The Mission of the Manhattan Institute is
to develop and disseminate new ideas that
foster greater economic choice and
individual responsibility.

Event Transcript
February 13, 2003


Spectrum Policy Reform in the UK and the USA, continued

[START TAPE 2 SIDE 3]

-- closer to that, then obviously the better the signals would be in the public sector and the better things might work out. Thank you.

MR. CHRIS SUELLENTROP: We have time for questions. Bring the mike down, Jim. Go ahead and identify yourself and ask your question.

MR. JIM SNYDER:  Jim Snyder from New America Foundation. I remember I read your report in March of last year. I had two questions. One, there were numerous references in the report to setting spectrum fees to give licensees incumbents to use their fees efficiently, and it kept being I think references to coming up with some type of market-based way of setting the fee, that you referred the reader to OffComm any number of times which I checked and it did exist. The key question is well, how actually do you set market-based spectrum fees by, I guess, using impunitive [phonetic] fees? Thereís no discussion in your lengthy report as to how actually to do that, which would suggest that thereíd be all sorts of room for gaining, politics. So how do you solve this difficult economic problem of setting those rates on a market base. So thatís one question.

The second one --

MR. MARTIN CAVE: I can tell you how to formulate it. I donít know if I can tell you how to solve it.

MR. JIM SNYDER:  Okay. Because that seems like a million dollar question.

MR. MARTIN CAVE: Yeah.

MR. JIM SNYDER:  All this talk about fees and then not have a methodology to set them thatís clear and not subject to an awful lot of political manipulation I think is an important question.

The second question is, to what extent were your recommendations regarding broadcasting dictated by the power of the broadcasters in England? How powerful are the broadcasters politically and how does that maybe distort your recommendations, and maybe Tom there has been a big advocate of the Negroponte switch which I think has been best manifested in the Berlin transition -- what do you think of the Berlin model for transitioning broadcasters from their spectrum?

MR. MARVIN CAVE: Well, let me deal with the second question first and then perhaps we might manage to bury the first one. The broadcasters are powerful in the U.K. framework, but the public service broadcasters in particular are powerful. Because itís a strong element of government policy reflected in the communications bill, that we should have more rather than less public service broadcasting. Now, thatís a bit of a puzzle. Because normally youíd think a multichannel world, lots of choice, lots of variety, the need to correct for market failures would be limited and therefore we could have less public service broadcasting. But in fact itís quite clear, you know, on the face of the communications bill, that the government wants to expand the provision of suitable programming by public service broadcasting rather than contract it. So in a sense, theyíve been on a roll. Because on the back of that desire to enhance the cultural education level of the British people, or whatever the reason is, on the back of that theyíve been able to say well, itís absurd for you to say have more of this stuff than the BBC, and at the same time charge us for spectrum. And so thatís really been the issue.

And within the government, the sort of noncultural lobby has said that the spectrums being used clearly has an alternative use; it should therefore be priced. And what we should be doing is not interfering in the input market, mainly for spectrum, but interfering in the output market by for example, giving the BBC a larger license fee or giving it access to more revenues which will then enable it to meet these services and also pay its bills.

Now, I said earlier that this issue is nested within a much bigger one. Namely, the governmentís desire to get terrestrial broadcasters off analog transmission onto digital transmission, thus releasing a reasonable, but not a gigantic amount of spectrum for other purposes. Which other purposes, incidentally, include the operation [Phonetic] of more public service broadcasting channels which you may find slightly surprising. Now, theyíre committed to doing that sometime between 2006 and 2010, but will only do so when 90% of the public have digital service available to them by some other means. And when a number of additional conditions are met. Yet the U.K. public has in fact not shown the desired appetite to acquire digital boxes which enable that condition to be satisfied. So they really are in a terrible box about this and they havenít been able to take the decisive action which other European countries, you know, Berlin for example, have been able to take in order to further that process.

There has been a suggestion that they might do it within the U.K. region by region, starting with a, you know, a less favored part of the U.K. like Scotland or something to see how it works out. But they havenít actually bitten that bullet.

As for the first question, how do you do it? Well, think of it this way: an efficient price might consist of the following three elements. Sort of a cost-saving element. You know, just as land near the townís worth more because you donít have to carry the cabbages so far, equally, spectrum, which is particularly good for some services, is worth more. Compared with some other set of spectrum that you might use.

The second element would be the sort of the rents associated with occupation of the spectrum. You know, the spectrum gives you rights within the downstream market to make excess profits. That makes the spectrum more valuable.

And then the third element might be some kind of option element. You know, having the spectrum enables, gives you flexibility, et cetera, et cetera. Now, we proposed in the report that the administrative price for spectrum should be based only on the first of those, namely, the cost saving capability associated with the use of a spectrum for particular service. And in particular, we said that it wasnít appropriate for a public service use of a spectrum to pay an additional component in the price which reflected the downstream exclusivity of within commercial markets. The reasoning for that was we didnít want to propose numbers that were frighteningly high. That could be sort of leveraged off the results at a 3G auction. Because that would have just killed the whole idea stone dead.

The other was that we hoped that as spectrum trading, I mean, in the commercial arena extended, then those rents would diminish and so the prices would come down and the element in the market price of spectrum associated with downstream exclusivity would disappear. Now, that leaves you -- and then we just dump [Phonetic] the option issue as being too complicated. That leaves you with the question, how do you calculate what I describe as the cost saving component of the opportunity cost of the U.K. [Phonetic]? Well, thereís one area where you could do that quite simply. In mobile telephony. Because youíve just got a simple tradeoff. I'm a completely ignorant person as far as these technologies are concerned, so if I say something thatís completely absurd, please donít disabuse me of it because I would find that deeply painful. Youíve got a tradeoff between base stations and spectrum. You know, if you have less spectrum you have to use more base stations. So you can therefore, by use of that tradeoff, produce some kind of implicit cost saving price for mobile spectrum. Okay?

Now, you can do it with mobile spectrum if you want to extend the use of that particular calculation for the rest of public service spectrum then you have to say something on the following lines: you know, if that spectrum werenít used for defense or whatever, whatever, whatever, it would be used for mobile communications. And therefore we should be looking at the implied price derived from the mobile calculation to the rest of the spectrum as well. Now, thatís the method that the government uses at present. Itís used since 1996. Itís obviously a helluva stretch. The argument are weak, to say the least. But it does give you some numbers. And it isnít entirely spurious. Itís about sort of 87.88% spurious, I think. But itís not entirely spurious.

But also, I think this is an important point, is itís not liable to political interference. Because once the numbers come out then those are the numbers which are used if charges are levied. So the argument isnít should the charge be a million dollars or a million-and-a-half. The argument is should the charge be positive or zero? And thatís where the battleground is. Thatís where the politics come in.

MALE VOICE:  [off mic] -- in fact, I think you may have an agency with a thousand economists and technologists, it would be as complex as the pricing for the television [off mic] pricing is ministered [off mic] I could be wrong, but I think itís going to be --

MR. MARTIN CAVE: Oh, itís --

MALE VOICE:  -- very complicated.

MALE VOICE:  Well, there was a four-and-a-half year old economist we were talking about earlier who would look forward to that [interposing]

MR. MARTIN CAVE: Yeah, by the time this process is finished sheíll be 75, I'm sure. But I mean, I think the interesting question is, you know, how to you stop doing that? How do you, in other words, manage to shrink the area to which administrative pricing applies, and expand the areas to which market pricing applies. Now, thatís something that we say in the report. You know, we hope thatís going to happen, but I didnít think that the government would tolerate proposals which made that very big step at the early stages. They were so concerned with the necessity to reserve spectrum for socially desirable uses, that making the availability of that spectrum, placing it at the mercy of market forces, run the risk of just killing the whole thing dead.

But I think you know, if given that theyíve accepted that, the next issue is transition. How do you manage to sort of leverage off the market prices to get to the prices that are set in the other really. Thatís very difficult. And itís something that will be considered very soon. But wonít be implemented for perhaps four or five years.

MR. CHRIS SUELLENTROP: Bob, you want to --

MR. BOB CRANDALL:  Bob Crandall. Sorry, I didnít--

MALE VOICE:  Might I suggest that itís hard to make progress as long as you keep referring to public service broadcasting and socially desirable uses. I realize you're a supporter of the Labor Party in the U.K., but couldnít we say something like Socialist broadcasting or something like that? [laughter] For the national peopleís radio here. [laughter]

But as to your administrative pricing, why canít we move more in the direction of allowing, whether they are socially desirable or not uses, those people now using it in public safety and national defense, to either lease out their spectrum or to sell their spectrum without any guidance from administrative pricing. Let the market take care of what prices would result and allow them to keep a substantial share of the revenues from so doing. Wouldnít that solve the problem?

MR. MARTIN CAVE: Well, we do recommend that in the report and the government accepts it. But, you know, the sixty-four thousand dollar question is will the government really allow them to keep the revenues? And my experience of finance ministries, generally speaking, is that they donít.

I mean, for example, weíve just had, about five years ago, the introduction of a public lottery in the U.K. and it was made quite clear that the profits associated with this would be used for -- entirely for incremental spending. And about three years later itís used to fund public hospitals with, you know, absolutely basic stuff. So in other words, itís just been assimilated into the public expenditure more and itís -- public revenue more, and itís very difficult to get a credible commitment. Thatís the real problem. Because you know, if you face a fiscal crisis then you just pinch the lot. I mean, thatís what governments do.

MR. CHRIS SUELLENTROP: Yeah, we have time, I think, for one more quarter. Bert? I just want to say, though, I can understand why Bob Crandall did lose his Brookings Institution parking space. [laughter] I think itís fairly evident. Bert?

MR. BERT HALPRIN:  Thank you. I'm interested in the issue of your approach to dealing with the public uses as well. This comes up again and again. And as you point out, the Ministry of Defense is forty percent of it. I mean, thatís where the big issue is. And you correctly point out that if you charge them more, but you just give them more money, itís meaningless. Now all you have is the cost of collection. So somehow you have to squeeze it. But the problem seems to me that since spectrum is leased or charged for by the government, not commercially, if you squeeze their dollars you now have hardware, software, computers, tanks manufactured by people all of whom are hiring folks saying buy this rather than spectrum. And you have a virtual guarantee, since thereís no equivalent marketing for the spectrum, that even if spectrum is overused now, that it will be underused in such a circumstance. The only way you can solve that is to say well, make all spectrum commercial, and instead of paying any fee which is set by the government, which isnít going to then, you know, have people going out and wining and dining the general saying doing it, itíll be sold. It will all be sold. What that means is that when the demand for it by defense gets higher, you know, if for example, we find out that Iraq has a nuclear weapon, the price theyíre going to have to pay is going to go up.

And one of the few things that anybody who thinks a government has a value for, itís to protect people so that they donít pay more for it at a point at which the national defense is more threatened. So even in theory itís not clear to me what possible value comes from, you know, or whether anybody has studied whether any increase in moving the balance is not going to be offset by -- and, you know, by a swing which is going to be worse in the other direction.

MR. MARTIN CAVE: Right. I think I understood the question in general rather than in particular. I think -- well, first of all, let me say that the Ministry of Defense has been subject to spectrum charges for four or five years now, and we did look and see if there appeared to be any response to those spectrum charges. And the answer basically was no. However, since the sum involved was, as I said, it probably wasnít enough to, you know, it might get to the level of a captain or a major or something, but it wouldnít get up to some senior official within the Ministry of Defense. You say, well, weíve got to do something about this. Itís real pretty small stuff. I think thatís pretty inclusive.

However, if you start with the motion which I think you're putting forward, that itís reasonable for the government to stipulate an output target for defense services, and that might change over time, you know, as the threat increases or decreases, then either you have to find a way of delivering the spectrum in order to achieve any enhanced output target, or you have to leave it to the military to sort of go out into the marketplace and buy the materials that they need in order to provide the high level of service.

Now, you know, in 99% of the cases thatís what happens. You know, if they need more communications engineers, we donít operate by conscription normally, you know, we operate by [Indiscernible] and raise the pay of communications engineers, why shouldnít that just be right across to spectrum? Well, my view is, you know, that in due course that might be a practical proposition. But I donít think that the British government or the British public are ready for it yet. And therefore, we have to mess it around or these boundaries, where the process is subject to precisely the kind of difficulties that youíve described.

MALE VOICE:  Just, thatís the second half of the question. I understand what youíre saying, that now it has no effect? What I'm saying is if you raise it to a level [inaudible] --

MR. MARTIN CAVE: Yeah.

MALE VOICE:  -- itís orphan third [Phonetic] [inaudible] produce a morbid [interposing]

MR. MARTIN CAVE: Well, okay. I think itís going to produce a more efficient use of spectrum in the following essential respect: suppose they go 40% of it. In fact, they have a bit less than that. Suppose they only use 25% of it. The rest of it is just unused. Itís kept for a rainy day. You know, why give it back? Who knows? We might need it, et cetera, et cetera. A higher spectrum charge might actually be enough to trigger them to hand it back. Now, how they would hand it back is a question that Bob raised. I think the magnitude of whatís available in excess to the defense department, Ministry of Defense, is such that just saying they could just sell it in the open market would produce revenues for them which would be quite politically unsustainable. So I donít think that thatís going to work. If theyíre going to have to hand it back theyíre going to have either to lease it because itís interruptible spectrum, that they might need to get their hands on it in the event of an emergency, or theyíre going to hand it back to the government to have it auctioned by the normal way with the government receiving the loot.

Now, in those circumstances, the presence of a non-trivial annual charge relating to that spectrum, provided that the other incentive problem that youíve described could be got around, in my opinion can actually make a contribution, even though a small contribution, to enhancing efficiency. So weíre really talking, I agree, about small gains, but I'm pretty certain that the effects will be positive rather than negative.

MR. CHRIS SUELLENTROP: And interestingly enough, the Chinese military was given some spectrum -- you may know this -- and actually built a CDV [Phonetic] network with it for civilian use, so where they understand property rights, you see, these things get done. [laughter]

Anyway, could you please join me in thanking Martin Cave for this excellent talk. [Applause]

And if our next panel could come up. Scott Woolley from Forbes magazine is going to be the moderator for this next session. Scottís been with Forbes for about six years and he spent some time at the Kennedy School before that. Did time, some people would say. And for the last three years has been covering communications issues, and in fact, had a cover story that many of you saw this last fall that was probably the most prominent feature in an American magazine of the spectrum policy, spectrum allocation issues. And a very fine piece. Scott?

MR. SCOTT WOOLLEY: Thank you. Just on the last point, on the defense spectrum, my favorite example recently was I saw the U.S. militaryís new effort to make their ships be able to be controlled wirelessly, so if the captainís not on the bridge, that he can control them. And the wireless technology they chose to use was WIFI [Phonetic], which I thought was surprising, given the amount of spectrum they have, that they would go with allegedly unsecured and unlicensed technology.

But moving on, in writing about this itís been very hard, first of all, to convince people that itís important, but itís also been hard to convince people that thereís something wrong with the wireless industry. Because I think most people, when they look at what wireless technology can offer them today that they didnít have two or three or ten years ago, they look at the pace and itís amazing to most people. But that said, there is this incredible potential that is untapped if this spectrum could be freed up. And I think itís created this mass of what nobody can really define, but most technologists know could exist, has built up to the point that we finally do have a constituency in Washington and around the country, coming to Washington to complain. And finally, we have a counter-balance to the political force thatís always existed of the incumbents who are very happy keeping the public out of the loop and confused as to what spectrum is and why itís important.

So that said, itís finally this spectrum report to me has some potential for going somewhere in this city. So with that said, I'll turn it over to our panelists. They range from an economist trained at Berkeley to a Stanford professor [laughter] and I guess weíll start out -- I guess weíll start with Berkeley. Coleman Bazelon is a vice-president, the Analysis Group/Economics. He was also at the time [Phonetic] at the CBO and while there analyzed the FCC auction process. So he should have some good perspectives on this.

MR. COLEMAN BAZELON: Thanks. Letís see if I can get my show up here. I should start by saying that I'm an economist, not a lawyer or an engineer. And thatís relevant, because I'm going to talk about interference issues, and I guess my take on it is that if the economists had listened to the lawyers about allocation issues, nothing would have happened, so I donít -- and weíve had some success there. So I'm ready to take on the engineers.

And I'm going to talk about some interference issues in the Spectrum Policy Task Force Report, and Greg is going to cover the allocation side of it.

So the bottom line for me is that the approach in the report taken to interference management is incompatible with the approach taken to allocation. I'm going to try and bring this, flesh this idea out a little bit in the coming slides. But if they take the approach that theyíre talking about to interference management it will thwart any chance at having market allocation of services in spectrum.

Just to review real quickly or preview, perhaps, on the allocation side, for services the current or the historical model is at the bottom, the command and control. And the FCC is trying to move away from that. The report suggests two parallel models, commons and exclusive use or property rights. And they recognize on the allocation side that the centralized command and control of how spectrum is used and of the services itís used for is not meeting the needs of spectrum users. And that we need to have a more decentralized and market allocation.

So let me quickly review what the report has recommended. There is also, you might not know, thereís a working group report, a working group on interference report that fleshes out the ideas a bit more.

The first thing we want to do is regulate receivers. The reason is obvious because poorly designed receivers canít tolerate interference and lead to inefficient use of the spectrum. So better receiver standards will allow for more efficient use of the spectrum.

Likewise, on transmitters, smarter transmitters will be able to economize more in the spectrum thatís used.

They think they should better define and regulate, and communicate the regulations on interference. Thereís a lot of discussion about how the terms are now well-defined. Whatís harmful interference, whatís acceptable interference. And that by harmonizing these definitions and then publicizing them more, theyíll be able to better regulate and manage interference.

Create neighborhoods of similar users. This is a pretty good idea. Itís already done to an extent in the sense that we donít mix fixed wireless links with mobile users much. But I guess the idea here is to broaden that idea of grouping similar types of uses in the same general area of the spectrum because the presumption is that theyíre going to be better able to manage interference between the licensees if the use is somewhat similar.

The most radical, or the biggest suggestion I think thatís getting the best talk is this idea of an interference temperature. This is sensible sounding. It says thereís some -- all radio systems have to accept some level of interference. Itís impossible to be completely interference-free. So letís define what that acceptable level is for each allocation. And then any uses that come in underneath that will be allowed. And the key here is to define what that acceptable level is so that you donít have to judge on a case-by-case basis about a new low power, usually a low power service thatíll come in underneath it. This fits in with the underlay idea, that you could imagine a lot of low power devices filling in that space below this interference temperature.

And another idea, which I donít think got as much attention perhaps is this idea they want to measure the interference temperature. Itís one thing to define it, but you donít know if itís being exceeded, you donít know how much dead space there is underneath it that can be used. And the interference working group proposes what could be a rather large project, which is go out across the country, across many bands and monitor and create a database of actual spectrum usage. Clearly, if you're going to manage interference you need to know whatís out there, so theyíre suggesting that we go out there and measure it. Although, like I say, it might be quite a large task.

I guess the problem with this is that the mindset of it is of the old allocation regime of command and control. That their recommendation, this is from the working group report -- this is the paragraph suggesting the interference temperature says that the commission should consider this. That itís a tool for the commission to be able to manage things better to set this. Not that a licensee should think about what its acceptable interference temperature is and what it can tolerate, but that this is a decision for the commission.

And then what I think really belies the mindset is that in putting the cart before the horse they suggest that the transition from analog to digital should be promoted by the commission, even by rule should be forced by the commission. Now, I think we all agree that in general a digital technologyís more efficient than an analog technology, and that you would expect to see a migration from analog to digital. But the commissionís purpose here, or the working groupís purpose here is, because itís easier to measure digital interference. Itís easier to predict digital interference. So they want the transition, not because itís in itself better technology, but because itís easier to manage it from the center.

So these are quotes from the working group report. Interference protection is related to rights and obligations. I think thatís absolutely right and we all agree. I give them an A on that. But then they go on to saying that if itís too lax you could lose some services and if the protections are too tight youíll prevent new services. And this is somewhat if the porridge is too hot or too cold. And theyíre still recommending that itís the commissionís job to find out just what that right level of interference is. And I think thatís, as I'm going to argue, a process that just wonít work. So, sorry.

I want to give an example of interaction between interference and allocations. This is the band that Gregg Skall talked about earlier. Early out [Phonetic] there was three commenters [Phonetic] and I think three companies that were interested in this band in particular. And their Astro [Phonetic], I'm not exactly sure what their service is. Maybe thatís the satellite location service. They have their own idea about technical standards. And again, I'm not an engineer so donít ask me to really take ownership of these numbers. Theyíre just pulled out of the report. The point is to show that theyíre different by each petitioner. ArrayComm has a different service with different peak output and in band, out of band emission limits. And what used to be called Microtracks [Phonetic] has its own idea about how itís used. Now, we heard that the commission decided not to set a standard, but the fact that the people who wanted to bid on these bands of spectrum were in there arguing about what the tactical standards should be and that those technical standards are tailored to the service allocation they want, had the commission gone ahead and in other areas they do, that would have ended up, the technical rules or interference rules would have ended up setting a lot of the allocation service decisions. And also, the FCC in the notice they put out said that they asked people about whether or not they should consider banning cellular architecture which of course, has a huge implication on service that can be provided in a band. This is a technique that was used in the guard band [Phonetic] auctions to prevent commercial mobile phone service from being allowed to be provided and theyíre, rather than outlawing it explicitly, they just said you canít have cellular architecture and that eliminated a whole lot of potential uses [users?] that there was an incentive to eliminate.

So I have at least an idea for an alternative approach that I'd like to put forth. And thatís that we have administrative law judges who are knowledgeable in spectrum issues, adjudicate interference disputes. This reminds me of special masters in water law. In a previous life I did work on California water policy, which is very much like radio spectrum. And, you know, special masters would be assigned to adjudicate the rights of groundwater reserves. There would be principles they use in adjudicating them, but nonetheless, you have highly specialized administrators or judges making these decisions. And the idea is that it takes it out of the political process.

The commission, without having to get into the congressional arena, is free to delegate to the administrative law judges decision-making power like this so long as you can appeal the decision of the law judges to the full commission. And here is where the hard part comes in. The commission would have to resist the temptation to hear the appeals from these law judges, or at least in most cases. And if they did that over time how interference is resolved between parties would -- rules would develop out of practice that would become predictable and understandable and the parties could then know what to expect if they had a dispute. Much like nuisance and land issues might develop. I'll admit up front that this will be a difficult and expensive start way to start off, but after this has been working for a little  while the rules would become more predictable and would be, I think, out of the political arena which would be the key part.

Take the analogy back to water, thereís, you know, the state will set what the water law is for a state. In the west prior appropriation is the rule which basically says whoever uses it first gets to keep using it. And in the east there was a reparian [Phonetic] law which basically says anybody who has riverfront property gets a piece of the river. Even if the way they want to use it comes in later, other users have to readjust their uses and share. How itís implemented though can be difficult. Different states would implement it different ways. Even at the reparian level sometimes it was measured by how much land you owned that touched the river; sometimes itís measured by how much linearly the riverfront of your land is. But the commission and policymakers would still have a role to play in setting the policy about how interference would be resolved. They just wouldnít do the actual details of resolving it. So they could set a priority [Phonetic] in use versus a commons approach to a band, or they could set a past practices versus a most efficient use. You could imagine some people argue that maybe more for political reasons than anything else, that analog television needs to be protected so the commission in its wisdom could say that past practices would define that band. That would give the incumbents complete protection. On the other hand, they could say that an efficient use or the efficiency of a use can be considered in whether or not it should be accepted.

So what I want to end on is that -- letís try to get this out of the political realm and into a judicial realm. The idea to me of two licensees -- of somebody owning a license and trying to trade it to another use or themselves putting it to another use, and an adjacent licensee saying no, you canít do it because it interferes with me, and the two parties then have to go to the commission and open a proceeding and, you know, I'll get rich off of that, but I donít think itís going to be good for spectrum management. Getting it out of that process to define these interference rights I think is going to be absolutely key. Thanks.

MR. SCOTT WOOLLEY: Our next panelist has seen the task of making FCC spectrum policy from the inside when he was a deputy chief economist there in the mid-nineties. Greg Rosston then moved on to Stanford where he teaches economics now.

MR. GREG ROSSTON:  Thanks a lot. I guess they said that the earlier panel was the bash the FCC panel. I think Coleman and I will make this bash the FCC II for at least part of it. So I apologize to FCC folks in advance for that. My slides, as you can see, are basically showing you the entire range of spectrum that weíre talking about, and thereíll be a continuous loop of the spectrum going forward for that.

My job has been sort of to look at the Spectrum Policy Task Force and think about how does it affect spectrum policy and whatís going to happen. Hopefully, it will shut off cell phones in conferences. That wasnít part of the recommendations, but maybe that should be one.

I'm going to divide my comments into two types of comments. One is sort of on the -- as a process observer, and then the second is spend more time thinking about the economics of the spectrum market. The wireless market, as everyone knows, is rapidly changing everything. And on the plane I didnít even think about this, but Scott should be really happy because I bought a copy of Forbes. Then I got on the plane and realized I could have saved my five dollars and just ripped the pages out of the one that was on the plane. But I did buy it. And itís got a list of the Midas touch -- the tech tips of top fifty venture investors. And I looked through this because itís important for my job at Stanford because we raise money, and a number of them are supporters. I also looked through them, but unfortunately, none of them are my relatives. But in this they ask these tech tips of these guys and James Way [Phonetic] whoís number twelve on the list, theyíve asked him whatís hot and whatís not, and he says under not, the whole 802.11 area is overfunded. But then number thirty-one on the list, Andy Radcliff [Phonetic] of Benchmark Capital says whatís hot is 802.11. so whatís clear to me is that these guys donít know exactly where to invest. You know, they have different opinions about whatís going on. Theyíre competing with each other about what theyíre going to do. And I think it has been and will be impossible for the FCC to know where technology and demand are going. And itís important for -- if youíre going to engage in a command and control or try to do this stuff in a way that doesnít let the market do it, that you have to do where technology and demand are going or set up rules that allow technology and demand to -- allow the flexibility to change in response to changes in technology and changes in demand.

So I think the FCC report does a good job of acknowledging this and setting up a framework that says we need flexibility. We need to have grant flexibility as much as possible to licensed users and unlicensed users. So I'm not completely bashing the FCC report. I think it really does, you know, sort of say the right things in a broad, general framework. So I think thatís important.

In some sense when I first read this, Tom had asked me what my reaction to the Spectrum Policy Task Force Report was, I said well, theyíve set up the framework, I in some sense drawing from the movies, this could be the feel good report of the year. It sort of says the right things and sets it up and sets the framework up, but I worry that itís going to suffer the same fate that lots of spectrum policy reports in the past have, including one that I helped write. It was an FCC staff working paper, and lots of things going way back in history, as Tomís lengthy tomes will attest, thereís been lots of people who have talked about spectrum reform. Evan [Phonetic] has been a big proponent of spectrum reform. And a lot of these have said hey, this is great. And the FCC, even commissioners say this is great. And then go ahead and say well, this is a special case. We have to make sure we keep broadcasting special, we need to make sure we keep other things special, and it doesnít necessarily work.

I think that the FCC has to go a lot further than what their Spectrum Policy Task Force says, and in some senses I know the FCC wonít go as far as the Spectrum Policy Task Force goes, so I'm sort of disappointed that the task force report didnít sort of take a really strong stand out on the edge. Because you know, the commission is going to cut back.

For example, just recently, and I'll acknowledge, I did some work for one of the satellite proponents that was talked about earlier in trying to get additional terrestrial rights for their satellite spectrum. Now, what the FCC could have done was given full rights to the satellite parties to use ancillary terrestrial broadcasts on their spectrum. They could have gone the opposite way -- they could have said no rights. Or they could have done two interim things which is, said well, thereís terrestrial rights that go with this, what Toni Cook Bush would say, and auction them to anybody who can use them. Or they could have done something where they said weíll give limited rights out. And the task force report says hey, you should be giving full flexibility out to people when you can. Well, the FCC, in a decision that came out after the task force report, went ahead and made a decision that said weíre going to give limited rights. Weíre not going to give full flexibility. Whoever they gave full flexibility to, they could have done full flexibility and ended up sort of not doing it that way.

So what the FCC often says in its reports and orders on everything from you need a report -- the umbone [Phonetic] that regardless [Phonetic] to everything else is we have to balance the interests. And theyíre balancing the interests of the parties which is not the same as maximizing the public benefits from the airwaves. And so they sort of -- and even in this task force report they talk about balancing the interests. So what I think needs to happen is the FCC needs to go ahead and take a bold step, and actually go ahead and tie its own hands and say weíre going to give spectrum out, weíre going to give full flexibility, weíre going to do this.

So thatís my sort of view as a process observer of the FCC and sort of someone whoís been involved in spectrum policy debates in the past.

Now I want to go on to some of the economics of spectrum policy. And the key issue obviously in spectrum policy is interference. And Colemanís talked about that some. I'm going to not talk directly about it, but talk -- so not about adjudication and other things. There are different visions for how you deal with interference. Economists have sort of long proposed a property rights regime to deal with interference and how to deal with a scarce resource. Or if you have a scarce resource like usable spectrum below five gigahertz, thereís a property rights approach. There have been people more recently whoíve been advocating a commons approach to spectrum policy because they think that the property rights approach rather than dealing with scarcity is creating artificial scarcity. And without property rights they think there would be no scarcity.

Kevin Ormoch [Phonetic] has come up with a great analogy that I think is very useful, which is this ocean analogy. The spectrum is like an ocean and we donít have to have property rights for shipping lanes or anything because thereís plenty of room for ships to pass. So that would be wonderful if we were in that situation. I have the fear, and I think maybe more likely the occurrence is that weíre in the harbor. And there are lots of ships. And thereís got to be rules and thereís got to be ways to figure out how to deal with this. And so sort of thinking about this is somewhat different.

Kevinís group is much different. Kevin had a confronta -- [Phonetic] [sound break] in Palo Alto a couple of months ago and the audience was about the same size as here. There was Ė every single person, or not every single person, because I didnít come with a laptop the first day I came, but the second time I came, had their laptops up. They were all blogging the whole conference. Not a single person had a tie. And so I think itís a very different group. And I think thereís an important way to try to get these people to talk together and try to see how our different visions of spectrum policy may work together or not.

The task force report does a very good job of explaining three different mechanisms for spectrum. And I'm not going to talk about the command and control part. I'll just sort of leave that out for right now. But what I consider the three different parts are the licensed, the unlicensed and under the noise floor or the ultra-wide band things. And a lot of people, the commons advocates tend to want the last two together. I think the report does a good job of disentangling the two. Both of these are predicated to some extent on low power and a small geographic coverage so that you donít cause interference. Theyíre both also based in an open entry model. And then they have Orion [Phonetic] time technical protocols and good behavior, and somebody has to go ahead and set the rules of the road. Some of these receiver performance rules and things like that.

The unlicensed differs somewhat from the ultra-wide band stuff in that unlicensed spectrum, as Martin talked about, requires that a dedicated block of spectrum be set aside for unlicensed use. And then that spectrum is unusable for a licensed user because itís filled with unlicensed --

[END TAPE 2 SIDE 3]

[START TAPE 2 SIDE 4]

-- operate under the noise floor and sort of on a non-interfering basis. So Martin brought up the important point that is sort of -- is addressed in a -- but glossed over in the spectrum policy task force -- how to make the determination between how much spectrum you put for licensed use and how much for the commons. I'll read these couple of quotes from the task force report:

The exclusive use model should be applied primarily, but not exclusively in bands where scarcity is relatively high and transactions costs associated with market-based negotiations of access rights are relatively low. And then they say the commons model should be applied primarily, but not exclusively, where scarcity is relatively low and transactions costs associated with market-based negotiation of access rights are relatively low.

I bet lawyers were involved in writing these. Thereís enough wiggle room in there to drive any proposal that Billy Tosian [Phonetic] wants right through that and figure out exactly what he wants to have done. It sort of says hey, thatís the right framework, but it gives the commission as much wiggle room as it needs to do whatever the heck it wants. Of course, they also give exceptions for satellites and broadcasters as well. So in some sense it doesnít tell us a whole lot about how it should be done. I think they actually do say well, we should -- then in a little more detail they get into we think that licensed use should be primarily -- that the spectrum below five gigahertz should be primarily, but not exclusively exclusive use and not above fifty gigahertz should be primarily, but not exclusively unlicensed. And so they sort of have the -- they continue to sort of be selective in what they do, but want to be able to selectively use unlicensed in the lower band and sort of give it enough room so that Barbara Boxer could introduce her bill saying you have to allocate more spectrum for WIFI and get people involved in this. Itís, you know, this is going to come up to a group where people like Gregg Skall are going to come in and lobby for their uses in front of the FCC or you have some problem, because licensed users who want to go ahead and get spectrum allocated to licensed use, well, they face an auction. If an auction were perfectly efficient and took away all the rents, no one would an incentive to lobby for licensed spectrum. I donít think they take away all the rents, because there is some idea that they do actually go ahead and bid in the auction and sometimes make money, although Martin worries about blood going to their head during the course of the auction. On the other hand, the unlicensed guys have the same problem in terms of lobbying for spectrum. They have an incentive to lobby if the unlicensed open entry model doesnít take away all the rents as well.

So the balance that the spectrum task force talks about between these areas, well, do we need a balance between two different groups that may not lobby enough to be efficient but also in my mind shouldnít be lobbying at all. The approach that I've advocated for spectrum has been an auction. And I think that there isnít any, you know, for a lot of spectrum there is the chance that if I think unlicensed is going to be great and thereís going to be a lot of WIFI people out there, that the equipment manufacturers who stand to benefit from this could actually go and buy licensed spectrum and have a licensed/unlicensed band. Where they could have it as open entry and charge five dollars a unit for the radios. Have a fixed cost. Have it low power. Say weíre the ones who set the standards. Weíre the ones who set the rules of the road. And actually participate in an auction.

When I was at the FCC people continually came up to us and said well, we have a great idea, but we canít afford to participate in the auction. We canít solve this coordination problem. Well, itís -- the private radio guys have participated in the guard band auction and got -- actually got spectrum and are actually doing something similar to this although for years they protested that they could never do it, it would never work. Auctions were not for private radio. They did it. I talked to Gregg Skall at the break. His company couldnít participate, you know, we needed this incredible break on spectrum, they needed it to be part of the public safety. Well, the auction was delayed because no one was willing to come up and say we would bid the minimum opening bid in the auction of eleven million dollars for five megahertz of spectrum. I have a feeling that the cost of the system was going to dwarf eleven million dollars. So I think that there are ways for unlicensed users to ban together and buy spectrum as well and have commons approaches. The trick is the FCCís got to make spectrum available and get it out in the market.

I want to talk a little bit about, if I have a little more time? You know, the tragedy of the commons. Economists talk a lot about the tragedy of the commons in unlicensed use. You know, this is -- a lot of the proponents of unlicensed spectrum say well, people will have an incentives to adopt more efficient technology and technology will work and people wonít do these things that are bad. So weíll -- itís, but in my mind thereís a difference between reducing scarcity and eliminating scarcity. And a lot of these technology things can have great potential to reduce scarcity, but may not eliminate it. I sort of have this fear that itís going to be Yogi Berra will be right. That itís going to get so crowded no oneís going to go there anymore.

Going back to Kevinís conference, one of the guys said I have a great way to know that I can always broadcast and get a clear channel on my WIFI. I just leave my thing on broadcast all the time. You know, that could be a real problem especially when you start getting commercial service providers like Boingo [Phonetic] and Team Mobile [Phonetic] who want to use this spectrum for commercial use and they happen to put up a system next to my house. Whatís going to happen? So I worry that there is this -- is this problem with the tragedy of the commons and it doesnít necessarily get solved by technology.

We have some examples of spectrum thatís gone for unlicensed or commons use before. And 802.11 has been incredibly successful. I have one in my house, a system, and it works really well. None of my neighbors have one as far I can tell because before I got it my laptop got no signal. So I'm one of millions. And thereíre lots of uses of this, but thereís also, as weíve heard throughout the day here, thereís lots of people who use cell phones too and value that. So that it doesnít necessarily mean just because WIFI has been very good that it necessarily is better than cell phone technology.

But thereís a couple of other examples. In the private radio bands where thereís open entry, they have a proceeding that I worked on when I was at the FCC called reforming. They had been there and they had this old technology in the private radio bands. And they came in in 1996 or so and said weíve got too many users, itís getting congested, we need to move to a better technology. We said great, move to a better technology. They said we canít do it. Why not? Well, thereís this incredible coordination problem. We have fourteen million users in this band and we canít get everybody to switch. Why not? Well, if I switch and I get more spectrum conserving technology the rest of you benefit. So I have no incentive to go and spend the money when I donít get any benefit from the better technology. Thatís the same similar thing to what could happen in these unlicensed bands. What happens five years from now when the 802.11 standard is found to be not that efficient and the band is overcrowded. Whoís going to have to mandate the new technology that comes in? Is it the FCC going to come in and tell everybody no, no, these WIFI cards that you bought are no longer good; you have to go invest in new ones. The FCC is not real good at telling people to get rid of their old technology.

The other is that thereís been a fair amount of spectrum foreign licensed use already. And if it truly is non-interfering they should be able to use a lot of the spectrum they have and there is some spectrum, unlicensed spectrum in the unlicensed PCS spectrum thatís close to lying fallow. And why is that? Well, because the FCC picked the wrong standards. So continuing through FCC picking standards, FCC having to be part of this process makes it pretty difficult.

The last thing I want to touch on since I'm probably going over my time, is the transition to a more market-based approach and more flexibility. The report lists five alternatives for how we can get there from here. And, you know, this is part of my thing, that the report, I think, should have taken a bold step. It says well, we should get a hundred megahertz of spectrum flexible -- we should have a goal of getting a hundred megahertz of spectrum of flexible use over the next five years. Thatís pretty small and pretty long and pretty soft, I think. Released at the same time was a paper by Evan Querrell and John Williams that said in two years we get a 438 megahertz out. So it sounds a lot better. I think that the bottom line is the task force report and the commission needs to take much more proactive action to get spectrum out into the public with flexible use, you know, no matter how it gets it out there, and get it out there so that people can start using it for location monitoring, for cell phone, for sending pictures, whatever it is, itís got to take this -- get it out there and get the spectrum flexible use defined so that people can use it.

So my major disappointment was that the task force report wasnít bold enough and wasnít aggressive enough.

MR. SCOTT WOOLLEY: We have about twelve minutes for questions.

MR. ARNOLD KLING:  Arnold Kling [Phonetic]. It sometimes seems to me that the -- you make -- that the Russians [Phonetic] way of making deregulation sound incredibly complicated. And the simplest approach might be easier. What if you just said that any spectrum license can be used for, you know, that itís up to the licensee to just, I guess itís go to the exclusive use model. And then just let the market sort out how these licenses get used and traded. I know that sounds like a mess, but it may be the quickest way to kind of get the, you know, get out of the regulation business.

MALE VOICE:  Take the television band, all right, where the licenses are defined on a six megahertz swatch in a certain geographic area and thereís a lot of dead space thatís there for interference purposes. Who owns that? If you say the current licensees own it then the channels start fighting with each other over who gets to take that space. If you auction it off you then have the continuing problem of how much interference protection do the channels really need under the current -- thereís a current standard with when thereís other television stations in the band as to how they sort that out and generally what the level of interference they have to accept. But if it was a cellular system thereís nothing on the ground now to figure out how much interference those stations would have to have. So you have first a problem of the dead space in between, and too, what I was talking about, resolving the conflicts between the licensees.

MR. GREG ROSSTON:  In addition I think there are, certainly Martin talked about which is how do you deal with windfall gains is something thatís, you know, since you say, in Washington very political. But in your view, okay, itís -- letís ignore that. The other is sort of the how do you redefine these rights to be full and exclusive use of this? For example, did the satellite guys have the full terrestrial rights for their spectrum? Well, you have geosynchronous satellites that are placed at different orbital locations that reuse the same spectrum. So you need to figure out a way to allocate that among the twelve satellites that can all see the United States and who gets the rights to use it. I think there are ways to do this, and I think this would be good.

The final point is the new idea of the ultra-wide band technologies that operate under the noise floor. How, you know, should you allow those or not? Ultra-wide band has a great potential for increasing entry without causing any harm, which could be really good if it can really work and really stay under the noise floor and no one sees them. And this is a great sort of reuse of the technology if that can be solved. And the question is would you be allowed to use that under this model of giving full exclusive use to people.

So those are the things. But in some sense, you know, giving, you know, just sort of getting out of the business is the right thing to do.

MR. MARK NADELL: Hi, I'm Mark Nadell [Phonetic]. I have a question for Coleman about your approach. One consequence of it that you may have considered, but I was wondering whether you had. You used the water analogy as a helpful metaphor, I guess. I was wondering whether youíd looked at the copyright legislation metaphor. And what I mean by that is you encourage more discussion among the parties, get the politics out.

Jessica Littmanís [Phonetic] written about how copyright works in this country, or it has for the last couple of decades. That the parties involved negotiate among themselves and then they bring it to Congress so that politics is out of it. But one consequence of that is the incumbents divvy up the pie among themselves and thereís a harmful consequence to new entrants who werenít at the table, and to the public in terms of social welfare.

Do you address that?

MR. COLEMAN BAZELTON:  Iím not familiar with, or hadnít thought about the analogy to the copyright. It strikes me that what you're touching on is the windfall gain instead of with the non-incumbents. Thatís an issue, and I donít know how you protect -- I have some ideas, but separate from how you protect the people who arenít on the ground there at the time, youíre still going to have this problem of how are the people that are involved going to get along. And thatís where I was focusing my comments.

MALE VOICE:  Maybe I could jump in and put that question Greg, to you. Do you see what would set up, now that the task force is proposing is setting up a fair fight between the licensed and the unlicensed folks, that is to say, give the unlicensed people a chance to prove their assertion that this tragedy of the commons problems really arenít going to be that big a deal and, you know, it strikes me, in the cellular industry now I think that this latest allocation might have 220 megahertz, but, of foreign spectrum, but some of the latest allocations from the FCC to unlicensed under six gigahertz I think is going to be three times that amount. So I'm wondering if, I donít know, just does it strike you as a fair fight between those two groups? A level playing field, I guess.

MR. GREG ROSSTON:  I donít know what definition of level playing field or fair fight is in this case. Yeah, they both have plenty of time to go and lobby the FCC is, you know, thatís where itís going to come out. And, you know, or go to Congress and get Senator Boxer to introduce bills mandating certain spectrum be allocated for WIFI. So itís, in some sense itís, maybe itís an unfair fight of a fight with different weapons than the fight that I would like to see which is the fight in the auction to get the spectrum. To fight, you know, using markets as opposed to using lobbying. So theyíre using the tactics of kind of lobby as opposed to how to prove it in the market.

MALE VOICE:  I seem to remember from my first year of public finance class that when you have a public good itís the sum of the marginal valuations that is how you value the band, right? So the question here is how does the sum of all these small unlicensed uses in a common area compare to the marginal valuation to a commercial provider? And I guess if a band gets overcrowded, right? The -- expand in a little bit so that you would have less congestion would increase everybodyís marginal valuation in that band. Beyond that, I've been at a loss as to how to implement that simple textbook rule about how to allocate this. I really donít know how youíre ever going to -- I havenít yet heard a good way to measure how youíre going to address that margin.

I've put out there the idea, although the both people I use this analogy to told me they didnít agree with it, but I'm going to try it on you guys anyway. That, you know, Central Park in New York is a public good in Manhattan, and I think everybody in Manhattan would like to keep Central Park. Thatís not the issue. The issue is, is it more valuable to take one acre out of Central Park and put it into commercial use or is it more valuable to take one acre out of commercial use and add it to Central Park? And itís, thatís what I see as the analogy to the trade-off with the commons problems. And for me, in New York it seems obvious that with land values in Manhattan that it would be a good idea to shrink Central Park a little bit and increase the commercial property a little bit. But [some laughter] apparently not everybody agrees with that. But again, itís hard to say how you would measure that trade-off.

MR. SCOTT WOOLLEY: Question in the back?

MR. PETER KISH:  Both of you are quite familiar with the process -- Peter Kish [Phonetic] with Intellectual. Once a friend of Tom Hazlett. [laughter] In any event --

MALE VOICE:  Once a friend of [interposing]

MR. PETER KISH:  Thatís right, thatís right. Anyway, I know both of you are very familiar with the process problems at the FCC. And in fact I think that underlay, to use a word thatís in vogue, a lot of your comments and critiques about how to go forward here. Isnít it the case that given that both the unlicensed approach and reforms and the exclusive use, flexibility, Querrell Williams approaches will have the potential for dramatically reducing scarcity, that the debate about which is better is really a false dichotomy. That, you know, both of these have trade-offs and problems. We certainly know that politically both of them have problems. But we donít know which is likely to succeed. And UWB holds great potential, but so does converting the MMDS spectrum and ITFS spectrum to a wireless Internet access exclusive use service. So why not press on both fronts?

MALE VOICE:  Absolutely. And I think the analogy is to, what Martin Cave was talking about, setting administrative prices on government uses of spectrum. The important point is to set positive prices. Thatís the first order effect. The second order effect is getting them just right. And I think you're absolutely right, that moving away from command and control is the key here. Exactly how much unlicensed versus exactly how much exclusive use is a second order problem.

MALE VOICE:  Yeah, I agree. I think that the, you know, my major complaint is that the report wasnít bold enough and itís going to take five years for the commission to have a, you know, to miss their goal of a hundred megahertz of flexible use. And so I think that thatís sort of, I definitely agree, they should be pushing forward much more on that.

I also think that, you know, investigating ultra-wide band technologies and the impact and understanding whether and how these may or may not interfere with existing license uses is great. If you can find ultra-light band technologies that do go on the underlay and are not non-interfering then open entry would be great for that. That we could have a lot more uses and that sort of thing. So I think we should push forward on both fronts as well.

MALE VOICE:  And didnít you used to be a friend of auctions too?

MR. SCOTT WOOLLEY: Great. Well, itís noon on the nose according to Sprint wireless time, so thank you all for coming. [Applause]

[END TAPE 2 SIDE 4]

 


Center for Digital Economy.

EMAIL THIS | PRINTER FRIENDLY

CENTER FOR THE DIGITAL ECONOMY CONFERENCE

Participants:

Introductory Remarks:

Thomas Hazlett

Gregg Skall

Bert Halprin

Toni Cook Bush

Martin Cave

Greg Rosston

Coleman Bazelon

Conference Agenda

Powerpoint slide presentations:

Dr. Coleman Bazelon

Professor Martin Cave

THIS TRANSCRIPT HAS NOT BEEN EDITED, AND MAY CONTAIN TYPOGRAPHICAL OR PHONETIC ERRORS.

 


Home | About MI | Scholars | Publications | Books | Links | Contact MI
City Journal | CAU | CCI | CEPE | CLP | CMP | CRD | ECNY
Thank you for visiting us.
To receive a General Information Packet, please email support@manhattan-institute.org
and include your name and address in your e-mail message.
Copyright © 2009 Manhattan Institute for Policy Research, Inc. All rights reserved.
52 Vanderbilt Avenue, New York, N.Y. 10017
phone (212) 599-7000 / fax (212) 599-3494