I once got into a heated debate (well, maybe I was the only one getting heated) at a school cocktail party with (IIRC) a former chief economist from the FCC about spectrum auctioning.
TL;DR: there's a cadre of policymakers (current commissioner Ajit Pai among them) who are dogmatic about applying Ronald Coase's famous 1959 article "The Federal Communications Commission", which argued that a property-based market pricing system was the most efficient way to allocate spectrum.
I tried to argue that the fundamental problem isn't allocation of spectrum per se, but maximizing aggregate throughput. Yes, there's no disputing Coase's argument if we stick to the premise of figuring out how best to subdivide and distribute a finite set of spectral bands in a top-down fashion. But from a theoretical technological perspective that context was nearly outdated by 1959. By the time of the conversation in 2010, OFDM and beam-forming were standard, validating as a practical matter David Reed's argument that spectrum efficiency (in terms of number of data channels and aggregate throughput) is ultimately a matter of computational power and not primarily limited by spectral interference per se. The better allocation policy, therefore, is to rely more on open spectrum and let technology battle it out.
That's still a marked-based approach; indeed, it's more market-based; it's just not property-based. But he just wasn't having it.
This DARPA challenge is great, but there are policymakers, including policymakers currently in charge of the FCC, who fundamentally disagree with the approach. In their mind, the private owner of spectrum can be relied upon to select the technology that makes maximum use of the spectrum. But this completely ignores rent-seeking behavior. There is no ideal allocation of spectrum; it's all contingent on many different variables wrt to the path of technological development, not to mention the basic fact that competition isn't perfect.[1] Private owners of spectrum have an incentive to seek the outcomes that maximize their profit even if it they result in less efficient usage of spectrum (e.g. less aggregate point-to-point bandwidth). This has been proven empirically because until recently it was WiFi that was at the leading edge of spectral efficiency as deployed, and it still would be were it not for government regulatory policies that prevent WiFi and similar open spectrum environments from scaling up.[2]
If we can't agree on the basic problem we're trying to solve, we won't be able to agree on the solution.
[1] Importantly, Coase's argument is premised on perfect competition. Sadly, some very influential academics and policymakers adamantly refuse to accept that violating such a basic premise could potentially result in completely different (i.e. very inefficient) outcomes. This is the epitome of dogmatic, non-scientific thinking, but as it's based on a true _theoretical_ model it's cloaked as a scientific, fact-based approach.
[2] Which is not to say that there isn't a role for gov't regulation. But many regulations would be different if we went all in on open spectrum policy. If I'm right then in the next few years WiFi or something similar will begin pushing the envelope again, forcing telecom operators to continue competing with open spectrum solutions; or spectrum efficiency will plateau again and data rates will slowly rise as spectrum owners pursue rent-seeking over technological investment.
TL;DR: there's a cadre of policymakers (current commissioner Ajit Pai among them) who are dogmatic about applying Ronald Coase's famous 1959 article "The Federal Communications Commission", which argued that a property-based market pricing system was the most efficient way to allocate spectrum.
I tried to argue that the fundamental problem isn't allocation of spectrum per se, but maximizing aggregate throughput. Yes, there's no disputing Coase's argument if we stick to the premise of figuring out how best to subdivide and distribute a finite set of spectral bands in a top-down fashion. But from a theoretical technological perspective that context was nearly outdated by 1959. By the time of the conversation in 2010, OFDM and beam-forming were standard, validating as a practical matter David Reed's argument that spectrum efficiency (in terms of number of data channels and aggregate throughput) is ultimately a matter of computational power and not primarily limited by spectral interference per se. The better allocation policy, therefore, is to rely more on open spectrum and let technology battle it out.
That's still a marked-based approach; indeed, it's more market-based; it's just not property-based. But he just wasn't having it.
This DARPA challenge is great, but there are policymakers, including policymakers currently in charge of the FCC, who fundamentally disagree with the approach. In their mind, the private owner of spectrum can be relied upon to select the technology that makes maximum use of the spectrum. But this completely ignores rent-seeking behavior. There is no ideal allocation of spectrum; it's all contingent on many different variables wrt to the path of technological development, not to mention the basic fact that competition isn't perfect.[1] Private owners of spectrum have an incentive to seek the outcomes that maximize their profit even if it they result in less efficient usage of spectrum (e.g. less aggregate point-to-point bandwidth). This has been proven empirically because until recently it was WiFi that was at the leading edge of spectral efficiency as deployed, and it still would be were it not for government regulatory policies that prevent WiFi and similar open spectrum environments from scaling up.[2]
If we can't agree on the basic problem we're trying to solve, we won't be able to agree on the solution.
[1] Importantly, Coase's argument is premised on perfect competition. Sadly, some very influential academics and policymakers adamantly refuse to accept that violating such a basic premise could potentially result in completely different (i.e. very inefficient) outcomes. This is the epitome of dogmatic, non-scientific thinking, but as it's based on a true _theoretical_ model it's cloaked as a scientific, fact-based approach.
[2] Which is not to say that there isn't a role for gov't regulation. But many regulations would be different if we went all in on open spectrum policy. If I'm right then in the next few years WiFi or something similar will begin pushing the envelope again, forcing telecom operators to continue competing with open spectrum solutions; or spectrum efficiency will plateau again and data rates will slowly rise as spectrum owners pursue rent-seeking over technological investment.