THE ROLE OF GOVERNMENT IN TELECOMMUNICATIONS

STANDARD-SETTING

Kathleen M.H. Wallman*

I. INTRODUCTION: IS THERE STILL AROLE FOR GOVERNMENT IN STANDARD-SETTING?

Two trends are dominant in telecommunica-tions at the moment. The first is the convergenceof networks to form a network of networks;1 thesecond is convergence among proprietary net-work concepts and technologies to form publiclyaccessible networks. Both trends depend uponthe reliable and timely development of standards.Without standards ensuring interoperability of di-verse technologies, the benefits of network effects,which make networks increasingly valuable in di-rect proportion to the number of users, will be di-verted and in some cases, never realized or distrib-uted.

Part I of this essay frames the question ofwhether governments have a useful role to play instandard-setting and explains why this question isrelevant and important. Part II examines the con-tinuing role of standards in the modern telecom-munications economy and why standards are im-portant. Part III examines the current waystelecommunications standards are set anddeployed, including analysis of what works welland what needs improvement. Part IV establishesa framework for analyzing government's role instandard-setting and examines five examples ofplaces where governmental entities elected to in-

* President and Chief Executive Officer of Wallman Stra-tegic Consulting, LLC; former Deputy Assistant to the Presi-dent for Economic Policy and Chief of Staff and Counselorto the National Economic Council; former Deputy Counselto the President; former Chief of the Common Carrier Bu-reau and Deputy Chief of the Cable Services Bureau, FederalCommunications Commission; current Chairman of the Na-tional Coordination Committee, a federal advisory commit-tee formed by the FCC to recommend certain public safetyspectrum standards. The author gratefully acknowledges thevaluable comments on drafts of this essay offered by manycolleagues and particularly the work of Kevin B. Lefton (B.A.,George Mason University, 1998; M.A. in Telecommunica-

tervene or not to intervene. Part V summarizesthe characteristics indicating whether a particularstandard-setting situation is a good or bad candi-date for government involvement.

The standard-setting field is abundantly popu-lated with individual companies, trade associa-tions, professional associations, consortia andothers playing prominent roles.2 Governmentsalso have a role of their own as parties to suchtreaty organizations as the International Telecom-munications Union, an arm of the United Na-tions. They also work through officially sanc-tioned delegates, such as the American NationalStandards Institute's participation in the Interna-tional Standards Organization. But are govern-ments operating in their sovereign capacities in-dispensable to this process or should this role berelinquished to the private sector?

As developed below, the answer depends on thecircumstances. The problem is identifying whichtype of situation is best handled by government,and which is best left to the private sector. It isimportant to view past experiences in order toidentify potential future inefficiencies in the allo-cation of resources. Specifically, if the private sec-tor can do the job better or at least as well, thereis no reason to divert government resources awayfrom functions that government alone is qualifiedor empowered to do. For example, there is no

tions, George Mason University, 2000), also of Wallman Stra-tegic Consulting, LLC, who contributed to this article.

1 The development of seamless connections among wire-less networks in the U.S. and the connection of those net-works to the wireline network are examples of this trend. Ithas become commonplace for a caller to originate calls onone network that terminate on another. The calling partyoften is indifferent to whether his or her calls terminate on awireless or wireline network.

2 See HARRY NEWrON, NEWTON'S TELECOM DICTIONARY

742 (15th ed. 1999) (identifying over 100 private and publicentities involved in standard-setting in the telecommunica-tions area).

COMMLAW CONSPECTUS

reason to direct government resources away fromenforcement functions and into those standard-setting activities.

A companion question is whether there is con-tinued utility in the organized standard-settingprocess in light of the advent of de facto stan-dards. In other words, is it inappropriate to allowindividual companies, or consortia of companies,to set standards that respond to market demand,particularly when there is evidence that the mar-ket process is swifter and more timely than organ-ized processes or government action?

Here again, as developed below, there may beextremely important benefits to an organized pro-cess that is not controlled, at least ostensibly, byan individual company or companies. The bene-fits may include greater international propagationand acceptance of compatibilities based on stan-dards and greater openness to competitive "un-derdogs." Nevertheless, it would be beneficialfrom a resource allocation perspective to be ableto predict whether a particular standard-settingsituation is better handled by the resources of astandard-setting body or should instead be com-mitted to de facto standard-setting.

Another critical question to address is whether,even if government standard-setting is slower, lessefficient or less responsive to immediate marketdemands, it is or can be "fairer" in the sense thatit is more inclusive, more open to minority views,or protective of minority interests or potentialcompetition from those not yet involved in thespecific market. Additionally, government stan-dard-setting may be more responsive to public in-terest ideals. These factors must also be includedin any analysis that seeks to discover when govern-ment resources, as opposed to private sector re-sources, are warranted.

II. THE CONTINUING AND RENEWEDIMPORTANCE OF STANDARDS

Standards have always been important referentsin industry, even as the subject matter of the stan-dard-setting process has become more complex.For example, the process for setting standards

3 See Mark. A. Lernley, Standardizing Government Standard-Setting Policy for Electronic Commerce, 14 BERKELEY TECH. L. J.745, 745-46 n.3 (1999) (citing WILLIAMJ. CLINTON & ALBERTA. GORE, JR., A FRAMEWORK FOR GLOBAL ECONOMIC COM-

MERCE (1997) <www.ecommerce.gov/framework.htm> [here-inafter CLINTON & GORE]).

governing pipefittings would be relatively accessi-ble and straightforward. The interested partiesand those affected by the decision would be rela-tively easy to identify; the alternatives would berelatively few; and the consequences of onechoice over another would be reasonably foresee-able. By contrast, the process by which knowledge-able people arrive at the standards that governhow cable modems work in a network configura-tion is more complex. The downstream conse-quences of a standards-setting process on cablemodems are potentially uncontainable. They af-fect not only the cable business, but also the busi-ness of delivering telephony and data services,and possibly the over-the-air broadcast business.The level of technical expertise needed simply tounderstand the available choices is considerable.

The importance of standards is also increasing,owing to the networked nature of telecommunica-tions and information technology innovations.Standards are needed not only to coordinate out-comes but also to ensure the interoperability ofthe infrastructure.3 This underscores the need forcompatibility standards as opposed to merely co-ordinative standards. Coordinative standards-such as a declared standard for railway trackgauge agreed upon after a prevailing standard haspropagated in the market-serve the purpose ofinforming new manufacturers or suppliers to themarket and thereby encourage future conver-gence of competing standards.4 Compliance isnot mandatory but may be viewed as a conve-nience for new manufacturers. These new manu-facturers are equally free, however, to divert fromthe standard to employ a more technically ad-vanced standard or for tactical reasons, to imple-ment a new business plan.5 Telecommunicationsand information technology systems, by contrast,demand compatibility with respect to signalingsystems and conventions, conversion protocolsand modulation in order to work end-to-end. 6 Al-though paradigm shifts, such as the shift towardinternet protocol ("IP"), are still possible, it is im-portant that deployed standards be mandatoryrules.

7

Precisely because these choices matter so much

4 See SUSANNE K. SCHMIDT & RAYMOND WERLE; COORDI-

NATING TECHNOLOGY: STUDIES IN THE INTERNATIONAL STAND-

ARDIZATION OF TELECOMMUNICATIONS 34, 41 (1998).5 See id. at 34.6 See id.7 See id at 34-35.

[Vol. 8

Government's Role in Telecomm Standardization

in an economy that is dependent upon informa-tion technology, renewed attention is being caston the standard-setting process. The stakes arehigh not only in the United States for U.S. compa-nies and users, but globally as well. The compati-bility of U.S. standards with those adopted inother countries can make or break U.S. entry intoforeign markets, and can make or break the via-bility of foreign competitors in U.S. markets andabroad.

First, it is important to understand exactly whata standard is. "Standard" is defined as a "[g] uide-line documentation that reflects agreements onproducts, practices, or operations by nationally orinternationally recognized industrial, profes-sional, trade associations or governmental bod-ies."8 Thus, standards may address a variety of ar-eas, including interference prevention, interoperability, and service or product quality. Interest-ingly, this definition is accompanied in the text bythe following annotation: "Note: This concept ap-plies to formal, approved standards, as contrastedto de facto standards and proprietary standards,which are exceptions to this concept."9 So even infiguring out what this catch-all term "standard"means, experts are constrained to acknowledgethat a process that aims to be organized, hierar-chical and consensus-based often is not. And thereason for the need to acknowledge this excep-tion that often is the rule underscores what manyparticipants in the standard-setting process andusers of standards have figured out by experience:control the standard and be the master of yourown success.

Thus, the question of who should set telecom-munications standards matters very much to a lotof people. The answer to that question posed inthe present tense-"Who sets telecommunica-tions standards?"-is the classic hedging re-sponse: "It depends." There are many partici-pants.

In some cases, dominant industry participantsset the standard de facto. The emergence of theMicrosoft Windows operating system as the fa-

8 National Communications System Technology & Stan-dards Division, Telecommunications: Glossary of Telecommunica-tion Terms (last modified Aug. 23, 1996) <www.its.bldrdoc.gov/fs-1037/>.

9 Id.10 See GEOFFREY A. MOORE ET AL., THE GORILLA GAME

10-11 (rev. ed. 1999)I I See id.

vored desktop operating system is an example ofde facto standard-setting. 10 In fact, the skill of cap-turing enduring market share and thereby achiev-ing commercial success by promulgating de factostandards, is touted as a key criterion that inves-tors should seek and evaluate in candidates fortheir portfolio. 1

One key to success is to establish control over aproprietary architecture in a particular valuechain. The preferable architecture for this strat-egy also is open and exhibits high switching costs.An architecture can be both proprietary andopen, even though this sounds a bit counterintui-tive. It is proprietary when it is developed as theintellectual property of an enterprise; and madeopen when the proprietor publishes enough in-formation about the architecture and its inter-faces to allow other companies to build products,write applications, create services or otherwiseadd value that interoperates with the proprietor'sproduct. By building dependency on the architec-ture and everything that relies upon it, the origi-nal proprietor makes itself indispensable to userswho do not want to endure the headaches and ex-pense of trying something else-hence, highswitching costs.

In some cases, the industry that will be usingthe standard sets the standard. The cable modemstandard, known as Data Over Cable Service Inter-face Specifications ("DOCSIS"), is an example ofa standard promulgated in this fashion. 2 DOCSISwas originated by the Multimedia Cable NetworkSystem ("MCNS"), which initially was comprisedof Comcast, Cox, TCI (now AT&T), Time Warnerand CableLabs. 13 CableLabs then administeredDOCSIS testing and the certification of vendorequipment for compliance with DOCSIS require-ments and interoperability.14

In some cases, a group of companies or indus-tries mutually interested in and affected by a pro-posed standard will act collectively to arrive at astandard, usually under the auspices of a recog-nized standard-setting group, such as the Tele-communications Industry Association ("TIA").' 5

12 See Cable Modem Standards and Specifications (visitedMar. 30, 2000) <www.cabledatacomnews.com/cmic/cmic3.html>.

13 See id.14 See id.15 See, e.g., TIA Standards Development (visited Nov. 23,

1999) <www.csrstds.com/stddev.html>.

20001

COMMLAW CONSPECTUS

Participation in such a group is an important pro-tection against antitrust liability, provided thegroup follows procedures designed to ensure dueprocess is observed and that the procedures willnot be perverted to serve the anticompetitive aimof excluding competitors from a given field.1 6

Such procedures are themselves standardized, tosome degree, under the supervision of the Ameri-can National Standards Institute ("ANSI"), whichfunctions as a nonprofit member organizationthat is supported by private and public sectors.1 7

ANSI does not create or develop standards; itviews, and if appropriate, approves the work of ac-credited developers who submit possible futurestandards on a voluntary basis.' 8

In other cases, companies work outside the es-tablished standard-setting organizations in con-sortia to form standards. An example of this ap-proach is the Bluetooth consortium, a group ofcompanies that has formed a "Special InterestGroup" around the project of developing stan-dards for wireless connectivity for communica-tions appliances. ' 9

Occasionally, standard-setting work is under-taken by voluntary, private groups established forthe purpose of collaborating on a specific type ofstandard. The Internet Engineering Task Force("IETF"), the standard-setting arm of the InternetSociety, is an example of a private standard-settinggroup. Founded in 1992, the Internet Society'smembers stepped forward to stimulate coordina-tion of technical standards that enabled thegrowth and promotion of the internet.20 Becauseinternet standard-setting work did not fall clearlyinto the responsibility of any existing organiza-tion, the IETF initially competed with the Interna-tional Telecommunications Union ("ITU"). How-ever, these groups eventually agreed to worktogether on emerging technology standards. 2'

In yet other cases, the federal government, act-

16 See id. (summarizing TIA balloting procedures for in-

terested parties).17 See id. See also An Introduction to ANSI (visited Mar. 31,

2000) <web.ansi.org/public/about.html>.I8 See id.

19 See Bluetooth Special Interest Group (visited Apr. 1, 2000)<www.bluetooth.com>.

20 See All About the internet Society (visited Apr. 1, 2000)<www.isoc.org/isoc/>.

21 See Sandra Gittlen, IE7 and ITU End Rivalry, Join onConvergence Standards (visited Apr. 1, 2000) <www.nwfusion.com/news/1999/0715ietf.html>; Joe McGarvey & PeterJudge, IT Week: IETF and ITU Find Common Ground (visited

ing on its own or in concert with other nationalgovernments or industry members, essentiallytakes over the standard-setting process. Some-times the government convenes a federal advisorycommittee for the purpose of obtaining organ-ized industry and user input, as it did in the caseof standards for digital television.22 Governmentstandard-setting has fallen somewhat out of favorin recent years in the United States, with the Clin-ton administration promulgating a stated policyagainst government intervention in standard-set-ting.

2 3

Often, the nature of government involvement isblurred. The ITU operates with heavy de facto in-put from member countries' industries and theirtrade associations.2 4 In 1994, the Kyoto Plenipo-tentiary Conference of the ITU gave formal recog-nition to the role that private parties long hadplayed and accorded them certain rights and priv-ileges.25 Hundreds of service providers, equip-ment manufacturers and user groups now for-mally participate in the work of the ITU. 26

With so many interested industry players, tradeassociations and voluntary associations ready, will-ing and able to help set standards or take over theprocess altogether, and with the prevailing biasagainst government involvement in standard-set-ting, it is important to ask what role governmentlegitimately should have in this space. The realityis that despite the stated policy preference, theU.S. government remains involved in domesticand international standard-setting. It is thereforeimportant to be able to discern good prospectsfor such involvement-that is, situations in whichthe investment of effort by government is likely todo more good than harm-from bad prospectsfor such involvement.

There are two questions embedded within thisquestion. First, is there anything wrong, generally,with the current processes to the extent that it

Apr. 1, 2000) <www.zdnet.co.uk/news/1999/30/ns-9171.html>.

22 SeeJOEL BRINKLEY, DEFINING VISION 30 (1997).23 See, e.g., CLINTON & GORE, supra note 3, at 20 ("The

United States believes that the marketplace, not govern-ments, should determine technical standards and othermechanisms for interoperability.").

24 See Krishna Jayakar, Globalization and the Legitimacy of

International Telecommunications Standard-Setting Organizations,5 IND. J. GLOBAL LEG. STUrD. 711, 729 (1998).

25 See id.213 See id. at 718-21.

[Vol. 8

Government's Role in Telecomm Standardization

successfully produces standards without govern-ment involvement? Second, are there cases wheregovernment involvement can produce an affirma-tive benefit?

III. THE CURRENT PROCESSES: TAKINGTHE BAD ALONG WITH THE GOOD

Several observations, some of them critical, canbe made about the current processes: first, the defacto process by which proprietary technology orintellectual property becomes the prevailing stan-dard has the benefit, from an entrepreneurial,capitalist perspective, of letting the market deter-mine what best serves users' needs. 27 This means,in effect, that the winners in the system are com-panies that are agile enough to identify opportu-nities to capture an architecture and powerfulenough to sustain control over the architecture.This is nirvana for companies meeting these crite-ria, but not so great for companies outside the cri-teria. In one sense, this might be chalked up tothe workings of the market, but the alleged "un-fairness" of this system has been a source of com-plaint and concern among the United States'trading partners that have not yet developed astrong culture of entrepreneurship.

Not every complaint deserves redress, of course,and this is likely one of those that does not merit aremedy. The problem is that countries that havesuch complaints are likely to apply self-help. Forexample, the government of China recently re-portedly determined not to allow Windows 2000to be installed on government computers so as toencourage the development and deployment of acompeting operating system called Red FlagLinux, a product of Chinese software develop-ers.28 In the end, this problem is essentially a com-mercial calculation for the owner of the proprie-tary standard-are the near-term rents that flowfrom the initial exploitation of the de facto stan-dard greater than the revenues that the companyis forced to forgo when, and if, a backlash leads toforeclosure of a market? If they are, then there isno premium in opening the initial process to con-

27 See id. at 721.28 See David Renniw, Chinese Fly Red Flag with Ban on

Microsoft, DAILY TELEGRAPH, Jan. 7, 2000, at 16. But see We'llKeep the Red Flag Operating, COMPUTING, Jan. 13, 2000 (report-ing that China denied banning Windows 2000 from govern-ment computers).

29 See Cable Modem Standards & Specifications (visited Mar.

sultation or other steps that might avert suchbacklash.

Second, the industry committee process bywhich standards are developed can take a verylong time. The endeavor to establish a cablemodem standard had an original goal of Decem-ber 1995 for publication. 29 Most committeeprocesses are designed to achieve consensus,which can be elusive when the commercial stakesare high and competitive interests are diverse.Often, one or more factions will delay a final out-come in hopes that technological or commercialdevelopments will overtake an undesirable result.

Whether government intervention in a long-lin-gering standard-setting process can make thingsgo faster is imponderable. It merely sets up thequestion, "faster than what?"

Third, there is a risk that the process will pro-duce an answer that will end up isolating vendorsand fractionalizing the market rather than unit-ing them behind a standard that is well acceptedby users. An example of this is the wireless net-work standards situation that has ended up pro-ducing one set of prevailing standards for the U.S.market (Time Division Multiple Access and CodeDivision Multiple Access) and another for most ofthe rest of the world (Global System for MobileCommunications) . 3 The result is that travelersfrom Europe and Asia generally cannot use theirmobile handsets in the U.S., and U.S. travelersmight as well leave their telephones at homewhen they travel abroad. Only now in the ThirdGeneration ("3G") of wireless telephony has theredeveloped the real prospect of a more rational re-sult for the future. 31

IV. DISCERNING WHEN GOVERNMENT MAYPLAY A BENEFICIAL ROLE: SHOULDA,COULDA, WOULDA, DIDN'T AND DID

To benefit from past experience, we must firstdiscern when government intervention in a stan-dard-setting process has been beneficial. Chartingpast experience in terms of when the governmentdid or did not decide to intervene, against

25, 2000) <cabledatacomnews.com/cmic/cmic3.html>.30 See Coran Carlson, Economist Oppose Single 3G Stan-

dards, WIRELESS WEEK (Mar. 30, 1998) <www.wirelessweek.com/NEWS/3G/threegl0.htm>

31 See generally Industry Settlement Advances Standards Processfor Third Generation Wireless Services, FCC News Release, Mar.26, 1999.

20001

COMMLAW CONSPECTUS

whether that decision turned out to be wise orbeneficial by the best available objective mea-sures, is one useful tool.

The grid below notes cases in which the govern-ment did or did not become involved in the stan-dard-setting process and the consequences,judged retrospectively, of that action or the con-trary action. Of course, in the real world, theremay be gradations of "yes" or "no," representingvarious roles that government actors can choose

to play in the standard-setting process, even to thepoint of simply issuing a threat of involvementwithout actually getting involved. Such threatsalso have an impact on the process, but they areharder to document. Judging is based on whetherthe decision was normatively correct or incorrect,weighing the public good generated by the deci-sion. Thus, in each case, it may be said that thegovernment "should" or "should not" have gotteninvolved.

Table 1

Government Should Have Gotten Involved:Affirmative Benefit Achieved or Forgone

Government Should Not Have Gotten Involved:Negative Impact Realized or Avoided

Government Did Get Involved Affirmative Benefit Achieved: Negative Impact Realized:* Public Safety Spectrum Interoperability * Cable Set-Top Boxes Digital Television

Standard Transmission Standard" Digital Television ("DTV") Transmission

Standard

Government Did Not Get Involved Affirmative Benefit Forgone: Negative Impact Avoided:* Wireless Telephony Standards e Cable Modem Standards

A. An Affirmative Benefit to a GovernmentRole

There are two types of cases in which a signifi-cant, affirmative benefit to government involve-ment can be foreseen. The first is one in whichthe user market is relatively small in the sense thatuser demand is not great, and there is a risk thatthe normal industry and trade associationprocesses will "overlook" or not timely address themarket. The word "overlooked" is not used pejo-ratively. A market is "overlooked" not only when itis not developed because the prospects of profitare not judged great enough to attract interest,but is also "overlooked" when it is relatively lessprofitable compared to other markets. Therefore,in such circumstances, it may take so long forcommercial enterprises to explore prospects for amarket that the market and users' needs are effec-tively not addressed.

The second type of case is one in which theuser market is very large, but there is a risk that aprotracted competition among proposed stan-dards will significantly postpone important bene-fits to consumers. The legitimacy of this case is en-hanced when the benefits are judged to beimportant not only on their own merits but alsowhen there are larger positive externalities thatmay be captured by acceleration of the adoptionof the standard. The legitimacy of this case is fur-

ther enhanced when a shift to a new standard-the adoption of which is judged to advance a pub-lic good-is intended to replace or overtake astandard that is embraced by a large embeddedbase.

Three examples readily illustrate these twocases:

" the need for standards to ensure interoper-ability on spectrum used by public safetyagencies such as police, fire and emergencymedical services;

" the need for a declared, unified standard forDTV transmission; and

" the discontinuities caused by the incompati-ble standards that have emerged in the areaof wireless telephony (although as developedbelow, the proper government role is not atypical affirmative action to set the standardbut rather taking steps to preserve opportuni-ties for competitive standards to develop).

1. The Case of Interoperable Public Safety Spectrum

In 1993, Congress responded to long-standingcomplaints from the public safety community thatit was dangerously constrained by a spectrumshortage, evidenced by acute spectrum conges-

[Vol. 8

Government's Role in Telecomm Standardization

tion .3 2 In their response, Congress directed theFCC to designate additional spectrum for the ex-clusive use of public safety users. 33 Congress spe-cifically directed the FCC to designate for suchuse 24 MHz of the spectrum that will becomeavailable in the ultrahigh frequency band (chan-nels 60 to 69) when the transition to digital televi-sion is complete. 34

In implementing this mandate, the FCC soughtto resolve another long-standing problem in thepublic safety wireless communications area: thelack of interoperability among jurisdictions andemergency services. 3- Thus, in critical emergen-cies such as the Oklahoma City bombing or theColumbine High School shootings, emergency re-sponders from different agencies and jurisdic-tions have been unable to communicate directlywith one another from one handset deployed inthe field to another. In a makeshift workaroundsolution to this problem, agencies sometimesadopt a hub-and-spoke configuration in a com-mon command center in which one representa-tive of each responder stands in the hub and di-rects instructions and information needed tocoordinate operations from operatives deployedat the end of the spokes, back to the hub, andthen back to the spokes .3 This anomaly is an out-growth of the way that spectrum historically hasbeen allocated to the agencies. Some spectrum isespecially well suited to some types of publicsafety organizations. For example, relatively lowfrequency spectrum has been favored by somestate police agencies, which have a need for verywide geographic propagation of communications

312 See Omnibus Budget Reconciliation Act of 1993, Pub.L. No. 103-66, § 6002, 107 Stat. 312 (1993) (as codified in 47U.S.C. § 309)(j)(10)(B)(iv)). See also Talking with Champi-ons-JamesJ Flyzik, Treasury, AccEss AMERICA (visited Nov. 8,1999) <www.accessamerica.gov/docs/publicsafety-champi-ons.html>.

33 See In re Development of Operational Technical andSpectrum Requirements for Meeting Federal, State and Lo-cal Public Safety Agency Communication Requirementsthrough the Year 2010; Establishment of Rules and Require-ments for Priority Access Service, First Report and Order andThird Notice of Proposed Rulemaking, 14 FCC Rcd. 152, 163-64,para. 15 (1998) [hereinafter Requirements for Priority Acess Ser-vice Order].

34 See id. at para. 2. See also FCC Adopts Rules for LicensingLargest Block of Public Safety Radio Spectrum Ever Allocated (Rpt.No. WT 98-24), FCC News Release, Aug. 6, 1998.

35 See Wireless Telecommunications Bureau, Public SafetyNational Coordination Committee Executive Summary (visitedNov. 8, 1999) <www.fcc.gov/wtb/publicsafety/ncc/ncc-es.html>.

signals.3 7 Once a public safety user becomes en-sconced in a particular band and becomes com-mitted to purchasing equipment designed to op-erate in such frequencies, it tends to stay put. Theresult has been a lack of interoperability in criticalsituations.

The public safety community has been pursu-ing this problem for several years within the struc-ture of several organizations that have both publicsafety officials and equipment industry officials asmembers, such as TIA. 38 TIA is an ANSI-accred-ited standard-setting organization. 39

Essentially two competing standards haveemerged that could be adopted to address the in-teroperability problem: P25 ANSI from NorthAmerica and Trans European Terrestrial TrunkedRadio ("TETRA") European Technical StandardsInstitute ("ETSI"), which was developed in Eu-rope. P25 originated in "APCO Project 25" work-ing groups and is referred to as APCO Project 25,reflecting APCO's project numbering system, andtoday is simply known as Project 25 or P25. P25 isdocumented in ANSI as TIA-102, shorthand forthe entire suite of ANSI 102 standards. The tech-nical review and approval process has been as-sumed by TIA, ANSI's accredited standards devel-opment organization. Phase I of the P25 ANSIstandard provides a common baseline for inter-operability for both current and future P25 imple-mentations. The TETRA standard was developedand approved by ETSI. TETRA has gained cur-rency in Europe, where it has been implementedas the standard of choice among public safetyusers.

36 See More Frequencies Below 800 MHz Needed for PublicSafety Use, COMM. TODAY, Aug. 13, 1998 (discussing the needfor additional towers when using frequencies above 800 MHzfor public safety communications).

37 See Robert Fenichel, APCO Project 25-Here, Now, andInto the Future (visited Mar. 30, 2000) <www.apcointl.org/bul-letin/bull/99/march/feature4.html>.

38 Another such organization is the Association of PublicSafety Communications Officials ("APCO") Project 25 coali-tion formed by the joint efforts of the National Association ofState Telecommunication Directors ("NASTD"), the Associa-tion of Public Safety Communications Officials Internationaland representatives of the U.S. government. APCO Project25 includes participation from U.S. local, state and federalpublic safety users, as well as users from Canada. See TIAPress Release, TIA Publishes Systems Bulletins and Interior Stan-dard on APCO Project 25 Uune 25, 1997) <www.tiaonline.org/pubs/press-releases/1997/97-51 .cfm>.

39 See TIA Facts at a Glance (visited Sept. 2, 2000).<www.tiaonline.org/about/overview.cfm>.

20001

COMMLAW CONSPECTUS

From a policy perspective, as opposed to an en-gineering perspective that might focus more in-tently on signal propagation characteristics andtradeoffs in meeting user needs, there are prosand cons to each of these proposals. First, there isthe timing issue. Project 25 is ready for implemen-tation since it successfully has made its waythrough the TIA vetting process and ANSI accred-itation. 40 But Project 25 Phase I calls for channel"widths" of 12.5 kHz, which the FCC, driven byspectrum efficiency concerns, views as less thanideal. 4' The FCC has been pressing for a solutionthat involves channels of 6.25 kHz. 42 TETRA re-quires 25 kHz channels and uses Time DivisionMultiple Access ("TDMA") technology with fourvoice slots. It has been approved by ETSI, but isnot ready for implementation in the U.S. becauseit has not been through the TIA process, the U.S.gatekeeper for implementation and adoption.43

Although there is cooperation between the twostandard-setting organizations, there is no "fullfaith and credit" arrangement that obligates or al-lows one organization to treat standards approvedby the other as approved in its jurisdiction. An-other proposal, advanced by Ericsson, requires12.5 kHz channels and uses TDMA technologywith two voice slots. It also has not been througheither ETSI or the TIA standards process. Like theTETRA proposal, it has been introduced into theTIA standards process. 44

There are also competitive aspects to the de-bate accompanied by significant tension on thepart of vying interests. The TIA-102 standard wasembraced early on by Motorola, the manufacturer

40 See FCC Releases Service Rules for Public Safety 24 MHz Al-

location, Some Agencies Able to Begin Using Frequencies Prior to2006, 53 LAND MOBILE RADIO NEWS, No. 40, Oct. 2, 1998(describing the FCC's allocation of bands into 6.25 MHz).

41 See In re The Development, Technical and Spectrum

Requirements for Meeting Federal, State and Local PublicSafety Communication Requirements Through the Year2010, Fourth Notice of Proposed Rulemaking, WT Dkt. No.96-86, FCC 00-271, paras. 14-18 (rel. Aug. 2, 2000).

42 See id.43 See TETRA Overview (visited Mar. 30, 2000)

<www.etsi.org/technicalactiv/tetra.htm>; APCO Interna-tional Press Release, APCO International Project 25 CommitteeApproves Ericsson, TETRA Proposals (Nov. 19, 1999) <www.apcointl.org/project25/p25.html>.

44 See Robert Fenichel, APCO Project 25-Here, Now, andInto the Future (visited Mar. 30, 2000) <www.apcointl.org/bulletin/bull/999/march/feature4.html>.

45 Motorola Press Release, Motorola Introduces First Project25 Compliant Conventional System (Aug. 13, 1996) <www.motorola.com/General/Press/PR960826_4262.html>.

with the largest market share among public safetysystem users. 45 Motorola has participated in bothstandards venues and currently offers compliant

equipment to each standard in the respectiveANSI and ETSI areas of the world. Other con-tenders in the field, Ericsson and Nokia, were ad-herents of TDMA technology; Nokia supportingTETRA, and Ericsson 46 supporting its own TDMAsolution. Over time, the process affected a rap-prochement between the two standards in thesense of making progress on compatibility issues.Nokia, Ericsson and Motorola have all set forthproposals or guidelines in the P25 process for theNorth American Phase II TDMA solutions. 47

The decision concerning what standard shall beadopted for operations in the spectrum desig-nated for interoperability rests with the FCC.4

1

This is because the FCC is responsible for "typeacceptance" of radios that will operate at the des-ignated frequency. Since no radios previouslyhave operated at this frequency-it is currentlybeing used for television broadcast-this is a newexercise. 49 The responsibility for type acceptancederives from Congress' mandate to the FCC that itadopt regulations governing potential radio fre-quency interference. 5

1° In some cases, the FCC has

also included interoperability standards as part ofits type-acceptance requirements. Even so, theFCC could essentially delegate the duty of choos-ing the standard to others, and this it did in a Sep-tember 1998 order.5'

In the Requirements for Priority Access Service Order,

the FCC declined to adopt a standard. 52 Instead,it directed the formation of a federal advisory

46 Ericsson's private radio systems operations were re-

cently acquired by Com-Net Critical Communications. SeeEricsson Press Release, Ericsson's Private Radio Systems businessacquired by Corn-Net Critical Communications (Jan. 7, 2000)<www.ericsson.dk/pressroom/PressReleases/PRindex S_Corporate.html>.

47 See Ericsson Press Release, Ericsson Offers Proposal toHarmonize Third-Generation Mobile Communications (Dec. 8,1998) <www.gsmdata.com/arteric.htm>; APCO InternationalPress Release, Project 25 Committee Approves Ericsson, TETRAProposals (Nov. 19, 1999) <www.apcointl.org/project25/p25steering.rtf>; Motorola Press Release, Motorola Proposes Unify-ing Guidelines for North American Two-Way Radio Communica-tions Standard (Oct. 25, 1999).

48 See 47 U.S.C. § 302(a) (2000).49 See Requirements for Priority Access Service Order, 14 FCC

Rcd. at 154, para. 1.50 See 47 U.S.C. § 302(a); 47 C.F.R § 2.901 (2000).51 See Requirements for Priority Access Service Order, 14 FCC

Rcd. at 154, para. 1.52 See id. at 156-57, para. 7.

[Vol. 8

Government's Role in Telecomm Standardization

committee named the Public Safety National Co-ordination Committee ("NCC"). 53 The charge ofthe NCC was to fashion recommendations to theFCC concerning a standard for the interoper-ability spectrum, among other assignments. 54

Originally, the FCC contemplated that the NCCwould undertake an independent standard-settingexercise, 55 but upon reconsideration, the FCC au-thorized the NCC to rely upon previous work ofANSI-accredited standard-setting organizations,such as TIA.5 6 Without authority to do so, theNCC likely would have had to become ANSI-ac-

credited itself, which would have required moretime in a process that already had become argua-bly overdue in light of the statute's 1998 deadlinefor licensing the spectrum to public safety enti-ties.

5 7

The NCC is in the process of preparing recom-mendations for the FCC, an exercise that beganin February 1999.58 Initial recommendations wereforwarded to the FCC in February 2000, with thelife of the NCC anticipated to last up to four yearsbeyond the date of its 1999 inception. 59 The re-port recommended that the FCC embrace TIA-102 ANSI as the standard for the new spectrum, adecision that will put in motion production plansfor handsets and infrastructure that operate at thedesignated frequency based on the TIA-102 suiteof standards.

60

There are numerous characteristics of the pub-lic safety market and the standard-setting processthat made the government's involvement in thestandard-setting process for the interoperabilityspectrum more sensible than the alternatives.First, the market for public safety equipment-in-frastructure, base stations and mobile units-ismuch smaller than the market for commercialmobile infrastructure and devices. It is also frac-tionalized. Purchasing decisions are made at thestate, county or local level, and are often made bydifferent agencies and processes at the various ju-risdictional levels. There are numerous formal

53 See id.54 See id.55 See id.56 See Development of Operational, Technical and Spec-

trum Requirements For Meeting Federal, State and LocalPublic Safety Agency Communication RequirementsThrough the Year 2010, Memorandum Opinion and Order onReconsideration, 14 FCC Rcd. 8059, 8065, para. 11 (1999).

57 See id. at 8064-65, para. 10.58 See id.59 See id.

and informal professional networks to share andcoordinate information about dealing with ven-dors. But purchasing decisions and purchasingpower are rarely coordinated. Perhaps the nextbest thing is a role for the government in aggre-gating the selection power for standards, whilemaintaining sensitivity to the unique and variedneeds of public safety users. These effects, how-ever, must be balanced with the competitive vital-ity of the public safety market.

Second, the government role was not a matterof starting from scratch. Particularly after theFCC's reconsideration order, the government'srole, played out in part through the NCC, builtsignificantly upon what had gone before in pro-fessional and commercial standard-setting groups.

Third, the public good stakes are high. Theperpetuation of communications problemscaused by the lack of interoperability is not quan-tifiable in impact; it is difficult to say what tollsuch problems have already had in draining or di-verting emergency response resources. But al-lowing such problems to continue is an uncom-fortable path; their perpetuation seems virtuallycertain to lead eventually to direct adverse conse-quences that will be laid at the doorstep of theinteroperability problem.

2. The Case of Digital Television

One of the most contentious public policy is-sues in telecommunications during the 1990s wasthe transition to digital television ("DTV"). The is-sues subject to debate were legion. One issue wasthe basis on which the spectrum should be madeavailable to broadcasters-whether it should beauctioned, as most other spectrum for commer-cial use was being handled at the time pursuant tocongressional mandate. 61 In the end, Congressdecided to allocate temporarily 6MHz of spec-trum to each broadcast licensee without auc-tions.62 Another issue was whether the broadcast-

60 See Public Safety National Coordination Committee Recom-

mendations to the Federal Communications Commission for Techni-cal and Operational Standards for Uses of the 764-776 MHz and794-806 MHz Public Safety Band Pending Development of FinalRules (Feb. 25, 2000) <www.fcc.gov/wtb/publicsafety/ncc/feb25body.pdf>.

61 See Mass Media Bureau, Policy and Rules Division, Fed-eral Communications Commission, Digital Television Tower Sit-ing Fact Sheet and Frequently Asked Questions (visited Feb. 12,2000) <www.fcc.gov/mmb/prd/dtv>.

62 See id.

20001

COMMLAW CONSPECTUS

ers should be required to use the spectrum toprovide high definition television ("HDTV")-es-sentially sharper, richer television images-orwhether broadcasters should be permitted flexi-bility to use the spectrum for a variety of advancedservices. 63 Eventually, it was decided to affordbroadcasters flexibility. 64 Another issue was thetiming of the surrender of the spectrum now usedfor traditional analog broadcasting so that thatspectrum could be auctioned. 65 Congress decidedto schedule completion of the transition to digitalfor 2006, with escape clauses in the event thatthere is evidence that viewers have not made atransition by buying DTV receiver equipment(new televisions or new set-top boxes) .66

All of these policy and allocative fairness issuesbuilt upon the successful execution of a process toreach agreement on the even more fundamentalissue of what transmission standard to use forDTV. A single standard was desirable because ofthe characteristics of the mass consumer marketfor electronics-asking consumers to consider up-grading a television that is working fine (or to buya set-top box for a television that works fine with-out one) is one thing. But asking them to make achoice among competing standards advanced bycompeting brands was deemed a step more thanregulators and the industry should expect con-sumers to take. The FCC's order adopting theDTV standard reflects robust debate among com-menters about whether FCC action was necessaryand whether the market could operate withoutsuch intervention to establish a standard.6 7 TheFCC's order recites concerns about the delays thatstartup, coordination and splintering could causefor DTV adoption. 6

1 But it concluded "we are notconvinced that these problems are so severe thatthey would absolutely preclude us from allowingthe market to operate without a set standard."69

63 See id.64 See id.65 See id.66 The Balanced Budget Act of 1997 allows broadcasters

to continue to use the spectrum for which they are now li-censed beyond.2006 if, for example, less than 85% of televi-sion households in a market have the equipment to receiveDTV signals. See Pub. L. 105-33, 111 Stat. 251, 258 (1997)(codified as amended at 47 U.S.C. § 309(j) (2000)).

67 See In re Advanced Television Systems and Their Im-pact Upon the Existing Television Broad. Service, Fourth Re-port and Order, 11 FCC Rcd. 17,771, 17,777-79, paras. 10-14.(1996) [hereinafter DTV Fourth Report and Order].

68 See id. at 17,776, para. 8 (explaining that "[s] tartup re-fers to the situation where everyone would be better off

Nevertheless, the FCC continued, "we are con-cerned that market solutions may result in morethan one sustainable transmission standard. '" 70

The consequences, the FCC concluded, would beunacceptable:

Such an outcome might result in compatibilityproblems and increase the risk that consumer DTVequipment purchased in one city would not work wellin another city; that a receiver would not display all thebroadcast channels in a city; or that a digital televisionset purchased one year might not work several yearslater. Such results would hurt consumers and make itmore difficult to preserve a universally available broad-cast television service. 7'

By the time the rules were finally adopted atyear-end 1996, the process had been underway formany years, going back to 1987 in terms of theFCC's involvement. In 1987, the FCC chartered afederal advisory committee, the Advisory Commit-tee on Advanced Television Service ("ACATS"), tomake recommendations concerning the transi-tion to what was then called Advanced Television("ATV"). 72 Initially, additional spectrum alloca-tions were not contemplated in deploying ATV. 73

Rather, the FCC envisioned upgrades on existingbroadcast frequencies.7 4 But the FCC decided afew years later that temporarily allocating an addi-tional 6MHz of spectrum for deployment of ATVwould better advance a less disruptive transition. 75-

By February 1993, ACATS reported to the FCCthat it could not recommend any of the four com-peting digital television systems then under dis-cussion over the others because all required morework and development.71' Later that year, theseven entities that had been working on the fourcontending systems (AT&T and Zenith Electron-ics Corporation; General Instruments Corpora-tion and the Massachusetts Institute of Technol-ogy; and a consortium composed of ThomsonConsumer Electronics, Philips Consumer Elec-

adopting DTV technology, but no one has the incentive tomove first. Coordination is the collaborative effort by, broad-casters, consumer equipment manufacturers, and programproducers that is necessary to introduce DTV. Splintering re-fers to the breakdown of consensus or agreement to use theDTV Standard.") (footnotes omitted).

69 Id. at 17,778, para. 34.70 Id.71 Id.72 See id. at 17,773, para. 4.73 See id.74 See id.75 See id.76 See id.

[Vol. 8

Government's Role in Telecomm Standardization

tronics and the David Samoff Research Center)formed "The Grand Alliance" to develop a final,unified standard.7 7 It voted its final recommenda-tion in November 1995.78

The path of implementing the standard hasbeen wrought with controversy since its adoption.Issues concerning- surrender of the existingbroadcast spectrum continue to roil. The rolloutschedule for tower extensions necessary to deployDTV has been fraught with delay. And the meritsof the standard selected have been subject to chal-lenge. In October 1999, Sinclair Broadcast Group,Inc. filed a Petition for Expedited Rulemakingseeking a reopening of the DTV standard. 79 It

urged the FCC to modify its rules to allow broad-casters to employ the Coded Orthogonal Fre-quency Division Multiplexing ("COFDM") stan-dard that ACATS had rejected in addition to theone it had recommended. 0 Sinclair urged thatthe standard adopted by the FCC did not reliablyallow indoor reception of DTV signals, wheremost people like to watch television."'

In February 2000, the FCC rejected the peti-tion, citing the work of the FCC's Office of Engi-neering and Technology, which had studied bothstandards and "concluded that the benefits ofchanging the DTV transmission standard' toCOFDM would not outweigh the costs of makingsuch a revision."8' 2 The FCC also concluded that"the Sinclair petition had done no more than todemonstrate a shortcoming of early DTV receiverimplementation," which manufacturers wereworking hard to overcome.83

Even taking account of these bumps in theroad, and even though pervasive implementationin a way that benefits "real people" is still a longway off, it is not too early to observe that govern-ment involvement probably did more good thanharm in the DTV standard-setting process. TheFCC's order, however, demurs taking credit forthis positive impact, stating "[t]his proceedingdemonstrates how competing industries, workingtogether, can develop de facto industry-selectedstandards that satisfy the interests of contendingparties."8 4

77 See id.78 See id.79 See Advanced Television Systems and Their Impact

Upon the Existing Television Broad. Service, Petition for Expe-dited Rulemaking, filed on behalf of Sinclair Broad. (Oct. 8,1999).

80 See id. at i.

But the DTV standard is not a classic exampleof a de facto standard. It was developed by a con-sortium of companies under the auspices of a fed-eral advisory committee. The output of the advi-sory committee was circulated for public noticeand comment and eventually codified in federalregulations. These characteristics present abun-dant contrasts with the process by which, for ex-ample, Microsoft Windows became the de factodominant standard operating system for desktopprocessors. Government involvement was an indis-pensable element.

The characteristics that made the DTV transi-tion suitable for a government role in selectingthe transmission standard are evident. First, theneed for the public and the broadcasters to movetogether in a sequenced transition made the pro-cess a good candidate for government involve-ment. The transmission standard affected infra-structure and. equipment purchase decisions onthe broadcasters' side of the equation and con-sumer purchase decisions on the viewers' side ofthe equation. One group cannot make decisionsuntil the other does, and the industries on thebroadcast side of the equation are easier to organ-ize for input on technical subjects than are view-ers.

Second, the decision on standards involvedcoaxing consumers beyond an investment in a col-lective embedded base that is quite large. Takinginto account the durability of receivers and thenumber of households that own more than onetelevision receiver, it is asking a lot of consumersto make the switch.

Third, the timing urgency created by anotherpublic policy force.made the DTV process a goodcandidate for government participation. Thatpublic policy force was the advent of auctions, andthe tether between the auction process and thebudget process. The proceeds of auctions had tobe factored into federal budget projections; thus,certainty about the timing of auctions was desira-ble. Without agreement about a transmission stan-dard, regulators could make no reliable projec-tion of when the transition to DTV would be

81 See id. at ii.82 FCC Denies Sinclair Petition for Rulemaking on COFDM

Standard, FCC News Release, (Feb. 4, 2000).83 Id.84 DTVFourth Report and Order, 11 FCC Rcd. at 17,772,

para. 2.

2000]

COMMLAW CONSPECTUS

complete and when, therefore, the spectrum cur-rently occupied by broadcasters would be availa-ble for auction.

3. The Case of Wireless Telephony Standards

Occasionally, the need for a government role instandard-setting is measured by the gap left whengovernment declines to play a role. The currentgeneration of digital mobile wireless service in theU.S. ("2G") operates under three standards:

" TDMA (Time Division Multiple Access),which divides a channel into timeslots andsorts user transmissions into these timeslots;

" CDMA (Code Division Multiple Access), aspread spectrum technology; and

" GSM (Global System for Mobile Communica-tions), a European standard for digital cellu-lar system that has gained some currency inthe U.S.8 5

In January 1998, the European Telecommuni-cations Standards Institute ("ETSI") agreed upona Third Generation ("3G") mobile wireless stan-dard that relies on a combination of TDMA- andCDMA-based standards.8a The standard is calledUTRA, an acronym for UMTS (Universal MobileTelecommunications System) Terrestrial RadioAccess.8 7 Wideband-CDMA ("W-CDMA") will beused for wide-area applications and Time Divi-sion-CDMA ("TD-CDMA") will be used for indoorapplications involving low mobility.88 This stan-dard was forwarded to the ITU for considera-tion.8 9

The United States, by contrast, took the stancethat the market can support multiple standardsand that over time the market will pick a predomi-nant standard. Accordingly, the United States for-warded to the ITU four proposals for co-existing3G standards. 9°1

There is a substantial and important role thatthe ITU appropriately can play in evaluating thesestandards. The choice was essentially between aunified standard chosen by ETSI and a menu ofstandards advocated by the United States. The

85 See GSM World, North American 1999 GSM Highlights

(visited June 26, 2000) <www.gsmworld.com/news/latest_news7.html>.

86 See UMTS Report No. 5, Minimum Spectrum Demand perPublic Terrestrial UMTS Operator in Initial Phase <www.umts-fo-rum.org/reports.html>.

87 See id.88 See id.

failure of the system, up until this point, to organ-ize submission and negotiation over national orregional standards is inefficient. The ITU's in-volvement in evaluating and recommending 3Gwireless standards can help mitigate or avoid thistype of diseconomy and inefficiency. The resultsof the recent World Radio Conference, under theauspices of the ITU, represents good progress to-ward a recommendation along the lines of flexi-bility and a technology-neutral multiband ap-proach. - I

This case makes the point that government en-tities can have different roles at different jurisdic-tional levels. The U.S. government's involvementin the process was essentially to take no positiverole in forcing a unification of the competingstandards proffered domestically in the U.S. andto urge the relevant international governmentalorganization, the ITU, to permit multiple stan-dards to develop rather than allow essentiallymandatory regional standards to prevail. Thus, onthe one hand, a national government elected notto become involved domestically, while urging agovernmental entity at a supernational jurisdic-tional level to become involved-but only for thepurpose of preserving competing standards.

The characteristics of this situation make it ripefor government involvement of a limited and atyp-ical nature at the international level. First, the na-ture of government involvement here is not inline with the other examples. The aim of the U.S.government's involvement appears to be an at-tempt to keep the decisional environment opento competing standards. The correctness of thisjudgment is supported by the relatively benign ex-perience of mobile telephone users in the UnitedStates. Although the absence of a single standardhas put some limits on the availability of roaming,the fact that several carriers have built and devel-oped networks with nationwide footprints has mit-igated this problem. Also, the development ofroaming agreements to cover areas with compati-ble networks outside the home carrier's networkhas helped. Thus, the absence of a unified U.S.

89 See generally UMTS Reports (visited July 2, 2000) <www.

umts-forum.org/reports.html>.90 See UMTS Backs 30 Harmony, ELECTRONIC TIMES, June

14, 1999.9 See WRC-2000 Wraps Up Work with Spectrum Deals, De-

spite Last-Minute Wrangling over BSS Replanning, TELECOMM.REI'S. (June 5, 2000) <www.tr.com/tronlone/tr/2000/tr060500/tr060500-04.htm>.

[Vol. 8

Government's Role in Telecomm Standardization

standard for first and second generation wirelessservices has been generally no worse than an in-convenience that has not noticeably retarded de-velopment of service or subscriber growth.

Second, the market to be served is enormous,in contrast with, for example, the public safetywireless community. This affords room for experi-mentation. The downside of "failed" experimentshas relatively less dire consequences for themostly social and commercial users of nonpublicsafety wireless services. Should a competing stan-dard not survive in the marketplace, the costs ofmigrating users to a system that employs a surviv-ing standard are diffuse and not borne directly onthe public budget, in contrast to the example ofpublic safety users.

Third, the benefits of variety and innovationthat generally accompany competition can beachieved in a way that is compatible with improv-ing the availability of roaming. With competingstandards being deployed in the world market-place, the incentive to coordinate can workamong network builders to promote roaming. If asingle standard prevails in one part of the world,by contrast, coordination becomes a take it orleave it proposition-other providers must buildto that standard or abandon the prospect ofroaming. The ubiquitous deployment of networksusing competing standards, at least until the mar-ket identifies which, if any, is superior for variouspurposes, thus produces enhanced incentives forcoordination.

B. Cases Where Government Involvement is ofMinimal or Questionable Benefit

Likewise, there are cases in which governmentinvolvement in telecommunications standard-set-ting provide little benefit. Set-top boxes and cablemodem standards are two examples.

1. The Case of Cable Set-Top Boxes

The transition to DTV has created several issueswith respect to the relationship of the cable indus-try to the broadcast industry. Many of these issues

92 See 47 U.S.C. § 325(b) (1) (1994).93 See id. at § 325(b) (1).94 See Cable Television Consumer Protection & Competi-

tion Act of 1992, Pub. L. No. 102-385, 106 Stat. 1460 (1992)(codified in scattered sections of 47 U.S.C. §§ 521-611(1994)).

go back to a fundamental policy decision thatCongress made in 1984 giving local broadcasterstwo powerful options in dealing with the localcable operator's carriage of the broadcast signal. 92

First, broadcasters had the option of invoking astatutory requirement that the cable operator"must carry" the broadcast signal on the valuablebasic tier of service that all cable subscribers re-ceive. 93 The statutory provision rested on the leg-islative policy judgment that cable would be sopreferred by viewers to over-the-air broadcast thatthe vitality of local broadcast might be threatenedif local broadcast channels were not widely availa-ble via the cable platform. 94

Congress also gave broadcasters a valuable sec-ond option. Broadcasters, as an alternative to in-voking the must carry obligation of cable opera-tors, may require the operator to enter intonegotiations for "retransmission consent." 9 5

These negotiations are intended to lead to com-pensation for the broadcaster in consideration forthe broadcaster's consent for the cable operator'sretransmission of the broadcast signal.

The must carry regime has been upheld by theSupreme Court against cable industry challengesthat it imposed an undue burden on cable opera-tors and violated their First Amendment rights. 9 6

The Court's rationale rested largely on its accept-ance that it was legitimate for Congress to adoptand enforce the must carry regime because of thepossible threat that would be posed to the vitalityand viability of local broadcast. 97 While acquiesc-ing in the Supreme Court's decision upholdinganalog must carry, cable operators have opposedthe way in which some broadcasters have con-strued the law-that operators would have the ob-ligation to carry both the analog and the digitalsignal for the duration of the transition and thatcable operators would have an obligation to trans-mit multiple signals if broadcasters chose to usetheir spectrum in such a way."

The question would not be so weighty but forthe messy edges of the transition. The transitionto DTV will not be complete until the broadcast-ers have ceased broadcasting on the frequenciesnow used for broadcasting analog signals.99 They

95 See 47 U.S.C. § 325(b) (1).96 See Turner Broad. Sys. v. FCC, 520 U.S. 180 (1997).97 See id. at 187-92.98 See generally Harris J. Aaron, Note, I Want My MTV: The

Debate Over Digital Must-Carry, 80 B.U. L. REV. 885 (2000).99 See generally Larry Bloomfield, Cable-Broad. Compatibil-

20001

COMMLAW CONSPECTUS

may have a respite from relinquishing that spec-trum in some markets until a significant percent-age of households have purchased television re-ceivers capable of receiving digital broadcastsignals. Until the cost of the receivers comes downsignificantly, such penetration is unlikely to occur.In various contexts, the FCC has stated its expec-tation that the spectrum will be relinquished by2006, but the statutory trigger is more complexand dependent to some measure upon penetra-tion of receivers.' 00

In the meantime, cable customers that take theplunge and purchase digital receivers will findthat they cannot receive digital signals via thecable platform unless they have purchased a set-top box that is capable of receiving the signal orhave purchased a digital cable ready receiver."" I

Neither of these is generally available on the com-mercial market yet. ' 0 2

The FCC is already deeply involved in commer-cial decisions about cable set-top boxes because ofSection 629 of the Telecommunications Act of1996.103 Section 629 directs the FCC to promul-gate rules that would foster the development of acompetitive commercial market in set-top naviga-tion devices so that cable subscribers could obtainsuch appliances from multiple sources.'1 4 At thesame time, Congress directed the FCC to protectmultichannel video providers' content security.'15In other words, to the extent that the set-top boxplays a role in preventing unauthorized receptionof signals, the FCC was directed by statute to makesure that this function was not sacrificed. 1116 TheFCC's order permitted the integration of the se-

ity-an Oxymoron?, BROAD. ENGINEERING, Apr. 30, 2000 [here-inafter Bloomfield].

100 See generally id.; see also Mark Wigfield, Cable, ElectronicsIndustries Okay Digital-TV Pact, Dow JONES INT'L NEWS, Feb.23, 2000, at 19:30:00 ("All viewers are to have access to digitaltelevision by 2002, and broadcast stations must return theiranalog spectrum license to the FCC by 2006.").

101 See generally Glen Dickson, Back to the Future, BROAD. &CABLE, Aug. 2, 1999.

102 See generally Bloomfield, supra note 99.103 See, e.g., id.104 See 47 U.S.C. § 549(a) (1994) ("The Commission

shall, in consultation with appropriate industry standard-set-ting organizations, adopt regulations to assure the commer-cial availability, to consumers of multichannel video pro-gramming systems, of converter boxes, interactivecommunications equipment, and other equipment used byconsumers to access multichannel video programming andother services.").

105 See 47 U.S.C. § 549(b) ("The Commission shall not-prescribe regulations under subsection (a) which wouldjeop-

curity and navigation functions for a limited timeuntil 2005, after which time it would be unlawfulto integrate both functions in the box.10 7 The pro-hibition is designed to foster competition amongmanufacturers in the navigational device market.

Two fundamental issues that remain unresolvedinvolve the requirements for DTV receivers to be-come cable compatible and licensing arrange-ments for protection against unauthorized copy-ing of cable content. The National Association ofBroadcasters has urged the FCC to step into thebreach, beyond the 'jawboning" in which it al-ready has engaged. The Commission has com-menced a broad proceeding threatening to imple-ment rules if agreement cannot be reached l08

Raising the stakes in this way can be equivalentto actual involvement if the threat of imminenceand inevitability is credible. But it is unnecessaryto achieve the desired result and at odds withother undertakings of the FCC that affect thetransition to DTV.

First, the debate about receiver and set-top boxcompatibility is closely tied to the debate over dig-ital must carry. At the end of the transition, it islikely that cable operators will end up carrying thebroadcasters' primary signals either because theyare forced to do so by the FCC or because theydeem it to be in their own commercial interest. Ineither event, this eventuality will require the cableindustry to be ready at the end of the transitionwith operating boxes in the home. The set-topbox problem is likely to heal itself as a side benefitof the conclusion of the DTV transition. Addi-tional intervention is not useful or needed.

ardize security of multichannel video programming andother services offered over multichannel video systems, orimpede the legal rights of a provider of such services to pre-vent theft of service.").

106 See In re Amendment of Parts 2, 15, 18 and OtherParts of the Commission's Rules to Simplify and Streamlinethe Equipment Authorization Process for Radio FrequencyEquipment, Report and Order, 13 FCC Rcd. 11,415, 11,4416,para. 2 (1998) [hereinafter Streamlining Order]; but see Commis-sion Advances "Navigation Devices" Rules Creating Consumer Mar-ket for Set Top Boxes: Allows Analog Deferral, FCC News Release,May 13, 1999 (visitedJune 26, 2000) <www.fcc.gov/Bureaus/Cable/News_Releases/ 1999/nrcb9009.html> (noting thatthe Commission is considering whether to accelerate thecompliance date to 2003).

107 See In re Implementation of Section 304 of the Tele-communications Act of 1996, Report and Order, 13 FCC Rcd.14,775, 14,776, para. 3 (1998).

108 See In re Compatibility Between Cable Systems andConsumer Electronics Equipment, Notice of Proposed Rulemak-ing, PP Dkt No. 00-67 (rel. Apr. 14, 2000).

[Vol. 8

Government's Role in Telecomm Standardization

Second, one of the greatest risks of sidetrackingthis benefit rests within the control of the FCC it-self. The broadcasters' proposals to enlarge themust carry requirement so that cable operatorswould be required to carry multiple signals duringthe transition or thereafter is a deterrent to reach-ing the point where carriage of the primary signalis acceptable. So far, the agency has not rushed toembrace multiple signal plans, and its continuedresistance supports the right outcome in the set-top box proceeding as well.

2. The Case of Cable Modem Standards

In 1999, a pitched battle came to a head be-tween cable operators and certain Internet Ser-vice Providers over the issue of whether the net-work that enables provisioning of high-speedcable modem data service should be open toothers, notjust cable operators on a wholesale ba-sis to provide customer service. 10 9 The leadingproponent of this viewpoint was AOL, arguingthat the proprietary cable platform would becomea bottleneck. 110 It was critical, they argued, thatregulators establish rules to prevent cable opera-tors from affording exclusive access to the plat-form to favored affiliates, such as Excite@Homeand Roadrunner."' Allies of this position in-cluded the Regional Bell Operating Companies("RBOCs"), which argued that the anti-bottleneckrules that applied to the local loop in telephonyjustly should be applied to the cable plant toachieve parity in the regulatory scheme-or elsethe regulatory burden should be lifted from theRBOCs' provisioning of high-speed data servicesin order to achieve parity.' 12

Proponents of the cable position argued thatcable modem service was nascent and had nomarket share that posed a plausible near-term

. 109 See generally, Earl W. Comstock &John W. Butler, Ac-cess Denied: The FCC's Failure to Implement Open Access to Cable asRequired by the Communications Act, 8 COMMLAw CONSPEcrus 5(2000).

110 See id.III See id.112 See id.113 See id.114 See id.115 See, e.g., America Online, Memorandum of Understand-

ing Between Time Warner, Inc. and America Online (visited May24, 2000) <media.web.aol.com/media/pressview.cfm?>("AOL Time Warner will effectuate such choice for consum-ers by negotiating arm's-length commercial agreements withboth affiliated (such as AOL) and unaffiliated ISPs that wish

threat of becoming a bottleneck.' 13 Opponentsresponded that it was important to fix the rulesbefore a bottleneck was ensconced, which likelywould happen given the attractiveness of cablemodem service and the strong marketing positioncable operators occupied with respect to their ex-isting base of video subscribers."14

Proponents of the cable position argued thatrules regarding cable networks being opened tocompetitors were best left to commercial negotia-tion. 115 In other words, cable operators advocatedthat, rather than having the government declarethat the networks must be open and set, the po-tential contracting parties should be permitted tofind commercially sensible terms for contracts be-tween operators and competitive providers with-out the overhang of government intervention. 6

At one level, the debate became focused onwhat the obstacles to opening the cable networkwere. Some argued that the main reason not toopen the networks was one of fairness: that- thenetworks were built with private investment andwere, as such, private property.' '7 The argumentran that the government must have a very substan-tial justification for demanding that a privateproperty owner share its property, even with com-pensation, with rivals. 1 " Here, there was no suchjustification some operators and their advocatesargued, because there was no evidence of a bottle-neck, in contrast with the substantial historical ex-perience of the local loop for telephony.

Others argued that technical obstacles madeopening the cable network at least difficult, if notimpractical. The argument was that there werefew practical points where interconnection of theincumbent and competing networks could beachieved without sacrificing quality of service orcustomer care, a problem that scaled in propor-tion to the number of competitive providers the

to offer service on the AOL Time Warner broadband cablesystems. Pursuant to such commercial agreements, AOLTime Warner will partner with ISPs to offer consumers achoice of competing broadband internet service offerings.").

116 See In re Applications for Consent to Transfer of Con-trol of Licenses from MediaOne Group, Inc. to AT&T Corp,Application for Transfer of Control and Public Interest Showing, CSDkt. No. 99-251 July 12, 1999).

117 See generally In re Applications for Consent to Transferof Control of Licenses from MediaOne Group, Inc. to AT&TCorp, Reply Comments of AT&T Corp. and MediaOne, Inc., CSDkt. No. 99-251, at 6 (Sept. 17, 1999) (noting, among otherthings, that AT&T has invested over $100 billion in its mar-ket-enktry strategy).

118 See id.

2000]

COMMLAW CONSPECTUS

incumbent had to accommodate."II 9

In the course of the debate came the assertionthat the root of the problem that made openingthe cable network difficult lay in the architecturesand embedded standards cable operators hadchosen to deploy their networks. This portion ofthe debate came to a head when GTE, Com-puserveClassic and AOL announced the results ofa trial they had conducted in Florida to refutecable operators' arguments about the difficulty ofallowing multiple ISPs to have direct access to cus-tomers via the cable platform. 120 GTE's news re-lease said that it had worked with several differentvendors using off the shelf equipment, and it hadmade changes to its own cable modem platformto allow competing ISPs to gain access to custom-ers.121 It summed the one-time required invest-ment as "less than a dollar a home passed."'1 22 Thetrial was criticized by Excite@Home, a principalprovider of cable modem-linked services and por-tals as an incomplete experiment:

Even if traffic can be directed to the proper ISP, thecable infrastructure is a shared network ... Because ofthe nature of the shared network, allocation of capacityis problematic . . . It would effectively turn cable's fatpipe into multiple thinner pipes like DSL[,] . . .[which] is grossly stb-optimal in the cable environ-ment. 123

The vocabulary of the open access debate hashad many variations over its life, many differentways in which the petition for regulatory interven-tion has been voiced. At its common denomina-tor, however, is the notion that the regulatorsought to step in to standardize the way in whichcompeting ISPs could access the architecture ofthe cable modem platform. The arguments havebeen aimed at federal regulators, policy-makersand legislators, and at local officials who superin-tend renewal of the franchise agreements that

119 See GTE News Release, GT7 Demonstrates Ease of CableOpen Access to Multiple ISPs; Clearwater Trial Shows One-Time In-vestment of Less Than $1 Per Home Would Provide ConsumerChoice (June 14, 1999) <www.gte.com/AboutGTE/NewsCenter/News/Releases/ClearwaterOpenAccess.html>.

120 See id.121 See id.122 Id.123 Exite@Home Press Release, The Technical Shortcomings

of the GTE "Open Access" Trial (visited Mar. 30, 2000) <www.home.net/source/techdetails.html>.

124 See OpenNet Coalition Press Release, Maryland, Kan-sas Join Open Access Bandwagon (Feb. 7, 2000) <www.opennetcoalition.org/news/949962651.shtml> (noting that open ac-cess legislation had been introduced in Md., Va., Kan., Pa.,Mich., Ill., Vt., Del. and Idaho).

permit operators to serve subscribers in a given lo-cale.

To date, some local officials have found thesearguments appealing,124 but the FCC has declinedto intervene on the ground that the cable modemservice industry is nascent and should developfree of added regulatory requirements. 25 Itwould take intervention at the federal level to im-pose a comprehensive change in the architecturescommonly deployed in the cable plant, and thereluctance to step in is a sensible result for severalreasons.

First, the evolution of the debate has turned outto validate the proposition that open access couldbe a sensible commercial proposition. After its ac-quisition of Time Warner, for example, AOLstepped back from its vigorous public advocacy ofopen access, having achieved the core of its objec-tives by commercial means.' 26 Other cable opera-tors also have announced commercial decisions toopen their networks.' 27

Second, the thrust of the case for governmentintervention was essentially an invitation to thegovernment to level the playing field between en-tities with different means for bridging the lastmile (cable modem vs. DSL; AT&T vs. GTE), orbetween entities with a means for bridging the lastmile and those without the means for doing so(cable modem vs. no owned infrastructure; AT&Tvs. AOL). A comparison is made to past congres-sional actions: In the Telecommunications Act of1996 Congress determined that it was appropriateto require the RBOCs to open their networks tocompetitors. 28 But that judgment rested on overa decade's observation that the local exchangehad become a bottleneck facility and without an

125 See FCC Chairman William E. Kennard, RemarksBefore the National Cable Television Association (June 15,1999) <www.fcc.gov/Speeches/Kennard/spwek921.html>.

126 See Peter S. Goodman & Craig Timberg, AOL Ends ItsPush for Open Access, WASH. POST, Feb. 12, 2000, at Al, 11; seealso Kathy Chen, AOL Changes Tune in Debate On Cable Access,WALL ST. J., Feb. 14, 2000, at B1.

127 See Letter from David N. Baker, et al. to FCC Chair-man Kennard (Dec. 6, 1999) <www.att.com/press/item/0,1354,2331,00>; see also AT&T Press Release, AT&T Commitsto Give Consumers Choice of ISPs for High-Speed Internet Access ViaCable, Fixed Wireless (visited May 15, 2000) <www.att.com/press/item/0, 1354,2320,00>; AT&T Press Release, AT&TCommits to Open Cable Access (visited May 15, 2000) <www.clec-planet.com/news/9912/991206att.htm>.

128 See, e.g.,18U.S.C.§ 251 (2000);18U.S.C.§ 271 (2000).

[Vol. 8

Government's Role in Telecomm Standardization

open access obligation, competition was not possi-ble.

V. CONCLUSION

Many aspects of the telecommunications indus-try, even as various segments become more com-petitive and less regulated, still have significant re-lationships with industry regulators and policy-makers at the federal level. This may be becausethe companies are themselves still significantlyregulated, are spin-offs of regulated enterprises,or because they rely on regulators and policy-mak-ers to police the behavior of their competitors.

In general, the case studies demonstrate thatthe most hopeful case for a productive govern-ment role is where the market is least likely toprovide a timely and useful answer, as in the caseof public safety spectrum standards. The leastlikely cases for a productive government role arethose in which the government is being called inby parties in interest to substitute for commercialnegotiation, as in the case of cable modem stan-dards and "open access." In between these twopoles are cases that have mixed characteristicsthat will make government involvement a closequestion. Overall, government should be reluc-

Because the industry and federal regulators andpolicy-makers continue to see each other andwork together so closely, there are temptationsand urges on both sides to involve government instandard-setting projects of all sorts. Some ofthese could just as well be left to the market toestablish the standard; some actually merit a gov-ernment role. Weighing the costs and benefits ofgovernment taking such a role requires significantexercises in judgment. A checklist approach willnot capture the nuances of that judgment. Study-ing and enumerating the salient characteristics ofthose processes, however, may avoid wasted timeand effort.

tant to get involved in setting a standard andshould establish a high bar for involvement on itsown initiative or on the motion of interested par-ties. Part of the test should always be whether gov-ernment can play a role within its competenceand whether it can achieve a quicker, sounder re-sult than the market left alone. These casesshould be few and far between. But that general-ization should not become a wooden rule thatkeeps government out of setting standards wherethe benefits are readily foreseen. Discerning suchcases will always be a matter of keen judgment.

Characteristics Suggesting Government Involvement Characteristics Suggesting Government InvolvementMight be Beneficial Might Not be Beneficial

" Atomized or comparatively small market for users of products * Prime driver of petition for government involvement is settlingthat will incorporate the standard once adopted inter-industry dispute

" Need to coordinate U.S. standard with foreign countries or * Decision sought would require imposing new substantiveforeign manufacturers regulations

" Identifiable public policy externality that makes unified standard * Redundant or at odds with other public policy endeavorsdesirable and makes timing urgent

" Significant prior work on the part of affected industries and * Resolvable by commercial agreement between or amongconstituencies; government not starting from scratch contending parties that will not violate antitrust laws.

20001