Friday, July 31, 2015

Q&A: Risk-assessment, harm claim thresholds and adjudication

In my testimony before the Senate Commerce Committee on Wednesday July 29, 2015 I recommended three spectrum management reforms. A summary and links to the written testimony and video are in an earlier blog post. This post offers some Q&A.

The three reforms were: (1) moving away from worst case interference analysis and using risk-informed methods that consider not only the consequences but also the likelihood of harmful interference; (2) providing more clarity about operators’ interference rights and obligations by specifying harm claim thresholds; and (3) giving any spectrum user the option of taking action directly against any other, either in front of an FCC judge or in a federal Court of Spectrum Claims.

1.1 When you talk about decentralization or shifting management to the marketplace, do you mean deregulation?


Implementing the tools I recommend does not mean that the FCC would give up authority.  It would be using its authority in different ways, such as using alternatives to command and control as a way to achieve its policy goals.

Doing this would simply recognize an inevitable consequence of our increasingly varied, crowded and dynamic spectrum reality: the need for private parties to take more responsibility for some aspects of routine spectrum management, notably adjusting rights boundaries and rights enforcement.

As a formal legal matter (caveat: IANAL) the FCC would still have the authority to change its mind, e.g. to roll back regulatory changes.

The goal is to enable private ordering and remove the FCC from a gatekeeping role where appropriate, e.g. in routine spectrum management.  The spectrum geek’s shorthand is "Coasian spectrum policy reform."

1.2 How would your three initiatives appeal to the deregulatory-minded since they all seem to require regulatory actions or institutions? 

All three facilitate a more economically informed view and allow more private ordering:
  • Risk-informed interference assessment lends itself to being used in combination with cost/benefit analysis since it takes a quantitative, full-system view of interactions between incumbents and newcomers.
  • Harm claim thresholds provide a solid basis for parties to negotiate adjustments to the FCC’s initial definition of rights; they are a basis for what’s called “interference trading”.
  • Private adjudication lets parties take action against someone they believe is harming their operation, without having to wait for the FCC to get around to it—or ties resolution of the dispute to some other, unrelated matter that is also before the Commission.

1.3 On the other hand, how do they appeal to those who don't think of centralization as a problem?

None of the proposals the FCC's authority:
  • Risk-informed interference assessment is just another (though I think better) tool to use in deciding allocations.
  • Harm claim thresholds retain and build on the definition of harmful interference in 47 CFR Part 2.1 by adding a quantitative test before a party can claim harmful interference—which would still be adjudicated using the definition in Part 2.1.
  • Private adjudication gives parties an alternative way to resolve interference problems. If the FCC is functioning well, it won’t be used.

Many who see an important role for government worry about the abuse of market power. All three proposals will probably tend to reduce the uphill slope of the playing field that faces new entrants.

1.4 What about sharing v. clearing, or licensed v. unlicensed?

The challenges we face apply regardless of whether one favors
  • spectrum sharing, clearing and reallocation, or some hybrid (like the AWS-3 blocks where cellular licensees have to protect weather satellite earth stations); or
  • licensed or unlicensed allocation, or some hybrid (such as the 3.5 GHz band where unlicensed devices will be controlled by a Spectrum Access System). 
These choices are important, but the answer won’t fundamentally change the underlying physics and governance challenges that come with squeezing more, and more diverse, services ever more closely together.

The remedies I propose apply regardless of these choices.

1.5 What is risk assessment?

Risk assessment sets out to answer three questions: What can go wrong? How likely is it? What are the consequences?

For example, when considering whether to install a burglar alarm system one might consider the various circumstances under which unwanted people might enter your house; how likely each possibility might be; and what might harm might befall in each case, from pranks and petty larceny to assault.

1.6 What’s the difference between worst case and a risk-informed assessment?

In engineering practice, risk is typically evaluated by considering the combination of likelihood and consequence for multiple hazards. By contrast, a worst case analysis focuses on a single scenario with very severe consequences, regardless of its likelihood.

For example, when deciding on the amount of domestic protection to buy, most consumers do not plan for a worst case like home invasion. Rather, they take a view—based on the likely threats in their neighborhood, the severity of outcomes, their personal need for security, and costs—of various options like deadbolts, burglar bars, intrusion alarms and steel doors.

1.7 What are the benefits of risk-informed interference assessment?

Quantitative risk assessment will improve the analysis of harmful interference, and lead to wiser trade-offs.

The benefits include:
  • Providing quantitative information to policy decision-makers who are balancing the benefits of a new service against its adverse technical impact on incumbents;
  • Providing a single framework for comparing different interference scenarios and assessments, in other words, enabling apples-to-apples comparisons of different kinds of interference; and
  • Enhancing the completeness of analysis and increasing the chances of identifying unexpected harmful interference mechanisms.

1.8 How can Congress encourage the FCC and NTIA to start using quantitative risk assessment?

Options for Congressional action include:
  • Making it part of the oversight process
  • By requiring risk assessment in legislation
  • By making sure that the FCC and NTIA—through auction revenues or otherwise—have the resources to advance the state of the art in spectrum management. 
  • By encouraging research on this specific topic by federal labs, NSF etc.
(On legislation: Various executive orders (Clinton 1993, Bush 2003, Obama 2011) have required risk analysis and cost/benefit analysis – but Executive Order 12866 and circular A-4 don’t apply to independent agencies. The CRS (2014) believes the the President could amend Executive Order 12866 and OMB circular A-4, or Congress could enact legislation.)

1.9 Why does one need to adjust the rules after they’re made?

The rules that govern how radios are operated aim to strike a balance between the interests of incumbents and the value of new services. Given the imperfect information available to the regulators, the balance is probably not optimal, and even if it was perfect to begin with, it will become obsolete as technologies, businesses and missions evolve. Adjustments to the rules are therefore inevitable and desirable.

1.10 What problem does harm claim thresholds solve?

The FCC does not generally define rights and obligations clearly, providing clarification only after lengthy rulemaking—and such clarifications are purposely ad hoc and cannot be generalized.

In the past, with fewer demands on the resource, the FCC could afford to proceed in this manner. Going forward, this model of spectrum management will become more and more costly to adhere to.

At the moment, adjusting the rules almost always has to be done by the regulators, which is slow and inefficient. We need to delegate as much spectrum management as possible from regulators to spectrum users.

1.11 What are the benefits of harm claim thresholds?

Setting a harm claim threshold will help parties to optimize spectrum boundaries and resolve disputes among themselves, without relying on the government.
  • It reduces uncertainty about the rights and responsibilities regarding interference for both interfering and affected parties, allowing them to plan and invest with more confidence.
  • It moves decisions about system design, including receiver performance, away from government to where it belongs: with manufacturers and operators.
  • It allows parties themselves to adjust operating rights and spectrum boundaries, which reduces rent seeking and the load on regulators.
  • It facilitates the resulting negotiations by providing an unequivocal starting point.
1.12 Why is federal spectrum, in particular, a problem for enforcement and adjudication?

The challenge is particularly acute when it comes to federal spectrum uses, since these services are vital to that national interest, they are growing for the same reasons as private uses and, in many cases, federal jurisdiction over spectrum bands is shared with the FCC.

1.13 Are you saying that all three initiatives must be adopted together as a package? If you could only recommend one, which would it be? Why?

The three each bring benefits on their own, but there are synergies as I described in my remarks.

If I had to choose one (and I'd rather not) it would be risk-informed assessment.
  • It’s the biggest bang for the buck, since it can affect all new allocations
  • There is a recognition at the FCC that there needs to be more use of statistical analysis, i.e. looking at the probabilities of various events, not just the worst case.

Harm claim thresholds are a close second, since the method is already in use, both by private parties (AT&T and Sirius XM) and by the FCC (3.5 GHz rules for Priority Access License systems).

No comments: