After major regulatory advances at WRC-95 and WRC-97, the broadband satellite industry is moving into a new phase. Rather than a handful of big questions to be decided on a global level, we now face a larger number of subsidiary questions, mostly on a national level. There are landing rights issues, including blanket licensing, band segmentation, and perhaps relocation of incumbents. There are bilateral coordinations between administrations. And there is the challenge of integrating broadband satellites into more general regulatory schemes dealing with intellectual property, privacy, interception requirements, and content regulation.
Because these issues are so numerous and so inherently national, no single regulatory issue is as important as the prevailing mindsets of regulators everywhere as they address these issues. Much will depend on what "everyone knows," and what "everyone knows" is often wrong. Regulators must be especially wary of two very dangerous illusions.
The first is the illusion of scarcity. By this I mean the illusion that, because regulators can no longer grant every single request for a particular resource, that resource must be scarce. This illusion runs rampant right now in the United States, where second-round applicants for geostationary orbital assignments in the Ka band earnestly discuss the "scarcity" of available orbital locations despite the fact that there are practically no Ka-band satellites in orbit or likely to be in orbit anytime soon. Not since the time of Galileo has there been such divergence between what is actually in the heavens and what people on Earth say about it. If our congested regulatory process suggests resources are scarce when in fact they are practically unused, then perhaps we should change the process. And in fact, some of us have called for the FCC to scrap the "processing rounds" that distort the U.S. licensing process.
The parallel problem at the ITU is the huge backlog of coordination materials awaiting publication. This problem has at least two distinct aspects – namely, the volume of filings and the speed with which they are processed – but the end result is the same: delay. Delay, of course, begets uncertainty, and operators respond to uncertainty by building flexibility into their plans. And one of the easiest ways to add flexibility is to file for more orbital resources than necessary. This overfiling, in turn, begets more delay, which begets more uncertainty, and so on. The ITU itself has correctly diagnosed the problem. There are many powerful incentives toward overfiling and virtually no potential negative consequences to create any offsetting disincentives. Efforts are underway at the ITU level both to reduce the volume of filings and to process them faster, and we can hope that these efforts will be successful.
It is interesting to speculate about the relationship between the U.S. problem and the ITU problem. The ITU backlog seems to exacerbate the problem at the FCC by creating a huge "window of unknowing" – a period during which it is difficult if not impossible for U.S. applicants to know where they stand at the ITU in the line for orbital resources. But I wonder whether anyone has ever tested the empirical link in the other direction. Could it be, for example, that ITU filings pick up in anticipation of FCC processing rounds? It is widely understood that the processing round system in the U.S. induces companies to file applications at the FCC many years ahead of actual need, despite rather thin commitments to the projects described in the applications. They do this to "keep their options open." Is there any reason to doubt that sophisticated observers outside the United States file at the ITU for exactly the same reason?
The second illusion, perhaps more dangerous, is the illusion of abundance. By this I mean the illusion that if a thing can be done once, then it necessarily can be done over and over again – and therefore should be done over and over again, without regard to cost.
This illusion infects much discussion of non-geostationary FSS sharing issues in particular. A few short years ago, most observers were convinced to a moral certainty that having even a single non-geostationary FSS system in a single band was nothing more than a pipe dream. Practically overnight, the prevailing wisdom changed. Now the assumption seems to be that co-frequency sharing between non-geostationary FSS systems is so easy that we should put three, four, or even five such systems in every FSS band. Accordingly, experts within the ITU are struggling right now to determine the how and the how many of such an aggressive co-frequency sharing regime. The stimulating intellectual debate carries over to processing rounds at the FCC.
It is remarkable that this has arisen as a major regulatory issue before we have a single operational non-geostationary FSS system. That is certainly not the way the geostationary side of the industry developed. The first privately owned communications satellites were not authorized until more than ten years after the creation of INTELSAT. The United States didn’t authorize private international satellites for more than ten years after that. And geostationary satellites had been operating for twenty years before anyone began to impose the constraints of two-degree orbital spacing – constraints which some operators refuse to accept even today. Some administrations today find it intolerable that any FSS band should be designated for the use of "just one" non-geostationary system. But in the 1960s when INTELSAT was created, no one said that if we could only have one satellite system it would be better to have none at all.
Even more remarkable is the fact that no one believes that we will actually have three to five non-geostationary systems in the Ku band, plus three to five more in the Ka band, plus three to five more in the V band. Why, then, are we putting ourselves through this? Our vision should always be fixed on the goal of deploying services in the real world, not on paper. If we cannot realistically project over a dozen non-geostationary FSS systems in the real world, why should our rules presuppose that many?
If co-frequency sharing were cost-free, it would do little harm to prepare for more success than we can reasonably forecast. But of course sharing is not cost-free, and by optimizing our rules for a dozen "paper" non-geostationary systems we impose burdens that may actually reduce the number of systems brought into use in the real world. The question is never whether co-frequency sharing is possible; it is whether such sharing produces benefits that outweigh its costs. This seems to be better understood every day where sharing between satellite and terrestrial services are concerned. If we forget this fundamental point with regard to sharing between non-geostationary systems, we may deal the non-geostationary segment of the industry a terrible setback before it even gets off the ground. We ought not to command all to run before even one has walked.
My partner Scott Harris tells me of a children’s story in which a princess sits by the water holding a precious golden ball. Looking down, she sees another golden ball in the water and reaches for it. But she discovers the second ball was only a reflection, and in reaching for the reflection she drops the ball already in her hand and winds up with nothing. That’s a story that many of us in the industry need to hear. As we work through technical and regulatory issues for exciting new services, we must never lose sight of our real goal, which is more satellite service to the world’s users. Not more paper satellites, but more real ones.
First Delivered In June 1999
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.
