Spectrum Spaces

Scientific American: "Microsoft, Google and several more of the world’s largest and most influential technology companies have found a way to provide wireless Internet access that is so fast it makes today’s Wi-Fi networks seem as sluggish as dial-up service. The prospect, however, has big media broadcasters up in arms, because this blazing-fast network access may hamper television signals when they go digital next year."

The trouble, basically, is that this wireless internet service uses the same portion of the spectrum on which TV stations will be broadcasting digital signals. We could resolve the problem if only there were some set of human institutions that allow scarce resources to come into the possession of those capable of deploying them in the most high-value way. Like, say, an auction rather than just giving the space away to TV broadcasters. But that's crazy talk.

Comments

We could resolve the problem if only there were some set of human institutions that allow scarce resources to come into the possession of those capable of deploying them in the most high-value way. Like, say, an auction rather than just giving the space away to TV broadcasters.

An auction obviously won't necessarily do that. The highest bidder may simply be the company with the deepest pockets, not the one capable of exploiting the resource in the most efficient way.

Posted by Mixner | June 16, 2008 3:22 PM

An auction obviously won't necessarily do that. The highest bidder may simply be the company with the deepest pockets, not the one capable of exploiting the resource in the most efficient way.

And how are you planning to define efficiency without reference to market pricing? Absent some major market failure (e.g., externalities due to pollution), allocating in a market based system has a lot of appeal.

Posted by richard | June 16, 2008 3:43 PM

Mixner is right, although letting the deep pockets win makes more money for the Treasury, since an FCC license is a franchise on a government owned monopoly (i.e. electromagnetic spectrum), the highest value for the license holder would be to to use that monopoly power to be the troll at the foot the bridge and extort rent from anyone who tries to pass by (kind of like patent law).

This complaint is really about broadcasters wanting to be paid off first. In reality, interference is a very common problem in the radio field but it is easily dealt with both legally and technically.

I understand that broadcasters got this no-auction license deal because the FCC is taking back their analog broadcast frequencies. I don't think legally you can claim a property interest in a government license (though if you look at how some states treat billboard zoning permits, I'm afraid that's an open question), but politically there's no way Congress was going to take billions of dollars worth of licenses away from broadcasters without compensating them in some way.

In the old days before auctions, broadcast licenses were handed out to the politically connected. Exhibit 1, KLBJ in Austin, until recently owned by one Lady Bird Johnson.

Posted by beowulf | June 16, 2008 3:44 PM

I've never understood why it made sense to broadcast digital TV to begin with, in an era where something like 88% of households have cable or satellite TV. (And I say this as part of the recalcitrant 12%.) Putting HDTV on the broadcast airwaves is like putting electric headlights and turn signals on a horse-drawn carriage.

Hell, why not shut down broadcast TV entirely in the those areas of the country where cable TV penetration is essentially 100%, and require that cable companies give everyone free access to the 'broadcast' channels for the next 15 years?

That would open up tons of former broadcast spectrum for a ton of alternative uses.

Posted by low-tech cyclist | June 16, 2008 3:54 PM

And how are you planning to define efficiency without reference to market pricing?

Why, by government decree, of course. Because history shows that governments are so much better at figuring out what's valuable to people than people themselves.

Posted by Mixner | June 16, 2008 4:00 PM

"Hell, why not shut down broadcast TV entirely in the those areas of the country where cable TV penetration is essentially 100%, and require that cable companies give everyone free access to the 'broadcast' channels for the next 15 years?"

Two reasons;

1. No such area exists. There are urban households with no cable connections, so they would have to be built.

2. You put companies that serve urban areas at a competitive disadvantage compared to their rural (incomplete penetrance) counterparts.

Posted by Joel | June 16, 2008 4:02 PM

Does economics have a different definition of efficiency than physical chemists do (or for that matter just about anyone else?)?

I can define efficiency without respect to market pricing: efficiency = how much energy I have to put in to produce enough in order to get a certain share of production in return (energy in / work out).

Of course, I measure my production (and what share of production I'm paying for) in terms of market prices. But the price is just how the market agrees, after a lot of haggling, negotiating, etc., on how to measure production, ain't it?

Isn't saying "you can't define efficiency without reference to market pricing" the same as saying (in the context of talking about Carnot engines): "you can't define efficiency without reference to joules or calories"? Or am I missing something?

Posted by DAS | June 16, 2008 4:22 PM

Fortunately, Coase says this isn't a problem. So that's a relief.

Posted by Tyrone Slothrop | June 16, 2008 4:33 PM

DAS,

It's efficiency in exploiting the resource to satisfy demand.

Posted by Mixner | June 16, 2008 4:35 PM

Be nice if the link worked, though. I'd like to read the article. Anyone have the correct one?

Posted by DrBB | June 16, 2008 4:39 PM

Okay, so now it works. Looks like the changed the article's name and just set up a redirect.

Posted by DrBB | June 16, 2008 4:44 PM

It's efficiency in exploiting the resource to satisfy demand. - Mixner

How does one quantify that? I guess there's a reason why I ended up in bio-physical chemistry and not econs ...

Posted by DAS | June 16, 2008 4:50 PM

The highest bidder may simply be the company with the deepest pockets, not the one capable of exploiting the resource in the most efficient way.

Heck, I'd say that the companies least able to exploit the spectrum already have control over it. Anyway, it's not like GOOG and MSFT are short on cash. They win the auction, they'll be able to invest in the R&D and whatnot. Same goes for NBC, CBS, and ABC.

Posted by daveNYC | June 16, 2008 5:18 PM

How does one quantify that?

Profitability.

Posted by Mixner | June 16, 2008 5:22 PM

Profitability. - Mixner

Again -- this is why I guess I'm no economist. I hear/read "efficiency" and I think "work out / energy in" (I think I got it backward above ... oopsies!). I hear profitability and I think "if I paid XYZ Corp $10 for the product but they paid their labor, management, et al., only $9 in total, that extra $1 is 'energy in' for which I'm not getting 'work out'" and hence I think "inefficiency"!

Economics sure are strange! ;)

Posted by DAS | June 16, 2008 5:37 PM

DAS,

I really don't think it's hard to understand. $10 represents the value of the product to the buyer. A company that can make the product for $8 is more efficient at satisfying the demand for the product than a company that has to spend $9 to make the product. The additional profit that the first company makes compared to the second company is a measure of that greater efficiency.

Posted by Mixner | June 16, 2008 5:54 PM

Joel, the argument is not crazy. At 12% of households without access to cable or satellite, we are talking maybe 9 million homes. Compensate Time Warner/Direct TV one year's worth of service (about $360) for each new line installed and you've covered just about every pre-existing gap. That totals out to a bit over $3 billion or about 15% of what the government raised in the last spectrum auction. Digital TV is a horrible waste of that bandwidth and will get killed off eventually when we realize how valuable that "real estate" actually is...

Posted by dan k | June 16, 2008 6:10 PM

What if I value the product at $9 but nobody offers it to me at that price? Of course, if everyone were like me, perhaps the supply/demand equilibrium would occur at $9, but it might take a while for the price to reach equilibrium, for example.

So, since I do want/need the product, I'll simply suck it up and just buy 9/10's of the product (supposing I'm a Bayesian utility maximizer). Actually, this could be a stable equilibrium price then of $10 even if I (and all like me) only ascribe a "value" of $9 to the product.

If the company can make the product for $8 as opposed to $9, that means the "parts" of the company are more efficient. But the process as a whole is of equal efficiency as I am getting the same amount of product for the $10 (ultimately some fraction of my labor) of energy I've put in -- the company itself, taking more profit, is less efficient, isn't it?

The reason why, using a more physical chemist friendly language, for-profit industry often is able to produce products with more "component-wise efficiency" than non-profit industries is that, given stiff-enough competition of suppliers and an elastic enough demand curve, the "total energy available" for production is pretty much a constant, so the desire of a company to maximize its total profit means that it must be able to have components (workers, processes, etc.) that run as efficiently as possible so that way at the end it can skim off as much energy as possible in terms of profits.

I guess the key question when talking about "efficiency" is "efficiency of what?"

Still, it's enough to make this biophysical chemist's mind dizzy -- as if trying to figure out why I keep getting negative numbers from what should be an unbiased estimator of a positive quantity is not making me dizzy already!

Posted by DAS | June 16, 2008 6:17 PM

If the company can make the product for $8 as opposed to $9, that means the "parts" of the company are more efficient. But the process as a whole is of equal efficiency as I am getting the same amount of product for the $10 (ultimately some fraction of my labor) of energy I've put in -- the company itself, taking more profit, is less efficient, isn't it?

No. The efficiency in question is the efficiency of the supplier at satisfying demand. If he can satisfy it for $8 rather than $9, he's more efficient. His additional profit is a measure of that higher efficiency.

Anyway, I'm done explaining it. If you still don't understand maybe someone else will take a shot.

Posted by Mixner | June 16, 2008 6:27 PM

If he can satisfy it for $8 rather than $9, he's more efficient. His additional profit is a measure of that higher efficiency.

Um, that's more like productivity. Company A can produce a widget for $8 while Company B produces it for $10. You say efficiency and I start thinking efficient markets, which is a very different critter.

Posted by daveNYC | June 16, 2008 6:45 PM

Anyway, I'm done explaining it. If you still don't understand maybe someone else will take a shot. - Mixner

Thanks to your kind and patient explanation, I understand it very well now -- I'm just in a dizzy tizzy (as if reading papers on unbiased estimators relating to the non-central x2 distribution isn't tizzifying enough ;) ) 'cause the use of language jars so much with my thermo-based understanding of the language. I guess it's a "two cultures" sorta thing.

Posted by DAS | June 16, 2008 7:07 PM

davenyc,

Productivity measures the efficiency of production, not the efficiency of satisfying demand. I may be very efficient at manufacturing, say, 8-track tapes, but if there's virtually no demand for them I haven't created much of value.

Posted by Mixner | June 16, 2008 7:19 PM

Your opinion of Obama's transportation policies.

Posted by Frank | June 16, 2008 7:37 PM

@Mixner

You got any linkage on that? My Google-fu turned up nothing.

Posted by daveNYC | June 16, 2008 8:39 PM

dan k - actually, it's even better than that. That 12% includes people (such as yours truly) who have cable TV wires running into their homes, but choose not to subscribe.

In cities and inner suburbs, when cable companies were first negotiating with cities for the right to a local monopoly, almost every city negotiated universal access as part of the deal. (You don't want to piss off some of your constituents who didn't get the same goodies as everyone else.)

In the outer suburbs and exurbs, penetration is essentially total: developers just don't build new neighborhoods without the cable company's installing cable as part of the construction plan. And it's been that way for the past 25 years.

So in built-up areas, the number of households without access to cable is really quite small, and your solution would not only work, but be far cheaper than you allowed for.

Posted by low-tech cyclist | June 16, 2008 9:19 PM

"An auction obviously won't necessarily do that. The highest bidder may simply be the company with the deepest pockets, not the one capable of exploiting the resource in the most efficient way."

This is part of what financial markets are for. And so it basically becomes a question of who is going to be better at identifying the likely highest-value user--the financial markets, or the FCC? And I'm betting on the financial markets, even if the FCC was doing its best to pick the highest value user.

Of course the FCC isn't really doing that anyway. What the FCC really wants is to exercise control over media content, but frequently what they want to do runs afoul of the First Amendment. However, Supreme Court doctrine as it currently stands basically carves out an exception to the ordinary First Amendment rules for broadcast content (the key case being Pacifica, one of my all time least favorite Supreme Court cases still in effect). And Pacifica provides a back door into at least some control of non-broadcast content, to the extent often the same content is also being broadcast.

So, the FCC is hell bent on preserving broadcast media content even if its remaining viewership is relatively small, since otherwise media content will pass entirely outside of the Pacifica exception. And that has absolutely nothing to do with figuring out the most efficient users of spectrum.

Posted by DTM | June 16, 2008 9:59 PM

This is part of what financial markets are for. And so it basically becomes a question of who is going to be better at identifying the likely highest-value user--the financial markets, or the FCC? And I'm betting on the financial markets, even if the FCC was doing its best to pick the highest value user.

The FCC isn't "identifying the likely highest-value user." It's just holding an auction. The space goes to the highest bidder, regardless of what the FCC thinks about him. And while you might hope that a small company with a good idea would be able to secure sufficient investment to outbid a Google or a Microsoft with a worse idea, there's a good chance that won't happen.

Posted by Mixner | June 16, 2008 10:14 PM

Mixner,

Actually, the FCC allocated some of the former analog TV spectrum to digital TV broadcasters without an auction, and is only auctioning the remainder. That unauctioned spectrum allocated to digital broadcasters is what Matt was referring to in the quote you selected for comment. If the FCC had instead auctioned all the former analog TV spectrum, I would have no objection.

Posted by DTM | June 16, 2008 10:55 PM

In the outer suburbs and exurbs, penetration is essentially total: developers just don't build new neighborhoods without the cable company's installing cable as part of the construction plan. And it's been that way for the past 25 years. - low-tech cyclist

My parents moved out to an exurb to get away from coastal humidity and smog. There is no cable where they are -- everybody has satellite, though, so I guess it's a moot point. But still, it's not the case that all new exurban development has cable access: at least in the So.Cal. desert, AFAIK, many do not.

Of course the larger issue with digital TV is the lack of availability of conversion technology (and the need for TV in certain areas given that local TV generally still exists whilst many "local" radio stations are owned by big networks that shirk their duty to provide emergency announcements, etc.) as Harry Shearer has been shouting about (but nobody else seems to have noticed). And what for? For what do we really need digital TV?

Posted by DAS | June 17, 2008 9:55 AM

"The prospect, however, has big media broadcasters up in arms, because this blazing-fast network access may hamper television signals when they go digital next year."

it interferes with broadcast TV signals?

That's a feature, not a bug.

Posted by Sock Puppet of the great Satan | June 17, 2008 10:42 AM