

In current social and political environments, a closer look is unlikely to arrive. So let’s put expected results aside.

What were the results? And how do those results compare to those which were promised?

Forward Auction Results

As of last month, 50 bidders committed bids for 2,776 licenses for total gross proceeds of $19,632,506,746. These are legally binding commitments, not checks in the mail.

The auctions lasted approximately nine months, and took four attempts (stages) to succeed. The definition of success is when the FCC + Treasury gather enough bid revenue from forward stagers to cover auction expenses, participating broadcaster compensation checks, a $1.75 billion repack fund, a few public programs, and various expenses. The Task Force named this break-even point the “Final Stage Rule.” Here’s a picture of the timeline.

*Note: 1.17.2017 is the date the FSR was met. Bidding continued til February 10th.

Initially, 99 bidders qualified to participate. By stage four, only 62 bidders remained, and 12 of them abstained, including Verizon and Sprint.

Who were the bidders? In terms of dollars spent:

You may object to binning bids in dollars instead of quantity of licenses won, also that this is a stupid pie chart, but excel was freezing up on me, and I have another visualization of bidding share by license count coming up in a future post.

The point here is only four bidders spent more than 87% of gross proceeds, and walked away with a bit more of the total licenses up for bid.

$19,632,506,746 in proceeds is a lot of money, but it’s revenue without costs.

From the $19.6 billion collected, $10 billion is drawn down and used to pay participating broadcasters to... lavishly retire? Then there are the costs of putting FCC attorneys and engineers to work for years. Then, $1.75 billion given to broadcasters who elected to stay in business and remain on the air to ease the burden of the virtual and physical relocation of transmitter infrastructure. (Except for LPTV stations, who receive nothing.) Finally there are a few important programs to receive a few hundred million.

All told, net proceeds from the auction will deposit $6 billion in the Treasury. These billions are per the Spectrum Act to be used for the purpose of deficit reduction, which virtually all sane U.S. citizens can agree is a good thing.

Result vs. Promised

$6 billion of deficit reduction is better than $0 billion, I guess.

Incentive auction revenues place it on the podium as the second largest spectrum auction in 21 years of spectrum auctioneering.

But a number six with nine zeroes after it should not be permitted to write over public perception.

The Commission, Treasury, and auction economists were all in consensus the incentive auction was sure to bring far and ludicrously north of $6 billion, especially after the legendary AWS-3 auction, and it was upon this consensus the auction was sold.

For its part, the Congressional Budget Office issued a formal response to request for clarification and risks raised by Nevada Congressman Dean Heller. The CBO letter claimed:

>“The Congressional Budget Office estimates that the net proceeds will probably be between $10 billion and $40 billion, with an expected value of $25 billion, the middle of that range.”

The CBO was not alone in its optimism.

Peter Cramton et al. penned this expert report, submitted to the Commission. He and his team at the University of Maryland reasoned:

>“The AWS-3 paired price of $2.72/MHzPop is a timely estimate of 600 MHz auction prices. This price implies forward auction revenues of $84.9 billion for the 126 MHz clearing target (10 blocks). There are good reasons to believe that revenues will be higher than $84.9 billion as a result of the better propagation characteristics of the 600 MHz band and the greater scarcity of low-band spectrum. The AWS-3 auction presents current market evidence that the 600 MHz auction will achieve revenues above $80 billion if 10 unimpaired blocks are auctioned.”

As I wrote in October of last year, after three clearing targets failed and the paired blocks + gap shrank to only 84 MHz to the fourth and final target, my quotation of Cramton here is not meant to skewer him personally, or his discipline. Cramton is a superb auction economist, perhaps among the best in the world, and he contributed meaningful insights intended to maximize the benefits of the auction to all stakeholders including the American People.

And like the CBO, he was not alone. 112 esteemed economists, including one Nobel Laureate, backed a letter to President Obama in support of the auction.

Even with these experts onboard, to sell the auction, the FCC went seersucker on broadcasters, who were understandably frumpy and confused about the idea. Mobile carriers needed assurance affordable contiguous license blocks would unlock. Non-contiguous blocks = no value for incumbent carriers. They also got the tech sector interested with database-controlled cognitive radios in unused “white space” TV channels. Google, Microsoft, Motorola, Qualcomm, and Broadcomm to name a few were interested, and submitted many comments and reconsiderations to ensure white spaces exist in all markets, and that WSDs operate in the guard bands and duplex gap.

Don’t get me started on wireless microphones.

Former Chairman Tom Wheeler was widely quoted as promising a “spectrum extravaganza” in sessions at CES 2016.

IDG’s NetworkWorld reports Wheeler said: “You're going to see lots of interest in selling the spectrum, and lots of interest in buying the spectrum."

>“Wheeler deflected questions about the financial implications of the auction, insisting that he's more interested in the transfer of spectrum it enables than the amount of money that changes hands,” continues NetworkWorld. “The best-case scenario for the auction is to see ‘lots of spectrum sold to lots of wireless carriers,’ Wheeler said. When asked what he sees as the worst-case scenario, Wheeler said it would be that ‘it all blows up and we're wrong.’ But he was quick to reiterate his confidence in the plan. ‘I'll go to the bookies here in Vegas and put my money down against that”

(psssst: gamble)

I hope I’ve made it clear $6 billion is underwater from even the most conservative predictions, by the most credible experts.

It caused the exchange of large sums of money for regional monopolies on the gridiron of wireless infrastructure - spectrum - between people and organizations numbering less than a few hundred - and the 600 MHz frequencies exchanged have a value to the many which is disproportionate to the value ascribed to it by the few who bought them. This has to do with physics, the impending dawn of 5G, and dynamics between the big four carriers who dominate mobile broadband in the United States.

Compare the promises to the results. They are not equivalent. Which makes this blog post from the Task Force after the conclusion a Rubik’s cube all its own:

>“Today marks the end of all clock phase bidding in the incentive auction. This is a noteworthy event for winning bidders and an appropriate moment to appreciate the auction’s success in using market forces to allocate spectrum to its highest and best use...With $19.6 billion in forward auction clock phase gross winning bids, the incentive auction will generate the second highest total proceeds of any Commission spectrum license auction in its 20-plus year history...The public stands to gain substantial economic benefits from mobile broadband utilizing the 84 megahertz of spectrum repurposed by this auction.”

I see the auction’s success in using “market forces to allocate spectrum to its highest and best use.” I do. That’s the brilliance.

But when the FCC via Task Force claims, “The public stands to gain substantial economic benefits from mobile broadband utilizing the 84 megahertz of spectrum repurposed by this auction,” this is loaded in many ways, and from my perspective, deeply untrue.

The FCC insulates itself by making this statement speculative, with the phrase “stands to gain.” The public may but will not necessarily gain. And so grammatically I cannot say it is untrue because no truth is asserted. Even so, I’m assuming this is a carefully worded and be-lawyered pat on the back for a job well done.

Let me be more precise in defining elements of discontent.

The brilliance of the incentive auction design and the magnificence of its execution is overshadowed by:

The sacrifice of the concept and actual decrease in/of public spectrum

Gross and disproportionate handouts to a small number of broadcast licensees who pocketed what is as close to free money as it comes, versus the disappointing sum deposited in the Treasury.

The limited utility for the mobile ecosystem of adding 70 MHz of 600 MHz spectrum to the portfolios of only two mobile carriers, with T-Mobile taking 45% of licenses nationwide

The sacrifice of the concept and actual decrease in/of public spectrum

What I mean by public spectrum is radio spectrum which is neither unlicensed nor owned or leased as property to private entities. It’s not held by the federal government under eminent domain or for military or other government use.

Public spectrum is set aside by a government for non-commercial or commercial-with-an-asterisk use to benefit everyone. The contents of its use are available and free to all citizens, and barriers to access are low or non-existent.

The contents’ purpose is education, entertainment, and/or emergency communications. The cost to consumers is free, free, and free. Even if my use of the term “public spectrum” isn’t exactly right, the VHF/UHF TV and radio broadcast bands are examples of this type of allocation whatever the terminology, and may be the last of them, which stretch back before the FCC was a thing, to the Radio Act of 1927, even as far back as Taft’s Radio Act of 1914. It is from the TV bands the incentive auctions took 40% of remaining public UHF spectrum, and auctioned it off as property to a pool of 50 bidders.

True, carriers are required to incorporate public safety messaging over their networks, under the Wireless Emergency Alert (WEA) initiative. The length and content of the message is limited both in size and type of information - text, photos and URLs, usually pure text - versus an uninterrupted stream of audiovisual information. You also need to own a cellular telephone and subscribe to service to get them. Not everyone owns a TV or radio, but those devices require no paid subscription for content, and are arguably easier to share than a cell phone.

Carriers are given grey lines on how to adhere to WEAs, but I can’t be sure about that, since the website for WEAs is vanished.

2) Gross and disproportionate handouts to a small number of broadcast licensees who pocketed what is as close to free money as it comes, versus the disappointing sum deposited in the Treasury.

Of the $19.6 billion raised, $10 billion is paid to participating broadcasters. This is a staggering ratio when you consider the Treasury - the bank for America’s 320 million citizens - is getting $4 billion less than 175 broadcasters. Many of the biggest payouts were swept off to holding companies or conglomerates like Sinclair Broadcasting, Pearl, NBC, FOX, etc.

Small commercial (and some public and educational) broadcasters participated too, but in all cases the check goes to the licensee(s). It is not distributed among the station’s employees, but to whomever or whatever legally holds the broadcast license. These licenses were originally awarded or assigned on a merit or lottery basis decades ago, free of charge.

Given only 175 stations were bought during the reverse auction at a much reduced clearing target, that $10 billion will be enjoyed by a privileged few.

When I first began wrapping my head around the incentive auction (it continues), four years ago, in order to explain their ramifications to the community of UHF wireless audio device users I used to blog for, I read an illuminating editorial by Harold Feld of Public Knowledge. Illuminating because of the clarity of his description of the auction, and the uniqueness of his position - that the incentive auction design as revealed was not a good thing.

I went through stages of alignment with Feld’s position over the years. Though I still don’t agree with all his ideas, I am indebted to Feld, who stands in the margins of telecommunications analysts, and against the grain.

3) The limited utility for the mobile ecosystem of adding 70 MHz of 600 MHz spectrum to the portfolios of only two mobile carriers, with T-Mobile taking 45% of licenses nationwide

All mobile carriers require spectrum to deliver service to customers.

Since radio spectrum of long, moderate, short, and very short wavelengths (inversely proportional to frequency) exhibit different physical behaviors, in order to remain competitive mobile carriers must hold strategically diversified “spectrum portfolios.” Different wavelengths are more or less useful for different applications, end-users, and operating environments, and the value of one wavelength or another changes over time according to handset and antenna technology, market forces, and whimsical beliefs.

In 2017, spectrum between 600-699 MHz is considered “low-band.” In the cellular/mobile industry it doesn’t get much lower than 600. Any lower, and the noise floor goes up, there are these crazies called HAMs whispering conspiracy theories on obsolete radios, and handsets get ridiculously large with clownish antennas. The wavelength of 650 MHz is about 18 inches.

Although someone got to calling UHF spectrum “beachfront spectrum,” by today’s standards and tomorrow’s predictions, 600 MHz is pretty lousy stuff for the cutting edge.

First, carriers cannot use 600 for the industry’s number one, two, and three priority today through 2020: small cell densification. Even at low powers, it’s hard to densify a network with small cells on low-band spectrum without multipath and intersymbol interference.

Second, the transitional bandplan for 600 MHz offers a 35 MHz uplink and downlink block for a total duplex pair bandwidth of 70 MHz. There is an unbreakable upper ceiling and correlation between the amount of available bandwidth and the amount of capacity - or information density - a network can modulate onto radio waves. For comparison, the 2.4 GHz ISM band, which is unlicensed and free-with-another-asterisk to use, has about 86 MHz of bandwidth depending on region and channelization.

So why did T-Mobile shell out $8 billion for 70 MHz of 600 again?

The simple answer is competition. The complicated answer is coverage (vs. capacity). Which is really about competition, too.

Prior to the auction, T-Mobile’s spectrum portfolio lacked low-band spectrum.

Low-band spectrum is not useless, not at all, it’s just not very useful for the “flexible mixed-use mobile broadband” and “5G forward” use-cases of the next generation of mobile connectivity in modern cities and suburbs the FCC brands it as. Unless you don’t have any low-band.

Large wavelength/low frequency spectrum is great for covering large areas of heterogenous terrain. It penetrates walls and windows with ease. This makes it valuable for both rural coverage, and urban coverage, but largely at the macro-cell scale. 600 and 700 MHz is able to cover more geographic area with the same amount of energy, using fewer base-stations. It’s more cost-effective in some situations, especially if the capacity demand on the network is low. If you don’t have low-band spectrum you cannot profitably reach rural customers, and you cannot easily handoff or maintain transmissions to/from a handset if the user goes indoors. If you are a mobile carrier and you don’t have any low-band spectrum, you can still provide spectacular quality of service to users, but in today’s monopolistic marketplace, where there are four and only four carriers, if any one of those carriers does not have a full spectrum portfolio including low-band they are at a disadvantage.

T-Mobile did not have much low-band, so they placed intelligent bets on 600. Every credible source I’ve heard from believes their investment will pay off.

T-Mobi - oh I’m sorry, why do I keep saying T-Mobile?

I should offer a more accurate portrait. I mean T-Mobile US, Inc, of which German behemoth Deutsche Telekom is controlling shareholder.

That’s right. All of those sexy models kicking and punching at nothing in the air while bold pink serif typeface flashes gigabyte offers across the screen?

German.

T-Mobile US, Inc. is its own entity, but it was Deutsche Telekom behind the incentive auction paddle, and it is Deutsche Telekom that enjoys a 9% discount on the Euro exchange rate when they shovel $7.993 billion into the FCC holding tank.

There are plenty of analysts out there who say “T-Mobile needs it [low-band] to survive.” There are others who see T-mobile-DT’s $8 billion exchange as a strategic bet to make T-Mobile US, Inc, a more desirable candidate for a mega-merger.

Of late, mega-mergers have stormed back into vogue.

Calculated strategy with the aim of mega-mergers may explain the opaque and expensive bids of Dish Networks, which bought almost as much 600 as T-Mobile-DT, for no obvious reason.

Tara Lachapelle of Bloomberg wrote an excellent article which asks the question “What in the world is Charlie Ergen thinking?” Lachapelle’s answer, to Ergen, CEO of Dish Networks:

>“Dish's stock took a beating earlier in the week amid investors' bewilderment. Some interpret the auction results as a sign that Dish may not be able to sell or lease the spectrum at a premium later on… But careful about betting against Ergen... I don't think Dish wants to build a wireless network so that it can go it alone, despite the company's interest in the possibilities of the "internet of things" (as detailed in a March filing). Let's be real, Ergen is building Dish for a major transaction. It's why he's hoarding valuable spectrum.”

Lachapelle also mentions the mobile carriers are setting their sights on 5G.

Yes, they are.

What Verizon and Sprint (and ATT in fact) did not pay for spectrum in the incentive auction, they doubled down on next generation R&D and infrastructure.

A few days after the results of the auction were released, Verizon announced a $1 billion supplier deal with Corning for 12 million miles of fiber optic cable annually for their “Fiber One Initiative,” which has been criticized. Around the same time, Verizon started a bidding war with ATT for Straight Path Communications which continues to rage. From Straightpath’s website:

>“Straight Path Communications is a forward thinking communications asset company. We specialize in maximizing the value of assets such as millimeter bandwidth licenses for 5G networks (Spectrum)”

The licenses of interest are contiguous 39 GHz millimeter wave licenses, which Straight Path may have bought in Auction 30 way back in the year 2000. Straight Path is not listed on the winning bidder summary. STPH may have bid under a pseudonym shell LLC (a common practice), or, hoovered up Winstar Communications which won most of the 39 GHz licenses, and went bankrupt in 2002. Even in 2002, think about it, 3G networks were all the rage, I was thirteen years old, the iPhone did not exist. In 2002, it was commercially impossible to create networks on frequencies that high, because no manufacturers made 39 GHz cellular hardware of any kind. Radio’s above 30 GHz were exceedingly rare, and non-commercial. They were found only in a handful of research laboratories, SATCOM, and the military complex. Even today, 39 GHz is very difficult to build cost-effective radios, processors, filters, and antennas for. Electrons become erratic and quantum electrodynamic-y when you push a waveform with a width of several millimeters into a radiating element to transform them into information carrying EM waves. Some of them disappear into another dimension, or whatever. Because of these physical hurdles, before 2016 39 GHz spectrum was had for a song - pennies on the dollar. Everyone knew mmWave could do something, for someone, someday, but it was only very recently that 1) engineers argued that mmWave spectrum is of commercial value, 2) the FCC proposed specific rules on mmWave to unlock commercial opportunity for 5G and other next generation wireless services.

Less than a year later, the two largest carriers are pushing Straight Path’s price up past a billion USD largely for mmWave spectrum that hasn’t been fully field tested for real, large networks.

T-Mobile has 39 GHz spectrum too. But they (DT I mean) just spent $8 billion on low-band to flesh out their portfolio. That’s $8 billion that won’t be going somewhere else, like 5G R&D, or rapid base-station roll-out on 600. T-Mobile isn’t stupid or slow, but their market-share and ~3 year battle to get ahold of low-band may mean they’re playing an increasingly hopeless game of catch-up in an industry that moves at slightly less than the speed of light in a vacuum.

ATT and Verizon together have 200 million customers in the U.S. The U.S. has a population north of 320 million people. Sprint and T-Mobile scrounge for the other ⅓. Both ATT and Verizon spent wads acquiring spectrum in the 700 MHz band almost ten years ago in 2008, and thus have had ten more years to play with low-band than T-Mobile.

The difference in propagation and capacity between 700 and 600 bands are insignificant. Antennas are a little bit shorter, maybe 700 travels less far through air and penetrates architectural material with less ease, but the difference is slight. We’re comparing apples to granny-smith apples. It’s UHF spectrum vs slightly lower UHF spectrum. There are no statistical destructive fading or intersymbol interference patterns in 600 that don’t exist in 700, or difficulty bouncing around inside the characteristic spatial distribution of foliated trees. There are no heavy spectral attenuation lines in either band. About the same quantity of spectrum and therefore available bandwidth exists in a 600 MHz sector as 700 MHz.

If Verizon and ATT want Straight Path so desperately for its portfolio of 39 GHz, we have at long last arrived at the final disgrace and mystery of the incentive auction outcome: mysterious bidders who showed up and disappeared with no intention of using the spectrum for tellecommunication whatsoever.

Who were they? What were they?

It’s very simple.

In the case of “Channel 51 License Co, LLC,” also known as “Channel 51, LLC,” the fifth largest spender at $858,704,549, their identity and ownership is explained by this chart:

Wait. Here we go:

There are two more scenarios in this deck, which I understand nothing about, except that “Channel 51, LLC” is an instrumental spore of the Australian Government Future Fund, a sovereign wealth fund managed by financial specialists placing investments for the benefit of the Australian people. http://www.futurefund.gov.au/investment To bid in the U.S. incentive auction the AFF pulled levers through Future Fund Investment Company No.5 Pty Ltd, formed to invest in US markets. The U.S. has sovereign wealth funds too, but there’s something offputting about a different country purchasing U.S. spectrum real estate it has no intention to use for telecommunications in the United States. I apologize to all of my Australian friends. Yes, your economy is gangbusters right now. This is still weird.

What about sixth place, Northwood Ventures? A private investment fund.

Northwood bid as “TSTAR 600,” and in a report on Northwood’s site they get to the point:

>“TSTAR 600, Jackson, MS, was formed to bid on and acquire radio frequency spectrum licenses in the 600 MHz band to be auctioned in the Forward Auction by the FCC in 2017.”

This is radical transparency compared to rampant smokescreens shielding identity and responsibility of the actors we see at last before us at the end of this long and complicated techno-econo-train.

If only everything - or even something - about the incentive auction was so transparently revealed, we might have stopped it from going off the rails, in the well-intentioned but ultimately drowned spirit of being the world’s “first.”

0 notes