mr ian alexanderbookstore.teri.res.in/docs/journals/regu-prelims 3-1 june...private bag rondebosch...

Assistant EditorK P EashwarTERINew Delhi, India

EditorLeena SrivastavaTERINew Delhi, India

��

Associate EditorMeeta MehraTERINew Delhi, India

Mr Ian AlexanderSouth Asia Energy and

Infrastructure UnitRoom MC10-749The World Bank, Singapore

Mr Ashley C BrownExecutive DirectorHarvard Electricity Policy GroupKennedy School of GovernmentHarvard University, USA

Prof. Anton EberhardInfrastructure Industries Reform

and RegulationManagement ProgrammeGraduate School of BusinessUniversity of Cape TownPrivate Bag Rondebosch 7701Cape Town, South Africa

Michelle Michot Foss, Ph.D.Executive DirectorInstitute for Energy, Law &EnterpriseUniversity of HoustonLaw Center, 334 Melcher HallHouston, TX 77240-6283

Prof. M Nurul IslamProfessorBangladesh University of

Engineering and TechnologyInstitute of Appropriate

TechnologyBangladesh

Dr Raymond LawtonDirectorThe National Regulatory

Research InstituteThe Ohio State University, OhioUSA

Prof. Mohan MunasingheChief Energy Advisor to the

Govt. of Sri Lanka10 De Fonseka Place, Colombo 5Sri Lanka

Dr R K PachauriDirector-GeneralTERI, India

Prof. S L RaoChairmanForum of Indian Regulators, India

Prof. Dr Lineu Belico dos ReisEscola Politécnica da Universidade

de São Paulo - PEA USPAv. Prof. Luciano GualbertoBrazil

Prof. Rohan SamarajivaVisiting Professor of Economics

of InfrastructuresFaculteit TBM, TechnischeUniversiteit Delft, The Netherlands

Dr E A S SarmaPrincipalAdministrative Staff College of IndiaIndia

Mr S SundarDistinguished FellowTERI, India

Mr Steve ThomasSenior FellowPublic Service International

Research UnitSchool of Computing and

Mathematical SciencesUniversity of Greenwich, UK

��

Regulators, policy-makers, and the making of policy:who does what and when do they do it?Ashley Brown . . . 1

Trends in the management of regulation: a comparisonof energy regulators in member countries of the Organi-sation for Economic Development and Co-operationCarlos Ocana . . . 13

Private sector participation and the poor: realizing thefull potential of transactions in the water sectorShane Rosenthal and Ian Alexander . . . 33

BOOK REVIEWS

A Good Beginning but Challenges Galore: a survey basedstudy of resources, transparency and public participationin electricity regulatory commissions in IndiaPrayasP Abraham . . . 59

Networking Knowledge for Information Societies: insti-tutions and interventionRobin Mansell, Rohan Samarajiva, and Amy MahanR R N Prasad . . . 63

Why is I J R G published? How is its purpose being fulfilled?A biannual publication, I J R G (International Journal of Regulation and Gov-ernance) is aimed at decision-makers, planners, consultants, corporate execu-tives, and researchers. The journal provides a forum for detailed andcomprehensive investigation, analysis, and review of such sectors as energy,telecommunications, water, environment, and transport. The papers in thisjournal will address regulatory and governance issues related to tariff, demandforecasting, competition, institutional and transitional market structure, etc.Only original articles will be published.

How often is I J R G published?I J R G is published twice a year, in June and December. The two issues makeup a volume.

How to subscribe and how much does it cost?A year’s subscription to I J R G costs Rs 1000 within India and $30 overseas atcurrent rates. Payment can be made by demand draft or local cheque drawn infavour of TERI, payable at New Delhi. We also accept payment byMasterCard, Visa, American Express, and Diners Club cards.

Whom to contact for more information?For information on editorial content, please contact

The Editor <[email protected]>, I J R G

For enquiries about subscriptions and advertisements, please contactInformation Dissemination Services

T E R IDarbari Seth BlockHabitat PlaceLodhi RoadNew Delhi –

Information for authorsSee pp. 75–76

© The Energy and Resources Institute 2003

Fax or E-mail [email protected]

Tel. or India + • Delhi ()

Web www.teriin.org

��

��

An international biannual journal, the International Journal of Regulationand Governance has been published by The Energy and Resources Institute,New Delhi.

SubmissionsThree good-quality hard copies of the manuscripts have to be submitted to theEditor. Manuscripts can be submitted electronically also.

Presentation of manuscriptsManuscripts must be in English and the length of original articles should notexceed 6000 words. They must be typed on one side, only on internationalstandard A4 size paper, with a left-hand margin of 40 mm. The main textshould be double-spaced with headings and sub-headings clearly indicatedand placed on the left-hand side of the text. All tables, figures, and equationsshould be numbered with Arabic numerals and the measurements should begiven in metric (SI) units. The manuscript should be arranged in the ordergiven below.1 Short title (10 words is the desired maximum length), subtitle (if desired)2 Author’s name, affiliation, full postal address, and e-mail, telephone, and

fax numbers (respective affiliations and addresses for co-authors shouldalso be clearly indicated)

3 Abstract (not exceeding 200 words)4 Main body of the text, suitably divided under headings5 Acknowledgements6 References7 Appendices/annexures (each on a separate sheet)8 Tables (each on a separate sheet)9 Figures (each on a separate sheet)

Shorter itemsThe following short items are also welcome and must be typed in the same wayas major papers.� Viewpoints (research notes and short communications) and case studies

(maximum 1500 words)� Book reviews (maximum 1000 words)� Comments to editors

ReferencesIn the text, the surname of the author(s) followed by the year of publication ofthe reference is to be given, e.g. (Hall 1993). Where there are several publica-tions by the same author(s) in the same year, use notations ‘1993a’, ‘1993b’,etc. Up to three authors can be mentioned in text references; four or moreauthors should be shortened to the first three authors’ names followed by et al.References must be listed alphabetically at the end of the paper (doublespaced) and should conform to the following style.

For journals: Davis G R. 1990. Energy for planet earth. Scientific American263(3): 55–62.For books: Carmichael J B and Strzepek K M. 1987. Industrial Water Use andTreatment Practices. London: Cassell Tycooly.For chapters of edited books: Sintak Y. 1992. Models and projections of energyuse in the Soviet Union. pp. 1–53. In International Energy Economics, edited byT Steiner. London: Chapman and Hall.

For grey literature: Togeby M and Jacobsen U. 1996. How conflicting goalsconcerning environment and transport influence the policy process? Paperpresented at the Conference on Transport, Energy and Environment, 3–4October, Helsingor, Denmark.

FootnotesFootnotes should be minimized, and where unavoidable, should be indicatedin the text by superior Arabic numerals, which run consecutively through thepaper. They should be grouped in order of appearance at the bottom of theconcerned page in numerical order and must be double-spaced.

General� Responsibility for the contents of the paper rests upon the authors, not

upon the editor or the publisher.� Papers are accepted for refereeing on the understanding that they have

been submitted only to this journal and to no other journal.� The author must not publish the same manuscript subsequently in another

journal. However, selected portions of the published paper may be repro-duced, provided that written approval has been received from the publisher.

Electronic manuscriptsOn acceptance, contributors are requested to supply the revised manuscript inelectronic format, on computer disc (3.5 inches) containing the final versionof the paper as well as the final hard copy. Make sure that the disc and hardcopy match exactly. Please observe the following criteria.� Specify what software was used, including which release, e.g. WordPerfect 6.1.� Specify what computer was used (either IBM compatible PC or Apple

Macintosh).� Include the text file (saved in double spacing) and separate the table, fig-

ures, and illustration files.� Keep a back-up disc for reference and safety.

CopyrightAll authors must sign the Transfer of Copyright agreement before the paper canbe published by TERI. This transfer agreement enables TERI to protect the copy-righted material for the authors, but does not affect the authors’ proprietaryrights. The copyright transfer covers the exclusive rights to reproduce and distrib-ute the papers, including reprints, photographic reproductions or any other repro-ductions of similar nature and translations, and includes the right to adapt thepaper for use in conjunction with computer systems and programs, including re-production or publication in machine-readable form and incorporation in re-trieval systems. Authors are responsible for obtaining, from the copyright holder,permission to reproduce any figures for which copyright exists.

ProofsOne set of proofs will be sent to the author before publication, which shouldbe returned promptly within 48 hours of receipt. The publishers reserve theright to charge for any changes made at the proof stage (other than printer’serrors) as the insertion or deletion of a single word may necessitate the reset-ting of whole paragraphs. Please note that authors are urged to check theirproofs carefully before return, since late corrections cannot be accepted.

Off-printsApart from one free copy of the journal to the authors, 10 free off-prints willbe supplied to the first author. A copy of the paper in PDF (portable docu-ment format) will also be sent to the author.

��

��

�� !��"#�$�� $��

!��"� %�� %�� %�� " �� &

'��" (�%�)*�� +�� ,�� )��" -��" ��! .��" /��

�� ,�� 0��" 0/'1 2� �� 3�(�� 3�4 ��

�� 5��

��

��6�� !�� " �� #�� 4�� " �� " �� !�� !��"#�$�� 7��!��#��8 ��9:6�� !�� " �� " ��% �� 4�� % ��!�� 4�� $ �� " �� !��" ��9:6�� ;�� 4��%�� !�� !��"9:6�� ;�� %�� " ��9:<� � ��" �� !�� *!�� 4��" �� 4� !�� !��" =��!�� !�� %�� !��!� 4�� %� �� !��"#�$�� 9:6�� !��"#�$�� %�� 4�� 9 >�� %�"�� !��4��" �� !��" �� 4�� !��" �� !��" ��9 +�� $�� !��;�� !��#4��" �� !��" �� #�$��9

'��" (�%��

��

��

Delineating between the roles of government policy-makers andindependent regulators is the subject of controversy and confu-sion wherever independent regulatory agencies have been estab-lished. Part of the controversy, of course, is the result of thenatural ‘shaking out’ process for newly established independentregulatory agencies in countries with no experience regardingthese institutions. Part of the controversy, however, is simplythat the boundaries between ‘policy-making’ and ‘regulating’ areinherently fluid and uncertain. Moreover, the very notion of dis-tinguishing between ‘policy-making’ and ‘regulating’ may wellpose a false dichotomy. Both policy-makers and regulators makepolicy. The distinction is that policy-makers define the funda-mentals and define the parameters within which policy-makingis delegated to regulators. It is more useful to think, not in termsof policy-making versus regulation, but, rather, as macro policy-making versus micro. For purposes of this paper, macro policy isdefined as the basic policy parameters. It constitutes thefundamental principles of the regulatory system. Micro policy-making is defined as the action of regulators, operating withintheir legal authority and consistent with the macro policiesformally enacted by the government, to apply, clarify, interpret,and fill in details left unspecified by macro policy-makers.

In flushing out the distinction, it is useful to keep keyconcepts in mind.1 Basic and macro policy must be set by the government2 Government policy must be set and altered only on a

prospective basis3 Regulators must follow and enforce policies articulated by

the government4 Regulators are creatures of the state and not necessarily of

the government5 Policy vacuums are inherent and expected6 Some policy issues require technical expertise to be resolved7 Regulatory decision-making, policy or otherwise must be

subject to appellate review.

��

The government has the power and the obligation to set basicpolicy. It not only has the capability, but also takes the action(s)that provides the regulatory regime with its legitimacy, credibi-lity, and legal authority. In fact, except perhaps, in the rare

�� !��"#�$�� $�� !��" �

��

circumstances where the regulatory agency, as in California, iscreated in the Constitution itself, regulators possess only thosepowers specifically delegated to them by the government.Governmental failure to coherently articulate basic policy willinevitably lead to instability, uncertainty, and blurred vision.Neither investors nor consumers will tolerate for long a regula-tory regime without basic form. The real question about thegovernment establishment of policy is about the level of detailprovided by government policy-makers, stability of the estab-lished policy, and the means by which policy is articulated andcommunicated.

The level of detail provided by the government is not a trivialissue. It is necessary that policy be articulated in sufficient detailto provide a level of stability and predictability adequate toattract capital and market participation. The general rules andparameters for discretion delegated to regulators need to bestated in sufficient detail to enable a general understanding ofthe nature of the regulatory regime. Indeed, it is in articulatingthe basic policies that the difference between macro and micropolicy is defined. Anything articulated in law or rule by thegovernment constitutes a macro policy. Any policy that regula-tors articulate in order to carry out their duties to implementmacro policy constitutes micro policy.

��

Macro policy should not be overly detailed for two basic reasons.The first being that regulators should be allowed the flexibilityneeded to adjust to inevitably changing circumstances. Marketsand circumstances evolve with time and it is prudent to enableregulators to make appropriate incremental changes. Thatdegree of flexibility internalizes modest changes into the regu-latory process and avoids undue politicization of issues oflesser magnitude. It is also a recognition that policy-makers arenot and cannot be prescient. It is not possible for them toanticipate all issues that require policy-making to resolve.Rather than attempting to micro manage all details, delegationof authority to regulators to fill in policy details seems sensible,particularly as government policy-makers always possess theultimate authority to change policy on a prospective basis,when they deem it appropriate to do so.

The second reason for avoiding overly prescriptive policyparameters is that some matters are too technical for policy-makers.

'��" (�%��

��

An excellent example is in the area of pricing. While it is impor-tant that the basic methodology be set forth on a policy level,the actual implementation and application of pricing principlesis an extraordinarily complicated matter. What level of exper-tise, for example, can we expect to find in a legislative body onthe relative merits of locational marginal cost pricing for elec-tric transmission services? The matter, while an important sec-tor policy issue, is self evidently too arcane, too technical, andtoo complex to expect keen insights from macro policy-makers.That being said, however, it is critical that the governmentarticulate at least a basic theory of pricing. It may range fromthe amorphous ‘just and reasonable’ standard enunciated inthe Federal Power Act in the US, to something slightly moreprescriptive, such as mandating price caps, benchmarks, rate ofreturn regulation, PBR (performance/incentive-based regula-tion), reasonable opportunity to recover prudently incurredcosts, or some other criteria. Even here the task is not easy. Forexample, enunciating PBR goals is much more general thandescribing the narrower framework of price caps. Further, it isimportant to consider the general consistency of the guidance.For example, the proposed US energy legislation currentlyunder Congressional consideration gets into details such as‘native’ load protection that goes against the open access provi-sions in unexpected ways. Very small changes in the wordingmay have profound, unintended, and often quite adverseeffects. The purpose is to provide investors and consumersalike, some insight into what they may reasonably expect fromthe pricing regime, not to put regulators into a strait jacket byrigidly defining every detail. Where policy-making requirestechnical expertise, nuanced shaping, and expertise, it isprudent to delegate it to the regulators.

��

Delegation of micro policy-making to a regulatory authoritymakes sense because no macro policy-maker, regardless ofprudence and vision, will ever be able to foresee all policyissues that might be encountered in practice. Consequently,there is an element of policy-making that will have to be donewhen unanticipated issues arise for which there is no pre-exist-ing policy, or where the policies, articulated in broad terms,requires clarification or fuller definition in application. Exam-ples include refined definitions of what constitutes improper

�� !��"#�$�� $�� !��" �

��

exercise of market power in electricity generation, or how toprice a newly unbundled telecommunications service that hadpreviously only been offered in a bundled basis with other serv-ices, or redefinition of customer classes based on unforeseenuses. It is theoretically possible that regulators, upon encoun-tering such a situation, could stop their decision-making proc-ess and seek guidance from government policy-makers.Unfortunately, doing so will almost certainly render the deci-sion-making process more labourious, time-consuming, andless effective. Moreover, there is no assurance that there will bean adequate response, much less a timely one. Certainly, legis-lative bodies are not famous for either timeliness or precision.It seems both more efficient and fairer to the parties involvedto simply authorize the regulators to make the neededdeterminations. If the judgement of the regulators provesfaulty, there will be many opportunities for them to reversethemselves, or for macro policy-makers to step in and articu-late a new policy on a going forward basis.

��

It is useful to point out that macro policy can come from twosources—one legislative and the other executive. Obviously,basic policy should be set out in law. That requires legislativeaction. The other possibility for policy formulation, all withinthe scope of authority provided by law, is that executive agen-cies such as cabinets, individual ministries, councils of minis-tries (for example, The National Energy Policy Council inBrazil), the president or prime minister himself, will enunciatethe policy. Basic infrastructure ministries, and perhaps otherinstitutions, may possess comparable levels of expertise as isfound in regulatory agencies. They, therefore, may well be ascompetent at analysing arcane technical matters as the regula-tors. The issue with executive policy-makers, unlike legislators,is often not the lack of understanding or expertise, but, rather,one of timing, transparency, politicization, and application ofdecisions. It is important, however, to keep in mind that there ismore than one level of delegation possible for micro policy-making.

While broad policy questions should be resolved by policy-makers, many areas of micro policy-making, within definedparameters, are best delegated to regulators. Doing so followslogically from one of the fundamental reasons for regulatory

'��" (�%��

��

independence. The state performs three basic categories offunctions: administration, legislation, and adjudication. It isimpossible to put regulatory agencies in any single category asthey perform administrative, legislative, and judicial tasks. Theyoperate agencies, buy supplies, enforce laws, manage personnel,and perform other administrative tasks. They set tariffs, promul-gate rules, enunciate micro policy within the authority delegatedto them, and perform other functions which are universallyapplicable and prospective in nature. Those two attributes areclassic legislative powers. Finally, they adjudicate disputeswithin their legal jurisdiction. Thus, regulators do not readilyfit into any governmental table of organization. Policy-makingis legislative in nature and is, therefore, a type of activity inwhich regulators routinely engage. Their ability to do so,however, is governed by the scope of authority granted to themby the government. Once that authority is delegated, and, untilit is rescinded, regulators should be free, subject to appellatereview, to apply their expertise and exercise their lawful author-ity free of governmental interference.

Macro policy-makers always possess the legal capability todictate policy to regulators. It is important, however, that whenthey do so, they act only on a prospective basis. The rationale forthat principle is twofold, decision-making coherence, and thelegitimacy/transparency of the process itself. The first rationaleis rooted in sound process management which has three basicelements.� Legal/macro policy formulation and articulation� Implementation/micro policy formulation� Appellate review.

It is an element of basic fairness that those who participate inthe process are able, to the extent possible, to know the rules andpolicies with which they will have to comply. It is, therefore, forthe sake of both coherence and fairness that the three elementsof decision-making be conducted in appropriate sequence bythe proper authorities.

Policy-makers, both legislative and executive, need to provideregulators with the policy framework within which they mustmake their decisions. By articulating that framework, theysimultaneously provide all parties due notice of the basic pa-rameters of regulatory policy and principles to be followed. Those

�� !��"#�$�� $�� !��" �

��

policies are set forth in general terms and in contemplation of theoverall objectives rather than determining the outcome ofspecific cases or fates of specific market participants. Whilevested interests will undoubtedly attempt to influence policydecisions, and certainly have a right to do so, it is important tokeep policy-makers fully focused on the broad goals and objec-tives defining the public interest, rather than on the specifics ofindividual case outcomes.

It is for the regulators to decide individual cases and to actu-ally apply the policies to specific factual contexts and players.In doing so, they will encounter matters that require detailedinterpretation of policy, or even fill in the blanks left by thepolicy-makers. In fact, for the most part, it is in the context ofspecific cases or set of circumstances that issues of micro policywill arise. It is an inherent and unavoidable aspect of regulationthat matters of micro policy, or clarification of broad policy, willarise in specific cases before the regulators. Whereas macropolicy-makers are often initiators of policy matters, regulators,more often than not, make micro policy in reaction to mattersraised in specific cases or disputes, or, in order to specificallyfulfill obligations imposed upon them by law. It is axiomatic,but true, that unforeseen issues or circumstances will arise,which the macro policy-makers did not, or could not,anticipate.

While regulators could, in theory, upon encountering a mi-cro policy matter, stop the process, throw up their hands, andask for guidance from government or legislative authoritiesbefore proceeding, the result, would be likely be highly disrup-tive, time consuming, and would almost certainly politicize theoutcome of very specific cases or the fulfillment of specificregulatory objectives. Those inevitable effects of such a proce-dure would likely negate the very raison d’etre of independentregulatory agencies. It makes better sense, therefore, to simplyallow the regulators to proceed with their decision-makingprocess. That being said, however, there needs to be a check inplace to assure that the regulators neither exceed their legalauthority nor violate policies that they are obliged to follow.That, of course, is the reason why there is an appellate process.If regulators, in deciding a matter, fail to follow obligatory lawsand/or policies or adhere to required processes, then theoffending decision should be reversed and reconsidered.

'��" (�%��

��

� !�� !��

There is, therefore, a logical sequence to deciding regulatorymatters. The first is the initiation of the entire regulatory proc-ess through the articulation of basic principles and policyformulations. It allows for public contemplation of basic poli-cies through the political process, but in a broad context with-out reference to specific cases or disputes. That is, undisputedly,the role of legislators, and perhaps executive policy-makers as well.Regulators can, and perhaps should, provide input to suchmatters, but are not empowered to take a decision.

The second part of the sequence is the implementation ofregulation. That process allows regulators to adjudicate dis-putes, fulfill legal obligations such as tariff setting, and, wherenecessary, to provide micro policy details in order to clarify orprovide detail on policy. The latter, of course, is the essence ofmaking micro policy. It must be carried out independently,transparently, and in an apolitical manner.

The third sequence is to assure that the second sequence, theregulatory process, is carried out in a manner not inconsistentwith policies and principle enunciated in the first process. Thethird sequence is, of course, the appellate process. In fact, thereare two appeal processes, one for resolving specific cases in dis-pute, and the other, for resolving policy issues on a prospective,going forward, basis. In the first type of appeal, a party who feelsaggrieved by a decision by the regulator may ask that an appel-late body (usually a court or a tribunal of some sort) reverse thedecision in that case. The appellate body, among its other obliga-tions in reviewing the decision of a regulatory agency, mustmake certain that the regulators neither exceed their authoritynor fail to follow polices and processes set by macro policy-makers. It is important to note, however, that this type of reviewis not to formulate new policy, but merely to assure compliancewith existing laws and policies. This form of appeal should becarried out in an independent, transparent, and apolitical man-ner. The other form of appeal to macro policy-makers, however,is merely to review relevant policy in order to determine whethera policy needs to be altered or supplemented. Because such anappeal can be carried out within the political process, any policydeterminations will affect only future matters. In other words, itcannot affect the outcome of specific cases decided by the regu-lators prior to the re-formulation of basic policy.

�� !��"#�$�� $�� !��" �

��

��

Apart from sound principles of decision-making, there isanother even more important reason for allowing regulators todecide matters of micro policy. That reason is the transparencyand integrity of the decision-making process itself. The integ-rity of the regulatory process is rooted in many elements, butimportant among them is the idea that the process be transpar-ent, fair, and independent of politics. As one observer noted,regulators are agents of the state, and not necessarily of thegovernment at the moment. In order to assure the integrity ofdecision-making, it is vital that the process is exactly as itappears to be. All parties have equal opportunity to access deci-sion-makers and to know what information and arguments theregulators consider while rendering their decisions.

The making of micro policy often arises in connection withindividual cases involving specific and discrete financial inter-ests, the process, like the judicial process, must be absolutelytransparent and, to the extent possible, divorced from politics.Investors see greater predictability, more dispassionate analysis,and fewer risk variables in the regulatory arena than in a politicalone. Similarly, consumers in many places have come to the sameconclusion, namely that they are better served by having anindependent, transparent, apolitical body making key decisionsregarding infrastructure than having case-specific mattersresolved in a political forum where they are likely to possessless clout than well-funded lobbyists from large companies.The views of political figures may well be considered by regula-tors as one set of inputs. Such views, however, must be communi-cated in a transparent, public manner. It should, however, be theregulators, alone, who are responsible for actual decision-making.In short, the process must be internally open and complete.

��

Unlike the making of macro policy, which is inherently political,the regulatory process should be free of politics to the extentpossible, because it usually involves weighing the interests ofspecific parties, and making technical judgments regarding theapplication of broad policy to a specific set of circumstances. Itis, therefore, inconsistent with the very basic regulatory concepts ofindependence, transparency, and depoliticization for regulatorsto defer to political authorities in rendering their decisions.

'��" (�%�

��

It is theoretically possible to construct a relatively trans-parent mechanism for political consultation by regulators onmatters of micro policy. Indeed, political authorities shouldalways have a means of transparently offering their views toregulators. The problem is not the transparent offering of view-points, but, rather, the non-transparent bypass of the regula-tory processes that seems likely to occur if regulators are not ina position to decide micro policy issues on their own. Partiesseeking to advance their own interests will almost inevitably,whenever it suits their interest, seek out political officials tosupport their point of view. It would, for example, be grosslyunfair to have all of the parties in a case present their evidenceand arguments to the regulators through the prescribed proc-ess while another party to the same proceeding seeks out theclandestine support of a minister or other high political figurein order to secure a favourable decision. Success in such amanoeuver would render the entire regulatory process in thatproceeding a sham. All of the evidence offered, argumentsmade, processes followed would be meaningless. It is for thatvery reason that independence of the regulators is, in fact, acritical element of transparency. No process can be deemed tobe transparent when the real decision-maker is someone otherthan whom it is supposed to be under the procedures, or, wherethe real reasons for a decision remain unrevealed.

��

While perhaps it cannot be said that the motives of regulatorsare always pure, the discipline imposed by the process can atleast compel transparency. The same cannot be said when theprocess becomes politicized. While the motives of the govern-ment in interfering may well be for such legitimate policy rea-sons as controlling inflation, promoting investment, promotingspecific resources, the opportunity of bypassing an established,transparent regulatory process by political officials also opensthe door to politicization, corruption and/or de-legitimization.It is important, therefore, as elementary fairness to all parties,for the integrity of the process, and for transparency that theregulators make the decisions themselves, and that any effortby the government or any of its officials to influence theoutcome only be carried out in ways that are open and trans-parent. Certainly, advocating legitimate goals can be donetransparently without embarrassment. More importantly, if the

�� !��"#�$�� $�� !��"

��

goals being advocated by political authorities are meritorious,then the government is always empowered to change policiesprospectively. It need not intervene in the regulatory process inspecific cases in order to effectuate policy. Doing so is to effec-tively alter the rules in the middle of the game. By makingpolicy on a prospective basis only, the integrity of the process ispreserved without sacrificing the ability of political authoritiesto make a policy.

"��

In conclusion, governments must set basic policy, macro policy-making. Filling in the details of that policy, micro policy-making, however, is an inherent part of what regulators have todo in order to carry out their mission. Policy-making by regula-tors, however, is restricted by two critical factors. The first isthat policy made by regulators is subsidiary to governmentpolicy and is done only under a delegation of authority fromthe state. Secondly, policy-making by regulators is incidental toand inherent in their duty to decide specific cases or disputes.That policy-making role is derived entirely from the fact thatmacro policy cannot be reasonably expected to anticipate allaspects of policy that will evolve for the regulatory process to befully functional. Gaps will have to be filled in and it is the regu-lators, with technical expertise and hands-on experience, thoseare best positioned to accomplish that. Their role in doing so,however, is subject to two checks.

The first is the appellate review that determines if the regula-tors were acting within their lawful authority, followed policiesthey were obliged to follow, whether they were acting reason-ably, and whether they followed fair and correct procedures. Thesecond check is that the government retains the ability to altermicro policy determinations. In order to safeguard the integrityof the regulatory process, however, it is vital that that power beexercised only on a prospective basis. Recognition of the realitiesand limits of regulatory policy-making will safeguard the processand allow for a more orderly, transparent, and predictable regu-lation, both in terms of process and substance.

6��

� ��!�� " �� 4�

�� 2�� )��

+��!�� #�!��

�� 2��/'5�' �� /�� )�� (�� 0��" ��

?��;� /!��

�� @�@A

��6�� !�!� �� " ��"�� 4�� #��" ��"�� " �� 2)�+ �2�� )�� +��!�� #�!��4� ��9 6�� %�� !��;�� " �� !�� " ��;�� " �� *�#��9 ��" ��;�� %�� % �� #�� !�� !�%�� ! %�� *�� 4�� 9 6�� !�!� �� %� B��9

<�� %�� !�� " �� &�� 4�� %� �� !�� " �� 4��!�� #�$�� #�� !�� 9 6�� #�� 4��$�� 4�� !�� !�!�� !�� " �� %�� " �*��9

/�� %�� *!�� !�%�� #!�� " ��& 6�� "�� " ��

1 A first draft of this paper was prepared while the author was with the Energy Diversi-fication Division of the International Energy Agency. Comments and suggestions fromPeter Fraser, Robert Priddle, and Caroline Varley are gratefully acknowledged. Theviews expressed in the paper are those of the author.

�� 2��

�� @�@A

�� 4�� " ��%�$��!�� "9 ��" �� " �� !�%� �� !�� !��#�� !�� B�� 4�� %�$� �� " !�� #%�$ !��9 '� �� *�� " �� *�� !�� 7��#��8 ��9 �� %�� " ��;�� *�� !� �� 9

6��

�� @�@A

��

Regulatory reforms bring with them substantial changes in themanagement of regulation, often including the creation of newregulatory agencies and a new distribution of functions andresponsibilities among regulators, ‘line ministries’, competitionauthorities, and other parts of the government. Regulatoryagencies in energy, telecommunications, and other industries,once only present in Canada and the US, were established inmany other countries as an integral part of the reform. Theorganization, decision-making structures, and processes of thenew regulatory agencies are often similar to those of tradi-tional utilities commissions and their federal counterparts inthe US. However, their role has been redefined and extendedto areas related to the introduction and advocacy of competi-tion, and the rationale for independence and neutrality ofdecision-making has also evolved with the advent of competi-tion. This paper studies recent trends in the organization, deci-sion-making structure, and functions of regulatory agenciesbuilding on a comparative analysis of energy regulators in 23OECD (Organisation for Economic Development and Co-op-eration) countries.

��

There is a global trend towards specialization of the regulatoryfunction. Specialization is implemented with two differentapproaches to the management of regulation. In one group,comprising 10 countries, there are independent regulatoryagencies with concrete regulatory powers. In another group ofcountries, the ‘line ministry’ (within the executive) retains allregulatory powers but delegates the management of someregulatory functions to a ministerial agency in five of the casesexamined and to an independent advisory agency in another fourcases. These bodies act in a manner similar to an independentregulator but have no final regulatory powers. Only five coun-tries have no specialized regulatory organization apart fromthe line ministry and this occurs in countries with virtually noongoing regulation of the industry or with minimal marketopening.

��

There is convergence in the design of independent regulatoryagencies. The independent regulatory agencies examined in this

�� 2��

�� @�@A

paper share a number of key elements regarding their corefunctions, decision-making structure, procedures, and inde-pendence safeguards. Kerf, Schiffler, and Torres (2001) simi-larly conclude that there is significant convergence acrossEuropean Telecommunications regulators. However, it is alsodemonstrated that the powers of these agencies vary. Someagencies work under a general mandate to regulate the indus-try while the functions of others are more restricted. The analy-sis suggests that differences in national approaches are partlyexplained by the regulatory framework adopted in each coun-try. Where the scope of regulatory functions is larger, regula-tors are inclined to have more power and independence. Onthe other hand, where regulatory functions and competitionpolicies are less developed, regulatory institutions tend to haveless power. There is consistency between the design of regulatoryagencies and the scope of regulation.

��

There are two alternative views on organizational changes thatare currently taking place in the regulatory function. One viewemphasizes convergence by highlighting that many independentregulatory agencies have been established in the electricity, gas,telecommunications and, to a lesser extent, water industries.This trend is observed in the OECD countries—Latin America,Eastern Europe, and some Asian countries such as India. Therapid expansion of regulatory agencies suggests that interna-tional benchmarking and the identification of best practicescould be effective tools in the management of independent regu-latory agencies. Berg (2000) provides a comprehensive review of‘best practice’ design principles for utility regulators. The otherview emphasizes diversity by highlighting the large variety ofapproaches to the design of regulatory agencies and other regu-latory bodies, including a number of countries where no regula-tory agency has been established. The power and independenceof regulators seem to differ across countries, particularly in sec-tors like energy where reforms are very recent. This would implythat the organization and management of regulation are largelycountry-specific issues, linked to the legal and administrativetradition of each country. Benchmarking and identifying bestpractices would not be effective in this context.

The evidence examined in this paper lends support, withsome qualifications, to the view that regulatory institutions are

6��

�� @�@A

not entirely country specific so that benchmarking and compari-son of practices and performances across agencies can contri-bute to a more effective management of the regulatoryfunction. There is evidence of increasing convergence in theregulatory institutions overseeing the energy sector. A similartrend has been identified in the telecommunications industry(OECD 2000). Furthermore, differences in the design of regu-latory agencies across countries reflect, at least in part, theregulatory frameworks and are not purely the result of legaland administrative traditions in each country.

There are, however, country-specific factors that seem to havesignificant weight in the choice of regulatory organizations. Forinstance, several early reformers in Northern Europe – Norway,Sweden, and the Netherlands – have established ministerialagencies for regulation rather than fully autonomous bodies. Inpractice, these ministerial agencies operate with a large degree ofautonomy, similar to an independent agency, but are formallysubordinated to the line ministry. This choice does not seem torespond to differences in the regulatory frameworks adoptedthat are roughly similar to those in the UK, the US or Australia.

# ��

This section provides an overview of the main design options forindependent regulatory agencies and the trade-offs involved.Design issues are discussed in Smith (1997), Smith and Shin(1995), and with a focus on process, in Brown (1996) and Berg(2000), among others. Green (1999) considers the interactionamong the various institutions involved in the regulatory processwith a focus on the British experience.

Independent regulatory agencies are defined as autonomouspublic bodies empowered to regulate specific aspects of an in-dustry. Regulatory agencies may also have judicial or quasi-judi-cial powers such as setting fines and penalties fornon-compliance or acting as an arbitrator in disputes among in-dustry participants. Independence, in this context, specificallymeans that the regulatory agency is protected from short-termpolitical interference. Political independence is primarily meantas a commitment to provide for a stable regulatory frameworkover time. This commitment protects investors against oppor-tunistic government intervention. The value of commitment isdiscussed in Spiller (1996). This is valuable even in a fully regu-lated market, which suggests an explanation for the emphasis on

�� 2��

�� @�@A

political independence in traditional regulatory agencies over-seeing investor-owned industries such as electricity in the US.In a deregulated environment, political independence may alsocontribute to other goals such as enhancing the transparencyand competitive neutrality of regulation, particularly when someof the utilities are publicly owned.

Regulators are also meant to be independent fromstakeholders in the sense that the regulated parties should not beable to influence regulatory decisions. This is necessary toensure that regulation is fair and does not favour one group ofstakeholders over the others. Almost all approaches to regula-tion are based on the principle that regulators should not be‘captured’ by the interests of industry players. See Noll (1989)and Viscusi, Vernon, and Harrington (1998) for a discussion ofregulatory capture and its economic implications. Unlikepolitical independence, which is an attribute of independentregulatory agencies, independence from stakeholders is soughtfor all public bodies involved in regulation.

A few countries have independent advisory agencies. Theseagencies are similar to independent regulatory agencies but haveno decision-making powers on regulatory matters. They advisethe ministry on a wide variety of regulatory issues, have monitor-ing responsibilities and authority in the resolution of disputes onissues such as network access.

In most regulatory systems, several organizations either dealwith regulatory issues or influence regulatory outcomes. Typi-cally, regulatory agencies split regulatory activities with a ‘lineministry’. In theory, the ‘division of labour’ between the minis-try and regulatory agency allocates policy-making, setting thegeneral framework and rules, to the line ministry. Implementa-tion of these rules is the responsibility of the regulatory agency.However, there is some overlap between policy and regulation.This overlap of regulatory activities may indeed be one of themain advantages of setting-up regulatory agencies. Laffontand Martimort (1999) argue that the co-existence of two ormore regulatory institutions may result in additional effectivemonitoring. It may also reduce the regulators’ scope for cap-ture by industry interests. On the other hand, the existence ofmany regulatory institutions increases the complexity of regula-tory processes, creates a need to protect the independence of theregulatory agency from other parts of government and requiresthe development of appropriate coordination mechanisms.

6��

�� @�@A

Autonomous ministerial agencies subordinated to the ‘lineministry’ have been set-up in some countries. These institutionsoperate on a separate budget, under an autonomous managementand may be subject to a differentiated legal framework but areultimately subordinate to the ministry. In practice, however,autonomous ministerial agencies in some countries operate witha substantial degree of independence.

Competitive authorities can influence regulatory outcomes,particularly when acting ex ante to prevent mergers and acquisi-tions that are deemed detrimental to competition or apply struc-tural solutions to remedy an anti-competitive industry structure.There is significant overlap between regulatory and competitionagencies in many areas including network access and pricing,and structural policies such as the unbundling of networks,mergers and divestitures.

Regulatory agencies can be designed in many different ways.Components include the role (or ‘mission’) they are assigned,their governance, the specific regulatory functions and processes,the resources and internal management of the agency, thestart-up strategy and other factors. The main components aresummarized in Table 1.

"��

��

The study covers 23 countries in Western and Central Europe,North America, and the Asia-Pacific. Institutional approachesto utility regulation can approximately be grouped into fourcategories according to whether regulation is managed exclu-sively by the ministry, the ministry and an independent advi-sory body, the ministry and a ministerial agency, or theministry and an independent regulatory agency. The basicapproach to regulation adopted by each country is summarizedin Table 2.

Independent regulatory agencies have been established in 10countries. As discussed in the previous section, these entities areautonomous bodies with specific powers, and are governed byone or several commissioners appointed for a definite and non-revocable period. The goals, powers, and activities of these regu-latory agencies vary to some extent. The (federal and state)agencies in the US, the UK, Canada, and Australia have a broadmandate to regulate industry and may act on almost all regula-tory and competition policy issues. The agencies in Ireland and

�� 2��

�� @�@A

��

��

��

� ��

��

��

��

��

��

��

��

!��

�� "��#��

��

�� $

��

!��

��$��

%��

&� ��

��

' �� (

�(�� ( ��

%��

��

� ��

) ' ��

) !��

) *� � �� $$��

) ' ��

� ��

) ��

) ��(��

) & ��

��+��

� �� "��

� ,��

� �� $��

� ,��

� ��

� ,��#�( ��

� -�� $��

��

� !��

� &� (�� $�� $�� $� ��

��$ ��

� ,�� $��

� ��

) �� $��

) *��"��$$��.��

) � ��

) ��

) ��

� &� ��

) ��"��#��

) /�� #�� ( ��

) � ��

� �� $

�� "��#��(��(��

�� $� ��

� �� $��

��

� �� $ ��

� ��

� % �� $ ��(�� $ �

� �� $��

� *��#��

� %� �� $� ��

� ,�+�

� ,�� $� ��( ��+ �

��

6��

�� @�@A

�� (��!*��

!�� (

!��

��

��

��

��

��

'��

'+��(� ��

��#

%��

!��

!��

& ��

01

0,

%��

2��

/��(��

/ ��

,��

-��

��

3��

,��

��

��

/�� 4��

,��+��

� ,�� #��

��

� ' ��+�� $� ��$$

� 0�� $��

� ��

�� (��

� *��

� 5��6��$ �� $ ��$ ��

� !��$$� �� $� �

��

5��

2��

��

��

,��"��

��

�� $ ��

�� 2��

�� @�@A

the Czech Republic are responsible for network and end-usertariff setting and network access issues, licensing and authori-zations. The remaining agencies have more limited and special-ized powers, mainly related to network and end-user tariffsetting and network access issues. This is the case in France,Italy, Portugal, and Denmark. These agencies have been estab-lished over the past 12 years, except in the US and Canadawhere they have been in operation for several decades. Many ofthese agencies have been created in step with market reforms.Table 3 summarizes the main characteristics of national andfederal independent regulatory agencies.

At the other end of the spectrum, the ‘line ministry’ is theonly organization directly involved in the management ofregulation in Germany, Japan, New Zealand, and Switzerland.In three of these countries – Germany, Japan, and New Zealand –there is no ongoing regulation of networks (that is tariffs and ac-cess conditions are negotiated by industry players), which, ar-guably, diminishes the potential role of an independentregulatory agency.

Independent advisory agencies have been established in Bel-gium, Greece, Luxembourg, and Spain. These agencies provideadvice to the ministry and are responsibile for monitoring andarbitration, but have no definite regulatory powers. In accord-ance with their advisory role, the areas of activity of theseorganizations are broadly defined to include most regulatoryissues. Governance and decision-making structures and inde-pendence safeguards are similar to those adopted by indepen-dent regulatory agencies. Table 4 summarizes the maincharacteristics of these organizations.

In the remaining five countries – Finland, Hungary, theNetherlands,2 Norway, and Sweden – management of day-to-day regulatory affairs is delegated to a ministerial agency for-mally subordinated to the line ministry and managed by apresident or director, appointed for an indefinite but revocableperiod (Table 5). Ministerial agencies specialize in regulatingmonopolies. Their main role is to manage network regulationincluding tariffs and access conditions.

Competition authorities have jurisdiction over electricity andgas markets in the majority of the countries examined. Formally,competition law applies to energy in all countries. However, there

2 There are plans to establish an independent energy regulator in the Netherlands.

6��

��

��

��

��

�

��

� ��

��

��

��

��

��

��

��

��

��

��

��

� ��

� �

��

��

��

��

�@

�@

A

�� !��

,� ��

- ��

3��(� $

��

�7��

& �� $

��

�� $

$��

�� $��

��

*��

�� "

��

�� 8

� ��

��

9

0�� :��

7��

5��;�

-��

/�� #

�� 8

�( ��

��#�� 8

��

��

*��

��

<

9

7��

�� $��

�� (��

� ��

��

� �� $

��

��

�� $

��

��

��

*��

��

=

:

7��

,��

3��

�� #

��

��"�� $$�

��

*��

��

>

>

/

,��

��

/�� #

��

��

*��

��

9

?

7��

'(��

��

� ��

/�� #

��$$��

��

$� ��"��

��$$��

��

��

*��

=� ��

��

@

0�� 9

7��

��

&��

��

�� #��

/�� #

�� 8

��

� ��

*��

��

@

9

/

5��

��

��=��

�( �� $

��

��

��

*��

��$ $�8

�� #

��

��

*��

��ABBA�

��

@

:

7��

,��(��

��

��$$�

*�� $$�

��

!"��

#��$

*��

��

:��C�>�� "

��

��

:

7��

'(�� (�

�� $� �(�

��

*�� $$�8

��

��

% � ��

*��

:

:

7��

%�� $ �� $��

$��

�(��

�� $ ��

��

�� 8

��

�( �� $$�8

��

��%�� ( ��

�� 2��

�� @�@A

�� !��

,� ��

- ��

3��(� $

��

�7��

& �� $

��

�� $

$��

�� $��

&��

*��

��

>

>

7��

,��(��

��

��$$�

��

��

��

� ��

��

'��D

*��

� ��

��

:

:

7��

,��(��

��

�� "

��

��

��

��

� �� 8

�� "

�� "

�� $$�

(��

*��

�� "

��

:

@

7��

,��(��

��

� ��

��

��

��

��

%��

*��

��

<

>

7��

,��(��

� ��

� ��BE:

��=BBB� $

��

��

��

��

��

��.��

� ��

��

are exemptions in some countries that may block enforcementin practice. The most common approach is that the competitionauthority enforces competition law, including cases of abuse ofdominant position, anti-competitive behaviour and mergers,and the ministry and regulatory offices manage regulation. This isoften complemented with some formal or informal cooperationarrangements to facilitate the exchange of information.

There are, however, significant departures from this ap-proach. In Australia, administration of competition law and mostregulatory issues are the responsibility of a single independent

��5(��#��(�� (��(��

��$�� (�� E

6��

�� @�@A

agency. This applies both at federal and state levels. In theNetherlands, competition law and regulation are also control-led by a single institution, namely the ministry. In the US andUK, merger policy is concurrently enforced by the energyregulatory agency and the antitrust enforcement office.

!��"

The design of the ten independent regulatory agencies identifiedin Table 2 is analysed below and, more specifically, identifies thecommon characteristics. This suggests a common blueprintfrom which regulatory agencies are designed and adapted tonational circumstances.

ObjectivesThe role of independent energy regulators is largely concen-trated in two inter-related areas. One is monopoly control. Mostregulatory agencies are responsible for price control and accessin the monopolistic segments – transmission and distributionnetworks – while avoiding any anti-competitive impact of thesesegments on competitive segments such as production or gen-eration and supply. The other is consumer protection. Regula-tory agencies often have responsibility for end-user tariffs andother conditions such as quality of service. However, some agen-cies have a more general mission to promote efficiency, includ-ing oversight of the competitive activities.

��

,� ��

��

$��

��

*��

3�� $

�� #

��8

�� #� ��

��

��

)��

*��

3��

��( ��+��

�� $$

��

$��

��

��

*� ��

*��

/�� #

��

*��+�

*��

��

��

/�� #

��

�� $

�� #��

��

$��

%+��

*��

/�� #

��

�� 2��

�� @�@A

JurisdictionRegulatory agencies usually deal only with economic regula-tion. Social regulation is the responsibility of other authorities.As an exception to this rule, the UK launched in 1998 a reviewof the role of regulators that aimed to shift the emphasis of theiractivity from efficiency towards distributional issues. The gov-ernment proposed changing the primary statutory duty ofregulators to one of consumer protection, including dealingwith fuel poverty and other social objectives. In Denmark, theregulator is charged with overseeing some environmentallymotivated objectives.

Competition law is not the primary responsibility of mostregulators, except in Australia where the competition authorityis also the ESI independent regulator. However, ESI regulatorsoften perform some supporting functions for the application ofcompetition policy, such as monitoring, providing informationand advice or bringing cases before the competition authority. Inthe US, the regulator has concurrent jurisdiction with theantitrust authority in merger cases.

Industry coverageElectricity and gas are regulated by the same institution in all ofthe cases examined except Ireland. This is consistent with minis-tries that also cover all forms of energy. This pattern of associa-tion of electricity and gas reflects significant and growinginterdependencies between the two industries, such as theincreasing use of gas for power generation and the integrationof gas and electricity firms. It also reflects the fact that both arenetwork industries facing similar regulatory issues, such asaccess to the network, and are subject to a similar regulatoryapproach (both industries are being deregulated worldwide).

Decision-making, appointment, andindependence of regulators

A commission governs a majority of independent regulatoryagencies, the exceptions being the Czech Republic and, untilrecently, the UK where there is a one-person regulatory board.However, the Utilities Bill in the UK established a collegialboard for the regulatory agency. Appointments are for a fixedterm of between three and seven years. Casual observation indi-cates that, with few exceptions, stakeholders are not appointedas regulators but this does not seem to obey formal rules. Formal

6��

�� @�@A

independence safeguards generally include the non-revocabil-ity of appointments except in extreme circumstances, such asserious misconduct or insanity, a separate budget, managerialautonomy, or a stable source of financing.

FunctionsThe functions of independent agencies vary. However, theregulation of the monopoly elements (networks) is typicallyallocated to the independent agency when it exists. The regula-tion of energy transmission tariffs is a core activity of all inde-pendent regulatory agencies, with the exception of Portugal.Another common activity of independent regulators is theregulation of end-user tariffs, with the exception of France andIreland. Regulating entry and exit into the industry throughthe issue of licenses and authorizations is the responsibility ofthe independent regulators in Australia, Canada, the CzechRepublic, Ireland, the UK, and the US.

To carry out these tasks, regulatory agencies are typicallyresponsible for monitoring market conditions, compiling andauditing company information, and pursuing and penalizingmisconduct. The more technical aspects of regulation, such assystem operation rules, are frequently left to industry bodies,while the more strategic aspects of regulation and legislation areconducted by legislative bodies and government. For instance,in the US there has been an increase in legislative activity byindividual states in areas that were traditionally handled by thePublic Utilities Commissions and stakeholders have taken aleading role in developing operational rules.

In federal countries, regulatory functions are split betweenthe federal and the state or provincial regulators, with the lattertypically being responsible for the regulation of retail markets,including distribution and retail supply activities. This is thecase in Australia, Canada, and the US. The split of regulatorypowers reflects the local nature of distribution and retail supplyservices versus the systemwide dimension of production,generation, and transmission.

Process and appealsThe procedures of regulatory agencies show significant similari-ties, which include� a decision-making process that includes an obligation to

conduct hearings and consultations with affected parties

�� 2��

�� @�@A

and to make reasoned decisions, and to make thesedecisions public;

� an appeals mechanism, which establishes that either anadministrative court or an ordinary court of justice is theappeals body;

� mechanisms to make these institutions accountable, whichinclude an obligation to submit a report of activities to theparliament or other political body, and some form of auditingand control of performance by the relevant administrativebody.

Most regulatory agencies are governed by a collegial bodywith an odd number of members appointed for a non-revocableterm and subject to strict conduct requirements in their rela-tionship with the industry during and after their term in office toprevent conflict of interests. The agency has jurisdiction only oneconomic regulation (it does not deal with health, safety or envi-ronmental regulation), covers electricity and gas issues and, inparticular, deals with the regulation of monopolistic segments ofthe industry and setting end-user tariffs. The agency also moni-tors the industry and provides advice to the ministry. Process isregulated to ensure transparency and neutrality of decisions.Finally, in federal countries, the state regulator specializes inretailing while the federal regulator primarily deals with whole-sale activities.

%��!!�� &

Independent regulatory agencies’ powers vary significantly fromthose with a general mandate to oversee and regulate the energysector to those with a specialized function such as regulatingnetwork access or end-user tariffs. Diversity is even larger whenministerial and advisory agencies are considered intermediateoptions between a fully independent regulator and a ministry-only approach to regulation. While the legal and administrativetraditions of each country are important determinants of thechoice of regulatory approach, some of this variation can beexplained by the regulatory framework in which the agency isset-up to operate.

The evidence examined in this section suggests the follow-ing three empirical regularities. First, there is no regulatoryagency in countries that have adopted a ‘light-handed’approach to regulation, such as New Zealand and Germany, in

6��

�� @�@A

which there is no ongoing price regulation and prices andaccess conditions are negotiated or proposed by the utilitiesregulated. In the other countries where no regulator exists,market opening has been minimal to date. The lack of a special-ized regulator may be explained by the few regulatory tasks tobe performed in countries where regulation is minimal ormarkets have not been established. In Finland, where access tonetworks is regulated in principle but there is no ongoing priceregulation, the powers of the regulatory agency are also veryrestricted.

Second, in countries that have vertically restructured theirindustries at the time of reform, regulators have relatively largepowers and independence. Some countries restructured theindustry simultaneously with the introduction of competition.Restructuring measures include the unbundling of competi-tive activities, such as generation and end-user supply, fromnetwork activities and divestitures to reduce market power.Australia, Italy, New Zealand, the UK, and some US states pro-vide examples of this approach, which requires that many regu-latory tasks be recurrently performed, including the auditingof regulated activities, monitoring access conditions, and set-ting prices for regulated activities. Strong regulatory agenciesseparated from the ministry and endowed with relatively im-portant regulatory powers have been established in most of thecountries following this strategy, such as Australia, Italy, the UK,and the US (the exception being New Zealand as discussedabove). The broad scope and powers of these organizations areconsistent with the large range of regulatory functions to beperformed.

Third, in the remaining countries regulatory organizationsshow variation that cannot be readily related to the regulatoryframework. For instance, as noted in the introduction, someministerial agencies may act rather independently in practice. Inexplaining institutional design among these countries, legal andadministrative traditions seem to play an important role. How-ever, the experience in the telecommunications industry suggeststhat perhaps there will be increasing convergence in the future, asreforms mature.

�� !��

Regulatory institutions are changing in step with the develop-ment of new regulatory frameworks. Institutional change

�� 2��

�� @�@A

reflects adaptation to a new regulatory environment character-ized by open markets, new regulatory needs such as transmis-sion pricing and increasing both regionalization and the linksbetween industries.

The allocation of power and responsibilities to differentorganizations makes objectives more explicit and decisionsmore transparent in each area of public intervention. It alsoprovides for a framework that supports neutrality in regulatorydecisions. There is, however, no such thing as an institutional‘free-lunch’. The co-existence of several institutions that havejurisdiction over the same industry creates complexity thatspawns the need for increased coordination among the variousauthorities involved and, possibly, greater compliance costs tothe regulated parties. Developing coordination mechanisms inan increasingly complex institutional setting is the key condi-tion to an effective reform.

The multiplication of organizations also raises concernsabout the efficiency of the public sector. Bureaucracies areexpensive, and tend to grow and self-perpetuate. Thus, it is es-sential that regulatory institutions be examined over time andthat their role and resources be continuously adapted. A keychallenge in this regard is to find the appropriate balancebetween general energy policy, industry-specific regulation,and competition policy. The relative weight of each of thesepolicies is gradually changing as competition in energy marketsprogresses and the allocation of resources to different policyareas may need to be adjusted in response.

Periodical reviews of the institutional setting must take intoaccount the changing boundaries of the energy industries. Theneed for harmonization across the gas and electricity industriesas well as the trading areas will continue to put pressure on regu-latory institutions. The scope of sector regulators may need toadapt in order to cope with these structural changes as has beenthe case in some countries with the merger of electricity and gasregulators. Further change can be expected in this very dynamicsetting.

Most independent regulatory agencies share a number ofcharacteristics. These common elements provide a blueprint forthe design of new agencies. However, this blueprint is incom-plete given that regulatory agencies differ widely on a number ofissues. Designing regulatory mechanisms will benefit greatlyfrom international experience.

6��

�� @�@A

� � � �� Berg S. 2000Developments in best-practice regulation: principles, processes, andperformanceElectricity Journal July: 11–18

Brown A. 1996Transparency in regulated industries: elements and importance[A Harvard Electricity Policy Group discussion paper.Draft. 20 May 1996]

Green R. 1999Checks and balances in utility regulation: the UK experienceIn Public Policy for the Private Sector: viewpoints on competition and regulation,no. 185Washington, DC: The World Bank[Available online at http://www.worldbank.org/viewpoint/HTMLNotes/185/185summary.html (accessed on 05 June 2003)]

IEA (International Energy Agency). 2001Regulatory Institutions in Liberalised Electricity MarketsParis: IEA/OECD

Kerf M, Schiffler M, and Torres C. 2001Telecom regulators: converging trends?In Public Policy for the Private Sector: viewpoints on competition and regulation,no. 230Washington, DC: World Bank[Available online at http://www.worldbank.org/viewpoint/HTMLNotes/230/230Kerf-0516.pdf (accessed on 05 June 2003)]

Laffont J J and Martimort D. 1999Separation of regulators against collusive behaviorRand Journal of Economics 30(2): 232–262

Noll R. 1989Economic perspectives on the politics of regulation, pp. 1253–1287In Handbook of Industrial Organization, edited by Willig R D and Schmalensee RNorth-Holland: Amsterdam.

OECD (Organisation for Economic Co-operation and Development). 2000Telecommunications Regulations: Institutional structures andresponsibilities[Document DSTI/ICCP/TISP(99)15/FINAL, 24 May 2000]Paris: OECD

�� 2��

�� @�@A

Smith W. 1997Utility regulatorsIn Public Policy for the Private Sector: viewpoints on competition and regulation,nos 127, 128, and 129Washington, DC: World Bank[Available online at http://www.worldbank.org/viewpoint/(last accessed 05 June 2003)]

Smith W and Shin B. 1995Regulating infrastructureIn Economic Notes. Country Department IWashington, DC: World Bank

Spiller P. 1996Institutions and commitmentIndustrial and Corporate Change 5(2): 421–452

Viscusi W K, Vernon J M, and Harrington J E. 2000Economics of Regulation and Antitrust, 3rd ednCambridge: MIT Press

-�� !��!�� !��

��;�� !��

�� %��

/�� '��*��C+�� D�� 0��" 0/'

�� @@��E

��'� �� 4� �� #�� !�� *!�� %�$� ��%�� 9 6�� 4��% �� !�� 4�� %�� 4��" �� !�� %�� 4�� %#�� *�4��" �� 9

' �� 8� �� !�#!�� !�� " �� !��#��" �� 4�� 9 >�� !��#�� 4� ��;�� !�� *�� %�� !�� %�%�� !�� !�� !�� !�#!�� !�� *!�� %� ��!�� !�� 9 )*�!�� 4�� 4� �� 9

C/�� )�� /�� '�� )��" �� 0�� 6�� >��

(��$ >�� +� 0/'

/�� '��*��

�� @@��E

��

Worldwide, an increasing number of national and local govern-ments are turning to the private sector to manage and expandtheir networks for water and sanitation services. More oftenthan not, the public sector has failed to operate services effi-ciently to keep up with the ever-growing demands. This isparticularly so in developing countries, where the urban popu-lation growth is rapid and government resources are severelylimited.

The trend of PSP (private sector participation) has raisedconcerns about how the poor fare under such arrangements.Some fear it brings tariff hikes and makes service unaffordable,or that private operators prioritize high return, easy-to-reachcustomers (Hall 2003). Others argue that PSP is an importantpart of a wider set of reforms, one that has the potential tobenefit the poor (World Bank 2003; Estache, Gomez-Lobo, andLeipziger 2000; WSP and PPIAF 2002). A comprehensivereview of the impact of the experience of primarily LatinAmerican countries of utility privatization and the impact onconsumers, especially the poor, has recently been published(Ugaz and Price 2003).

A growing body of evidence suggests that private firms arewilling and able to address the needs of the poor if they are givenincentives and flexibility to do so (WSP 2001a; WSP 2001b;WSP and PPIAF 2002; Gray 2001). Governments embarking ona process of reform make decisions when engaging the privatesector that in one way or another affect the operator’s behav-iour. What should the private sector’s mandate be? How should theprivate sector be remunerated? How should the private sector beregulated? The way in which these questions are answereddetermines whether, and how much, the poor can benefit fromPSP.

The purpose of this paper is to examine how transactions canbe designed to benefit the poor.1 It is concerned mainly withtransactions for the provision of network services, but alsoaddresses the impact of these transactions on markets foralternative services.

1 Definitions and perceptions of poverty differ across countries and localities. Here werefer to those groups and individuals whose well-being is low relative to others.

-�� !��!�� !��

�� @@��E

� �� '��

Markets for piped water and sanitation services are differentfrom conventional markets, in a way that they cannot be servedefficiently by multiple firms competing with each other on anongoing basis. Due to the large investment required to set up adistribution network, and the declining cost of serving eachnew customer, these services are most efficiently provided by asingle firm.2 Markets with these characteristics are known asnatural monopolies.3

The economics of scale associated with a natural monopolyserve as a barrier to competition, giving the incumbent firm‘market power’ to charge prices higher than it could commandunder competition. This is perceived to be at odds with the inter-ests of consumers and is the basis for regulating the industry. Itis possible to make a case of no regulation (or very limitedregulation), as there is always an alternative form of supply—even if it is the one currently being utilized. This gives rise to aform of contestability, which should help ensure consumersface prices that are lower than those faced prior to the entry ofthe new operator while also providing the new operator with anopportunity to earn profits, thus giving an incentive to enterthe market. It could be argued that this creates the greatestincentive for system expansion (see discussion later). Whensome element of subsidy is involved the case for greater regula-tion becomes stronger. This discussion, however, goes beyondthe scope of this paper. Consequently, governments often limitthe provider’s prices through controls, sometimes referred to asconduct regulation, which usually establish the prices that canbe charged. Various approaches can be followed, the two stand-ard approaches are price caps (multi-year price paths, oftenreferred to as RPI–X or incentive-based regulation) and rateof return approaches.4

2 Declining average cost is not a sufficient condition for natural monopoly to exist.Rather, natural monopoly occurs where the cost function is sub-additive with respectto the cost functions of other firms.3 In considering the monopolistic nature of such markets, it is important to distinguishservices for distribution of water from the supply and retail of raw water; markets forsupply and retail may not have natural monopoly characteristics. This paper is con-cerned mainly with distribution. For further discussion, see Klein and Roger (1997).4 It is not the purpose of this paper to discuss the differences between the various formsof price control. An explanation of the differences can be found in Viscusi, Vernon, andHarrington (1997).

/�� '��*��

�� @@��E

Where price caps are used, the monopolist will be tempted tolower the quality in order to minimize costs, and governmentsmust therefore regulate this. Governments can require opera-tors to use certain materials or methods for constructing anetwork (input-based regulation), or to adhere to particularstandards for customer service (output-based regulation).Minimum standards for water pressure, water quality andcontinuity of service are common.

Not all markets for water and sanitation services are naturalmonopolies, especially in developing countries. Markets forhandcart water delivery and emptying of septic tank, for exam-ple, lack the economics of scale associated with a piped networkand are contestable (even if no actual competition takes place).Competition helps to select the most productive firms andprovides ongoing incentives to respond to consumers’ needs.This has advantages over a regulated monopoly. Competitivemarkets need little or no conduct regulation, thus placing asmaller burden on government resources.

Competition cannot be taken for granted. For example, watervendors in Jakarta are thought to collude with one another, whilein Manila a local on-seller of piped water services operates withlittle competition (Crane 1994). The role of the government insuch cases is to reduce barriers to entry. These barriers takenumerous forms, including non-competitive behaviour, restric-tive legal requirements (including licensing), under-developedcredit markets, and limited access to technology.

(�� ) �� *�� !��

What is it that governments wish to achieve for the poor? Theanswer to this is not as straightforward as it may seem. If the goalis declared as the availability of affordable services that meet thedemand of residents, then how ‘affordability’, ‘services’, and even‘residents’ are defined needs to be addressed. Does ‘afforda-bility’ refer to a proportion of household income spent onwater, a government-mandated level, or something elseentirely? Are only network services to be considered, or arealternative services also acceptable? Do squatters and seasonalmigrants count among the ranks of residents? While answers tosome of these questions depend on local politics and priorities,it is possible to make several generalizations about the poor andtheir predicament.

-�� !��!�� !��

�� @@��E

One of these is that in many cities – for example, Africa(Collignon and Vezina 2000); Jakarta (Crane 1994); Manila(PCWS 2000); Buenos Aires (Hardoy and Schusterman 2000) –the poor tend to live in areas that are not served by a citywidenetwork. Instead, they rely on small-scale providers such ashandcart vendors, water kiosk operators, and tanker trucks.Efforts to improve service by extending the network are typi-cally hindered by the physical layout of poor neighbourhoods,and by socio-economic constraints including a weak landtenure (ADB 2002).

Another generalization is that households that lack aconnection pay more per unit of water than their better-offneighbours for a network service. A growing body of empiricalevidence shows that the poor are willing to pay as long as itwould cover the ongoing cost of piped distribution (WSP1999; Foster, Gomez-Lobo, and Halpern 2000). In many cities,the poor lack access to formal credit markets and thereforehave difficulty affording upfront connection fees (World Bank2001a; Walker, Ordonez, Serrano, and Halpern 2000).

Given this, the poor might benefit from PSP in three mainways.� Downward pressure on prices Efficiency reduces costs; this

frees resources so that downward pressure can be placed onprices, making the service more affordable.

� Expansion of network coverage The quality of piped servicetends to be superior to alternatives used by unconnectedusers, and its price per unit is invariably lower due to theeconomics of scale (Kerf 1996). Qualities such as reliabilityand reduced levels of contaminants are particularly impor-tant for low-income households, who are less able to affordstorage and have inferior access to health care.

� Service levels that meet the needs of customers The introduc-tion of PSP is an opportunity to enhance the operator’sincentives to respond to consumers’ needs, especially thoseof the poor. Circumstances in many low-income neighbour-hoods demand that service providers be innovative in themethods and materials they use.

These outcomes would benefit households to varying degrees,depending on whether or not they have access to a connection.Those with connections clearly have more to gain from lower

/�� '��*��

�� @@��E

prices than from expansion. Likewise, cities with low coveragemay have more to gain from expansion than from lower prices.

Indeed, policy-makers face trade-offs between prices andcoverage. The cost of network expansion must somehow bepaid for—passed on to consumers, either directly throughprices or indirectly through the government in the form oftaxes.5 Depending on the extent of cost savings from efficiencyimprovements, a tariff hike might be necessary. Turning thisaround, prices can affect investment decisions by acting as anincentive, or disincentive, to expand.

+�,��-��.�!�� !��!��) ��

A contract’s potential for achieving pro-poor objectives islargely a function of the efficiency gains it can bring about,because these are the basis for expanded coverage and downwardpressure on prices. This section discusses efficiency and its rela-tionship with contract design.

��

Before considering the impact that different contractual formscan have on meeting pro-poor objectives it is helpful to distin-guish among the three types of efficiency.

Allocative efficiency is achieved when the most productive firmsare engaged in service delivery—efficiency in terms of the serviceprovider.6 In a competitive market, these are the firms thatoutcompete their less productive rivals. Where the market is anatural monopoly, a competitive bidding process can be used toidentify and engage the most productive firm; competition on theplaying field is replaced by a contest for the playing field.7

Installing the right firm (or firms) is not enough, however.They must also face ongoing incentives to maximize productiveefficiency. Competition provides the basis for this in a multi-firm marketplace, though interventions may be required to en-sure incumbents do not engage in anti-competitive behaviour.A natural monopoly lacks the competitive pressure of the

5 While the tax burden could theoretically fall on the same group of consumers, it ismore likely to affect a broader group. This will depend on the tax base and fiscal policy.6 Allocative efficiency is also relevant when considering consumption, which is dis-cussed in this context in the section on using efficiency gains to place downwardpressure on prices.7 For a discussion on franchising and privatization, see Dnes (1995).

-�� !��!�� !��

�� @@��E

marketplace, so incentives are instead built into a contract withthe monopolist.

Dynamic efficiency refers to the inter-temporal benefits ofinvestment decisions. Contractual incentives to invest soonerrather than later clearly affect the poor, who tend to depend onthe larger outlays needed to expand primary and secondary net-works. Similarly, contractual incentives to consume a greater orlesser volume of water inevitably affect the availability of rawwater.

��

Contracts for public service provision are essentially an enforce-able set of rights and obligations that limit profits or priceswhile specifying required outputs, and as such their design helpsdetermine the monopolist operator’s productive efficiency. Tosee why, consider four stylized contractual forms that are preva-lent in the water sector: management, affermage, lease, andconcession (Table 1).8

� Management contract The private operator receives a simplefixed-fee and has limited operational responsibility. It is notrequired to make investments. The duration of managementand service contracts is usually 3–5 years. In the case of anenhanced management contract, the operator receives a feethat is adjusted by a set of performance benchmarks so as togive the firm an added incentive to achieve specific goals.

� Affermage Under an affermage the operator has the right toa fixed proportion of each unit of water sold. The operatorhas broad operational responsibilities but is not usuallyrequired to invest in expansion. Affermage contracts com-monly last 8–15 years.

� Lease contract The operator has the right to all revenue inexcess of a fixed fee it pays the government, but does not usu-ally have obligations to invest. Leases are typically written fora period of 8–15 years.

� Concession The operator has the right to all revenues andcan make investments. Concessions have a relatively long du-ration, usually 25–30 years. In some cases, the winning opera-tor makes either an upfront payment or an annual paymentmay be required.

8 Several documents provide a description of the major contract types and theirvariants. See World Bank (1997); Kerf, Gray, Irwin, et al. (1998); and Delmon (2001).

/�� '��*��

�� @@��E

' ��

��F

,��

�$$��

3��

' ��

%�� ,� ��

3�� "

��E�,�� $ �

��$��#��(��

� �� #�E

��

�� $� ��

�� E

%�� $ ��

�� (�� $

�(�� E

��

�(��

�� (��

� �� $� ��

�� E

%�� $ ��

�� (�� $

�(�� E

3��#� ��(�

�� ( ��

�� (��

�� (��

�(��( ��$��

�� (�

�$$��$��

�� (�� E

��

�(��

�� (��

� �� $� ��

�� E

%�� $��

�� (�� E

��#�� (��

��(��

�$$��8� ��

�� (��

�� (��

�(�� $� �(�

�� (��

�� $$��

��E

' �� $ �

��

��$�� (��

�� E

�� $ �� $�� $

�� (� �(�

�� E

&�� ,��

�� ,��-� %�� $�� $��

�� $ ��"��

��

�� $��"��("

��#�

��.�� E� 5(��

� �� (��

�� (�� $$��

�� (�

�� $$�� $��

��(��(� ��

$��

� � �(��

��( �� E

5(��$$��$��

� ��$��

��$ ��

�� $$��E

�� E� 5(�

��

��

�� $��$��

� �� (�

$�� $��(�

��$�� (�

�� (��

��( ��E

&��$ ��

�� $$��E

�� E� 5(�

�� (��

��

� �� $��$��

��

�� $� �� "

�� $��

��(�� (�

� �� ( ��E

"��

� (��

��

��+�

� ��

�G��+��

��

��

/��

,� �5 ��

&��

' �� ;!� ��

,��

��#

/��

��

� +��.�

��

-��

��

'��

��

'��

'�� G��

-��

'(��

�� 5�� $��

�� H,&��&&!�%� �ABBA

-�� !��!�� !��

�� @@��E

While there are innumerable variations in each of these con-tractual forms, the three defining characteristics are the basisfor remuneration, the scope of responsibilities, and the length of thecontract. Together, these determine the potential of a contract toyield gains in productive efficiency.

The basis for remuneration can be structured in three basicways.1 Fixed fee, where the operator is paid a set amount for provid-

ing particular services.2 Per unit share, where the operator is paid a fixed amount for

each unit of water it sells.3 Residual claimant, where the operator is allowed to retain all

revenues above a fixed fee it pays to the government.

The scope of the operator’s responsibilities can range fromlimited responsibility for a part of the operation in one section ofa city, to full control over all assets and all areas. While, oftenthere is a correlation between this mandate and the basis forremuneration, this does not have to be the case. For example, alease can allocate complete responsibility to the operator tomanage the system, including the freedom to make investmentsthat expand the customer base. Similarly, a concession can givethe operator limited responsibility over a particular set of opera-tions, or a certain section of the city. In practice, it is expectedthat the basis for remuneration will drive the scope of the man-date. For example, the more the operator expects to benefit fromefforts to minimize costs, the greater the control it may seek overassets.

The third defining characteristic is the duration of thecontract, though in fact this can be seen as an extension of thescope of the operator’s responsibilities. The longer the operatorhas rights over revenue, the greater the potential benefits itperceives.

Productive efficiency will be greatest when the operator isgranted rights that mimic ownership. The higher the proportionof revenues retained by the operator, the more effort it will maketo lower costs and raise output (Grossman and Hart 1983).

Why, then are all contracts not ‘high-powered’ arrangementsthat maximize productive efficiency? One explanation for thepopularity of less powerful contracts is that they allocate morerisk to the government and thereby lower the ostensible cost ofcontracting with the private sector. In the case of a management

/�� '��*��

�� @@��E

contract, the government takes on all the commercial risk.Since the private sector need not hedge against such contin-gencies or otherwise incorporate them in its costs, it can chargea lower price. One downside to this arrangement, which affectsall existing and potential consumers, is the lower potential forefficiency gains.

Another explanation for the range of contractual forms is thatit is not always possible to choose the most efficient form of con-tract. Many governments are not in a position to award leasesand concessions, or to implement the reforms that must precedethem. Decision-makers and consumers are uncomfortableceding responsibility of ‘essential services’ to a private firm.They perceive an unnecessary loss of control and an unwar-ranted transfer of resources—concerns that may be wellfounded if the transaction is not well designed.

#��

Efficiency is pointless unless achieved in the production of serv-ices demanded by customers. The freedom of service providersto respond to customers’ needs and to have control over thechoice of inputs and methods are addressed in the section‘Giving service providers the freedom to respond to demand’.Two fundamental issues that warrant consideration here.

Allowing for a range of service levels can be difficult in thecontext of monopoly regulation. Where price controls are used, acorresponding service level must be specified. This, however,restricts the ability of the operator to respond to consumers’needs. While many service levels could be specified, in practicethis would be cumbersome to administer. Limiting the opera-tor’s rate of return could help, but this has other drawbacks.Rate of return regulation can generate undesirable incentives. Inparticular, firms will use too much capital relative to otherinputs, and thereby compromise productive efficiency (Averchand Johnson 1962). Where coverage targets are used in concertwith price controls, these lead to the same dilemma. It is difficultto measure and verify expansion if the operator can satisfytargets with a range of service levels.

The operator’s freedom to innovate is also important. Aresome contractual forms better at this than others? In short, yes.Remuneration under low-powered contracts is structured overa short period of time, and does not provide incentives to useinputs designed to provide returns over a long term. A logical

-�� !��!�� !��

�� @@��E

response is for the government to demand higher qualityinputs by describing these in the contract. However, in doingso it limits the freedom of the operator to innovate.

In sum, high-powered contractual arrangements havegreater potential for delivering benefits to the poor since effi-ciency and net revenue growth provide the basis for lowerprices and increasing coverage. However, there is no guaranteethat the poor will benefit. Other stakeholders including richconsumers, the operator, and the government itself can beexpected to compete for these benefits and are often betterplaced to win them. Van den Berg (2000) examines the alloca-tion of benefits under several water concessions in Argentinaand finds that transaction design is an important determinantof how much consumers’ gain. Restrictions on the types ofservices an operator can offer, or the inputs it can use, alsodetermine the outcome for the poor. The next section exploressome of the policies that impact the interests of the poor.

� ��-�� !��

The previous section dealt with the determinants of a contract’spotential to deliver benefits to the poor. The following factorsare central to realizing this potential.1 Efficiency gains are used to place downward pressure on

prices The ‘efficiency dividend’ can be distributed in manyways.

2 There are incentives for expansion in poor neighbour-hoods Incentives for expansion are no guarantee the poorwill benefit; in fact, several factors may lead to an anti-poorbias.

3 Multiple service levels are encouraged The poor benefit frominnovation and multiple service levels.

4 All service providers are free to respond to demand Transactionsthat hinder service provision by alternative enterprises can dothe poor a disservice.

$��

Most public utilities survive on the basis of an external subsidy,and many draw down on physical capital as maintenance isneglected. When these practices come to an end, typically dur-ing a process of reform, higher prices are inevitable. Efficiencygains, however, work in the opposite direction. As costs fall and

/�� '��*��

�� @@��E

collection rates improve, the need for revenue is reduced.Depending on the relative strength of these two forces – costsavings and cost recovery – prices under PSP can go in eitherdirection (Gray 2001). Box 1 illustrates some examples of effi-ciency gains under PSP.

Regardless of which direction prices move, efficiency gainscan always be used to apply downward pressure. An importantstep towards this goal is to award the contract on the basis ofcompetitive bidding. Where a contract is allocated using a lesscompetitive method, firms are not likely to offer their best price.Instead, each bidder will attempt to capture some share of theexpected efficiency gains for itself. This can be distinguishedfrom the other benefit of competitive bidding, that of ensuringthe most productive operator is awarded the contract.

Where the private sector is made to surrender these gains, thegovernment will find itself with discretion over their allocation.It can pass them on to the existing network customers by plac-ing downward pressure on prices, use them to expand coverageto previously unconnected areas, or allocate them to any othergroup by spending revenue in a particular way. Contractsawarded on the basis of lowest tariff channel benefit the pricelevel.

�,$ )�� " �� !�� !��!��

� �� (�� '�� 4�� !��" �!�� " ��

%�� %�� FF@9 6�� 4�� !� ��

�� @9@ �� FFA �� 9� �� FFE �� #

�� %�� !�!�� 4�� %�� G9H� �� FFA ��

G9@H �� FFE9 '� � �� !��#

��" �� 4��" �� G9� �� FFA ��

G9F �� FF�9

� �� 4�� !��"�� !� �� %��

�� H �� H9� �� * �� !�� !��#

!�� 9

� �� $ -�� !��"�� !� �� %�� #

�� G9@ �� E �� "�� 9

� �� !�� !�� " �� !�#

��" �� 9

� ' ��!�� !�� !�4�� !�� %�� !�#

�� '�� -�� !�� %�� !��

�� " �� !�4�� 9

/�� " �AGG��

-�� !��!�� !��

�� @@��E

The tariff structure can be designed to further allocate effi-ciency gains among different groups of network customers. Anapproach adopted by many governments, especially in develop-ing countries, to use an IBT (increasing block tariff). A perunit premium is charged at higher consumption levels whilediscounting prices at lower levels. Low-income households arethought to consume smaller amounts of water and thus benefitfrom this arrangement.

In practice, IBTs do not always achieve their goal (Whittingtonand Boland 2000). One reason is that the poor tend not to beconnected to the network and therefore lack access to the sub-sidy. Low-income households that are connected frequentlyshare their service with neighbours and as a result are chargedhigh volume rates intended for the rich. Even where a house-hold is the sole consumer, it is often subject to a fixed chargebased on a minimum volume of water far in excess of what ituses, which again leads to higher per unit rates. Another draw-back of IBTs is the disincentive they give operators to expandin low-income areas. How can affordability be achieved moreeffectively? While efficiency is the most important step, some-times the poor need extra help. The first task in designing asubsidy scheme is to identify the target group, understand theconstraints it faces, and articulate clear objectives based on thisunderstanding. Information about consumer demand is essen-tial, as this reveals gaps between willingness-to-pay for servicesand the cost of their delivery.

Subsidies can be designed to target assistance toward con-sumption or access. Where demand is sufficient, but consum-ers lack the resources to connect to the service, the subsidyshould be designed to facilitate access rather than consump-tion. Where willingness to pay is insufficient, a consumptionsubsidy can be justified—although consideration of the qualityof the service being offered should also be undertaken. In bothcases, efforts should be made to reach as many intended ben-eficiaries as possible, while minimizing assistance to those whoare not part of the target group. Subsidy schemes should alsobe designed to target beneficiaries with maximum efficiency,and be implemented with the least possible administrativecost.9

9 For a discussion on subsidy design, see Brook and Smith (2001); World Bank(2001b); and Coady, Grosh, and Hoddinott (2002)

/�� '��*��

�� @@��E

��%��

Low-powered contracts, such as management contracts, havelimited potential for attracting outside investment but cannonetheless be used to commit the private sector to lay pipesand provide connections. Where a government adopts thisapproach to expand service in low-income areas, it must iden-tify priority neighbourhoods and fund the requisite invest-ments. This requires not just financial resources, but alsoinformation about the location of the poor along with theknowledge of appropriate methods and technologies.

High-powered contracts transfer these responsibilities to theoperator, which bases its decisions about investments onexpected profits. This naturally leads it to expand in neigh-bourhoods where costs are low and revenues are high, startingwith the most profitable first. As the operator goes about assess-ing the return from expansion in each area, it takes intoaccount many factors. Not all of these are found in the con-tract. Consumer demand and population density, for example,or the placement of existing infrastructure are simply given—and can have a significant impact on decision-making.

Other influences on expansion are very much a part of thetransaction. Chief among these is the tariff structure. Where anIBT is in place, the return in high consumption areas will beboosted as higher rates are levied. As low-income neighbour-hoods have lower demand for water, expected profits will belower than in better-off neighbourhoods, all other things beingequal.

The exception is where remuneration is based on a per-unitshare basis, as in a typical affermage contract. Here the paymentto the operator is based on the volume sold regardless of thelevel of revenue. The operator is insulated from tariff levels andstructures, except where these affect demand (Box 2).

Even where an area is deemed profitable, legal restrictionsmay prevent an operator from serving it. Examples includeregulations or contractual clauses that ban service in illegalsettlements, or which make it difficult for residents lacking toconnect (Calaguas and Roaf 2001; WSP and PPIAF 2002).

In practice, high-powered contracts rarely rely only on incen-tives linked to profit from revenues to ensure expansion topoor areas. Some contracts require that a certain proportion of

-�� !��!�� !��

�� @@��E

consumption takes place in each ‘block’ of customers under anIBT; or coverage targets are established specifying service to aminimum proportion of the population. Targets are usually setout as the percentage of households or population to be servedby a network connection, and are backed up by penalties for non-compliance. In order to encourage the operator to focus on thepoor, targets are sometimes set for each sub-municipal area.

This approach was adopted in the lease in Dakar, and in theconcessions in Buenos Aires and Manila where there are largeperi-urban poor areas (Rosenthal 2002; WSP 2001a). It isineffective in cities where the poor are distributed in very smallpockets among the better off, or where targets are set lowenough for the operator to be able to avoid unprofitablecustomers and still achieve the prescribed coverage. Whetherprecise coverage targets are effective is a moot point. Operatorsmay try to move slowly in those areas that will be least remu-nerative, knowing that breaching a target may not lead to regu-latory or government action, especially when the only actionopen is the suspension or cancellation of the contract. It hasalso been the case that operators argue that costs had not beenfully known prior to winning the contract and using renegotia-tion as a way of cutting coverage targets for unprofitable areas.

Investments in expansion yield benefits over long periods,and decisions about their distribution and timing are made onthe basis of expectations about net revenues over the life of the

�,0 '� �� " !�#!��&

�� !��4�� %�� 4�� #

�� !��8� �� 4�� #

�� %#�� 4�� 4��$ ��

��9 '� �� !�"�� !�� 4�� !� ��

�� %�� !�"� ��

�� 9

6�� %� �� %�� 9 2�� !�� !#

�� 4�� !��8� !�"��9 6��

!�� !�� " ��$� � $��

4�� 8� �� " !�� 9 >��

�� !�� !�"�� 4�� #

��" �� " 4��$ �� !�"�� *�� 9

/�� '��*��

/�� '��*��

�� @@��E

concession agreement. This requires a certain amount of fore-casting of future conditions and growth patterns in a givencity; some areas will develop faster than others.

A long-term perspective warrants consideration of changesto tariffs and other contractual terms during the course of aconcession. Periodic negotiations with the government usuallyattract attention from the media, pressure groups, and the pub-lic, not in the least because they concern prices paid for servicesperceived as essential to the public’s well-being. As the operatorexpects the outcome of negotiations to be influenced by out-side pressures, it may decide to alter expansion decisions toinfluence these.

Finally, the private sector operator may look to influencesbeyond the immediate contract. For instance, it may face pres-sure from its owners to make investments that deviate from theprofit-maximizing pattern of expansion. Indeed, the long-termoutlook of a company that owns all or part of a concession candepend on a range of factors beyond the return it gets from asingle concession. It may perceive an interest in maintaininggood standing with certain interest groups, prospective clients,and shareholders—for example by building a reputation as a‘socially responsible’ company. This can translate into decisionsaimed at reducing the return on one contract in an effort toraise it on others.

In sum, the pattern of expansion is determined by many fac-tors, most of which are difficult to influence through transac-tion design and regulation. Tariff structure may play asignificant role, and is certainly an important policy instru-ment. While there is little empirical evidence showing its rela-tive importance, we suggest that tariff structure has a stronginfluence on expansion. IBTs warrant particular attention asthey explicitly match prices with consumption and therebymake one group of customers more profitable than another.

&��

The poor tend to live in areas that are not easily served throughconventional network approaches. As a group, they depend ona variety of service levels (Box 3). Low-income households aretherefore affected when restrictions are placed on the methodsthat service providers are allowed to use, and the levels of serv-ices they can offer. Standards that prohibit the use of materials

-�� !��!�� !��

�� @@��E

or methods needed for expanding in certain areas limit the op-erator’s ability to meet demand. Similarly, a transaction thatmakes alternative providers illegal raises costs and limits theavailability of service levels (Baker and Tremolet 2003).10

The network operatorAs seen, prices charged by an operator are controlled in order toprevent non-competitive pricing due to the operator’s monopo-listic position and, as such, standards are far from irrelevant. Itfollows that some level of quality must be specified in order tomake the price regulation meaningful. Otherwise, the operatorcould use its monopoly power to lower the quality as a way ofraising net revenue.

A further justification for setting a minimum quality of serviceis to avoid negative externalities associated with poor publichealth or environmental degradation. Especially where low wa-ter quality and interruptions in service could lead to an out-break of disease.

10 These issues are also discussed in detail in Johnstone and Wood (2001).

�,� >�" �� !�� !�� !��&

'� � ��! �� !�� %�� #!��

!��" �� %��9 /�� !�� %�� !�� #

�� 4" !��9 2�� 4�" �� 4�� #

�� "�� !9 5��" ��

��#�� " �� %#�� %�� !�4��

��!��9

)�� "!�� !�� %�� #

�� %�� 4� �� %�� B��" �� !��

��B��" �� 4�� 4��" ��

��9 6�� % �� " %�� %�� 4��

�� $ %�� !��4�� %��

�� %�� 4� �4�� % �� !�"��

�!�� !�� 9

/�� 4�� !��4�� 4" ��#�� #

�� 4� �� 4" ��%#�� 9 <� �*�#

!�� " 4��

�� %�� !�� !�� " �� " %��

��$ �� 9 /��" ��%#�� #

�� !�� !�� 4��" %��

%�� #�� " 4� �� %�� %�� B��"9

/�� '��*��

�� @@��E

Standards should be carefully managed because they may beburdensome for the poor. Higher levels of service quality areassociated with higher costs, which are inevitably passed on toothers—often to consumers through tariffs. At some point thesebecome unaffordable for the poorest, who are among the firstto turn to cheaper services that go unregulated for quality. Thiscould reduce the extent of service which ultimately is the meansby which public health and environmental goals are achieved(Brook and Smith 2001).

One way to increase the efficacy of standards, while mitigatingtheir burden, is to focus on outputs rather than inputs. Mostcontracts specify goals for water quality, continuity, and pres-sure. Since these deal with the quality of the end product orservice they can be thought of as output standards. Standards canalso cover technical specifications for engineering works, such asthe minimum diameter and depth of pipes. Use of input stand-ards like these are important in cases where commercial risk isnot passed on to the operator, or where the term of the contractis short. In most cases, however, output standards are a prefer-able way to regulate service qualities as they describe desiredoutcomes, while leaving decisions about methods and means tothe operator.

Flexibility can also be achieved through a careful definition ofcoverage obligations. Allowing the operator a choice of servicelevels to achieve targets will reduce overall costs, while makingpoor areas more attractive. It will have the flexibility to adapt tolocal circumstances (Baker and Tremolet 2003). As an example,concessionaires in Manila have the choice of offering standpostsin low-income areas instead of private connections to meet theircoverage obligations (Rosenthal 2002). For the purposes of cal-culating progress toward targets, each standpost is equated withservice to 475 people or about 100 households.

Alternative providersIncentives can also be used to encourage an operator to workwith alternative providers to install secondary and tertiary net-works. The concessions in Manila, for example, define coveragein a way that residents are considered ‘served’ for the purposesof the contracts no matter who supplies the service. This has re-sulted in the installation of independently operated piped net-works in some areas, using water purchased from the mainnetwork. The concessionaire benefits in two ways from the

-�� !��!�� !��

�� @@��E

arrangement. Firstly, it makes progress towards coveragetargets, and secondly, it is paid for the water.

Facilitating the entry of alternative providers is not always soeasy. Where coverage requirements are coupled with an IBTstructure, operators depend on profitable customers to offsetlosses from unprofitable ones. Unburdened by an obligation toserve the latter group, alternative providers may ‘cherry pick’ theprime areas and clients. Consequently, high-powered contractsoften grant an exclusive right to the operator to serve all custom-ers.

Problems of exclusivity: The problem with exclusivity is that itcan stifle competition and innovation, both of which bring eco-nomic and social benefits that are likely to outweigh any finan-cial savings. As seen, unconnected poor households depend onstreet vendors, tanker trucks, water kiosks and the like. The con-tinuation of these services is important for customers who can-not be scheduled for network connections until several yearsinto the term of a contract—if at all. Therefore, exclusivity isbetter avoided or restricted. If granted and strictly enforced, itwill prevent alternative providers from offering services in areaswhich might never be connected to the network – due to theirlocation, land tenure status or terrain – or where network expan-sion is not scheduled for many years to come.

Exclusivity is usually unnecessary, or can be modified toeliminate its impact on the poor. Since networked services re-quire large investments in infrastructure, there is already a sig-nificant barrier to new entrants who wish to offer the same levelof service as the main operator. Off-network providers mustcompete against prices made low by the network operator’s largeeconomies of scale, so once an operator expands the conven-tional network into a given area, it is unlikely to have its marketshare threatened. If alternative providers can offer a cheaper andmore appropriate service they should not be prevented fromdoing so. Exclusivity is seldom necessary or justifiable.

If warranted at all, exclusive rights are only applicable in pro-tecting the operator’s share of profitable customers. Since poorareas may be unprofitable due to an IBT structure, granting anexclusive right to serve them is pointless. Services to the poorcan be distinguished from services to the non-poor—in mostcases by geography or service type. Exclusivity need not apply tolow-income areas.

/�� '��*��

�� @@��E

One step to reduce the impact of exclusivity is to restrict itsapplication to network services, as this will enable off-networkproviders to continue serving their clientele. Another is to allowon-selling of water past the metered point, in order to facilitatekiosks and other forms of small-scale water vending. The mostimportant change is to allow for exceptions where a third partycan provide acceptable services cheaper than the operator. Thiscan be particularly effective if coverage targets can be met withthe services of small independent network providers. As in thecase of Manila (Rosenthal 2002), the operator may find itselfwith an incentive to encourage the involvement of low-cost pro-viders. A similar analogy would be the community participationseen in the electricity sector in Orissa (Ramanathan andHasan, 2003).

An issue that arises in such situations is the regulation of al-ternative providers. Most operate outside regulatory frame-works and are not committed to official tariffs except aspurchasers. This frees them not just from quality and environ-mental standards (Box 4), but also from the pricing constraintthat often makes low-income neighbourhoods unattractive forthe network operator. This is particularly the case if alternativeproviders are not paying the full economic costs (for example,

�,1 �� B��" %�� %�$ %�� #��

�� " �� %�$ %�� $#

�� !�� %�$�9 >�� " ��!��#

��4��" �!�� B��" ��

�� 4�� "!��" �� 4�"�� !��

��9 �� !��!�� !�� %�� %�� #

�� 9 ' �� B�� !�� #

�� $� ��$� �� " ��%�$ ��4��

%�� !�� 4" � ��" � �� 9 6�� %��

�!�� " �� " �� !�� %��

�� !��9 6�� 9

2!�� !�� %�� !�� B��" ��

��!��9 �� !�� !�� 4��1 �� #

�� " �� 4" �� !��9

>�� %�� !�� B��" �� 4��" �� !��!�� %��

�� !�� " B��" �� !�� " ��

�� %�� %�� 9 �� " !�4�� !��

�� "9

-�� !��!�� !��

�� @@��E

scarcity and environmental value of groundwater or treatmentcosts of wastewater). Thus, the use of alternative providers maynot maximize welfare benefits to society.

While alternative providers may be more flexible and able torespond to the particular characteristics of the market amongthe poor, this does not necessarily mean they are the most effi-cient providers. In fact, if pricing were less restrictive the mainoperator may well be in a position to offer less expensive andhigher quality services than alternative providers. The issue thenreturns to pricing.

"��

As governments contemplate reforms involving the private sec-tor, they are faced with a choice of contractual arrangements.Some arrangements have more potential than others for meetingthe needs of the poor, but realizing this potential requires a well-designed transaction. Efficiency gains must be used to placedownward pressure on prices; incentives for expansion must in-clude poor areas; and service providers should be granted free-dom to meet the specific needs of poor and non-poor alike.

As for doubts about the private sector’s interest in serving thepoor, these are well-founded only where transactions are ill-con-ceived. IBTs, for instance, can make expansion in poor neigh-bourhoods unprofitable and thus undesirable. Exclusivity canlock out service providers that are essential in poor neighbour-hoods. Where operators are not given the freedom to choose howthey serve such areas, their ability and interest in serving thepoor will surely be encumbered.

While technically straightforward, for decision-makers andtheir constituencies this involves new ways of thinking and achallenge to vested interests. They must be prepared to addresspopular concerns about the possibility of allowing foreign com-panies to operate ‘essential services’, fears of higher prices andjob losses, and scepticism about the private sector’s willingnessto serve the poor.

This is not to suggest that concerns about higher prices andjob losses are unfounded. Depending on the situation, efficiencygains may not offset the true costs of maintenance and expan-sion. Prices may rise, and cost-cutting will certainly lead to joblosses. By planning carefully, however, governments can identifycost-effective and socially responsible ways of addressing theseeffects.

/�� '��*��

�� @@��E

��'� ��

The views expressed in this paper are solely those of the au-thors. The authors would like to thank three anonymous re-viewers for the constructive comments provided.

��!��Ackerberg D and Botticini M. 2001Endogenous matching and the empirical determinants of contractformJournal of Political Economy 10 (3): 564-591

ADB (Asian Development Bank). 2002Beyond Boundaries: extending services to the urban poorManila: ADB

Alexander I and Rosenthal S. ForthcomingAllocating Commercial Risk: approaches used in the utility andinfrastructure industries utilizing the basis for remuneration

Averch H and Johnson L. 1962Behavior of the firm under regulatory constraintAmerican Economic Review 52(5): 1052–1069

Baker B and Tremolet S. 2003Regulating the quality of infrastructure services in developingcountriesIn Infrastructure for Poor People: public policy for private provision, edited byIrwin T and Brook PWashington, DC: World Bank

Brook P and Smith W. 2001Improving access to infrastructure services by the poor: institutionaland policy responses[Background paper for the World Development Report 2001]Washington, DC: World Bank

Calaguas B and Roaf V. 2001Access to water and sanitation by the urban poor[Paper presented at the Development Studies Association Conference,Manchester, 10 September 2001, organized by the Development StudiesAssociation]

Coady D, Grosh M, and Hoddinott J. 2002The targeting of transfers in developing countries: review ofexperience and lessons[Mimeo]Washington, DC: World Bank

-�� !��!�� !��

�� @@��E

Collignon B and Vezina M. 2000Independent Water and Sanitation Providers in African Cities: fullreport of a ten-country studyWashington, DC: Water and Sanitation Program

Crane R. 1994Water markets, market reform and the urban poor: results fromJakarta, IndonesiaWorld Development 22(1): 71–83

Delmon J. 2001Water Projects: a commercial and contractual guideNew York: Kluwer Law International

Dnes A. 1995Franchising and Privatization[Public Policy for the Private Sector, Note No. 40]Washington, DC: World Bank [Private Sector Development Department]

Estache A, Gomez-Lobo A, and Leipziger D. 2000Utilities Privatization and the Poor: lessons and evidence from LatinAmericaWashington, DC: World Bank

Foster V, Gomez-Lobo A, and Halpern J. 2000Designing direct subsidies for water and sanitation services–Panama:a case study[Policy Research Working Paper 2344]Washington, DC: World Bank

Gray P. 2001Private Participation in Infrastructure: a review of the evidenceWashington, DC: Private Sector Advisory Services, World Bank

Grossman S and Hart O. 1983An analysis of the principle–agent problemEconometrica 51(1): 7–46

Hall D. 2003Water Multinationals: no long business as usualGreenwich: Public Services International Research Unit

Hardoy A and Schusterman R. 2000New models for privatization of water and sanitation for the urbanpoorEnvironment and Urbanization 12(2): 63–75

Johnstone N and Wood L. 2001Private Firms and Public Water: realising social and environmentalobjectives in developing countriesCheltenham: Edward Elgar Publishing Inc.

/�� '��*��

�� @@��E

Kerf M. 1996Economic regulation of water companies[World Bank Policy Research Working Paper 1649]Washington, DC: World Bank

Kerf M, Gray R D, Irwin T, Levesque C, and Taylor R. 1998Concessions for Infrastructure: a guide to their design and awardWashington, DC: World Bank

Klein M and Roger N. 1997Competition in Network Industries: Where and When to Introduce It[Public Policy for the Private Sector Note No. 104]Washington, DC: World Bank

Komives K. 1999Designing Pro-Poor Water and Sewer Concessions: early lessons fromBolivia[Policy Research Working Paper 2243]Washington, DC: World Bank

Laffont J and Matoussi M. 1995Moral hazard, financial constraints and sharecropping in El OuljaReview of Economic Studies 62(3): 381–399

PCWS (Philippines Center for Water and Sanitation). 2000Vendors and Small-scale Independent Providers: 30 years of urbanwater experiences in the PhilippinesManila: International Training Network

Ramanathan K and Hasan S. 2003Privatization of Electricity Distribution: the Orissa experience,pp. 39–50New Delhi: The Energy and Resources Institute

Rosenthal S. 2002The Design of the Manila Concessions and Implications for the Poor[Background Paper prepared for PPIAF/ADB Conference on InfrastructureDevelopment – Private Solutions for the Poor: The Asian Perspective,Manila, 28-30 October, 2002]

Ugaz C and Price C W (eds). 2003Utility Privatization and Regulation: a fair deal for consumers?Cheltenham: Edward Elgar Publishing Inc.

Van den Berg C. 2000Water concessions: who wins, who loses, and what to do about itViewpoint 217Washington, DC: World Bank

-�� !��!�� !��

�� @@��E

Viscusi W K, Vernon J M, and Harrington J. 1997Economics of Regulation and AntitrustCambridge: The MIT Press

Walker I, Ordonez F, Serrano P, and Halpern J. 2000Pricing, Subsidies, and the Poor: demand for improved water servicesin Central AmericaWashington, DC: World Bank

Whittington D and Boland J. 2000Water Tariff Design in Developing Countries: disadvantages ofincreasing block tariffs (IBTs) and advantages of uniform price withrebate (UPR) designs[Draft report]Washington, DC: World Bank

World Bank. 1997Toolkits for Private Sector Participation in Water and SanitationWashington, DC: World Bank

World Bank. 2001aThe Design of Pro-poor Subsidies in Urban Water and SanitationServices in India: Maximizing the social dividends of reformWashington, DC: Public Private Infrastructure Advisory Facility

World Bank. 2001bContracting for Public Services: output-based aid and its applications,edited by Brook P and Smith SWashington, DC: World Bank

World Bank. 2003Infrastructure for Poor People: public policy for private provision,edited by Brook P and Irwin TWashington, DC: World Bank

WSP (Water and Sanitation Program). 1999Willing to Pay But Unwilling to Charge: do willingness-to-pay studiesmake a difference?[A field note]New Delhi: WSP–South Asia

WSP (Water and Sanitation Program). 2001aThe Buenos Aires Concession: the private sector serving the poor[Private Sector Serving the Poor series]New Delhi: WSP

WSP (Water and Sanitation Program). 2001bDurban Metro Water: private sector partnerships to serve the poor[Private Sector Serving the Poor series]New Delhi: WSP

/�� '��*��

�� @@��E

WSP (Water and Sanitation Program) and PPIAF (Public–PrivateInfrastructure Advisory Facility). 2002New Designs for Water and Sanitation Transactions: making privatesector participation work for the poorWashington, DC: World Bank

' �� (�� 4��

�� "

4�� " ��

��!��" �� !�4��

!��!�� "

��" ��

��

-��

�E !!9

- '4��<�� -�%� /��" ��

�� F��A

This report covers a wide range of issues regarding the function-ing of the ERCs (electricity regulatory commissions) in thecountry. The study includes 13 ERCs and covers issues relatedto independence and autonomy, empowerment, accountability,transparency and public participation, quality of professionals,and social sensitivity of the ERCs.

It clearly brings out that governments continue to hold con-siderable control over the ERCs through the selection processof the chairman and members, financial control, etc. ERCs arenot free to hire staff of their choice at salaries that they com-mand. The staff of ERCs is mostly appointed on deputationfrom the government. The study suggests that these proce-dures be modified and improved to ensure timely appoint-ments with sufficiently long tenures to ensure better selection.

2 � �� ! �� "�

Another aspect highlighted in this study is about the completedependence of the CERC (Central Electricity RegulatoryCommission and many SERCs (state electricity regulatorycommissions) on government funding. The Prayas survey hasrevealed that many ERCs have received less than 70% of the

(��$ ��%

(��$ ��%�

�� F��A

budget proposed by the Commission and some have evenreceived 17%–38% of the budget proposed by them. Financialautonomy of ERCs is necessary to enable them to function in-dependently. Lack of adequate financial freedom is bound tohamper the autonomy and effectiveness of the commissions. Itmay therefore be useful to allow the commissions to levy somefee or surcharge on sales, so that they need not depend on thegovernment for funds.

�� !��

The study brings out a major deficiency in terms of the presen-tation of annual reports to the legislatures. Although the presentlaw requires presentation of annual reports to the concernedlegislatures, many ERCs have not submitted any reports, or thereports submitted after considerable delay. Some annualreports were very brief and sketchy whereas they should havebeen comprehensive and exhaustive in order to instill confi-dence in the general public, as transparency is the hallmark ofreforms. Timely submission of reports would enhance theiraccountability. Many commissions have even failed to conductthe statutory number of Advisory Committee meetings.

The study reveals that most of the documents of the Commis-sion are not available in local languages which would allow thegeneral public to understand the justification and implicationsof various orders. This is true especially in the case of tarifforders, which have a significant effect on the general public.The study suggests that all proceedings of the ERCs should bemade available in local languages to ensure maximum partici-pation of the general public in a vital sector such as power.

��

Submission of accounts and correct information to the regulatoris a major constraint. Non-submission or partial submission ofinformation for tariff fixation, metering, billing and energy au-dit, capital expenditure, and investments, completely exposesthe inadequacy of the utilities in compiling such vital informa-tion. PPAs are also not submitted for scrutiny and the newPPAs are being signed without approval of ERCs, which areserious deficiencies in implementation.

Another issue of concern is non-compliance with the direc-tions and targets fixed by the SERCs while fixing tariffs. Invari-ably these targets have not been achieved, primarily due to the

(��$ ��% �

�� F��A

lack of reliable data with the utilities themselves. The utilitiesfail to submit the required data, undertake studies, and delay inimplementing action of key performance targets. Frivolous liti-gation is another strategy sometimes adopted by the utilities,which could delay crucial regulatory investigations. It is at timeslike these that the utilities are unwilling to allow commissionsto scrutinize their functioning. It is necessary for the utilitiesto comply with the directions and provide severe penal provi-sions for non-compliance of directions of ERC. It has alsobeen observed that many commissions have no library or areading room with all documents of the commissions.

The study also reveals the lack of a proper system to informthe general public of proceedings, other than the public hearingprocess. It cites the example of MERC, which informs recog-nized consumers’ representatives of all proceedings and evendirects petitioners to send all documents to the recognizedconsumer representatives. A practice worthy of emulation. Thestudy also highlights the insufficient use of web sites, asrelevant information is not available on the web site of manycommissions.

The study reveals that the commissions seem to have madevery limited efforts for ‘operationalizing’ the principles of trans-parency and public participation. Very few commissions havetaken any initiative to institutionalize public participation as aresult public participation in the regulatory process isrestricted only to the public hearings conducted during thetariff revision process.

The study shows that the institution of regulatory commis-sions in the power sector has made a good beginning in terms ofbringing in more transparency and public participation. But,the commissions will have to make proactive efforts to ensuremore meaningful public participation and full transparency intheir functioning. It is observed that some state governmentshave not reconciled to the process of independent regulation.Regulatory reforms cannot progress unless there is total govern-ment commitment to regulatory reforms. Support from thegovernment, utilities, and consumers is very essential for effec-tively addressing these challenges.

It is necessary to recognize that reforms in the power sectorwere borne out of compulsion and not by conviction, which mightexplain the tardy progress made in implementing the reforms,though they were initiated nearly a decade ago. The regulatory

(��$ ��%��

�� F��A

legislation is new to India and many ERCs have been created onlyrecently. It is necessary to recognize that the commissions haveto function in totally unfamiliar grounds and there are no prec-edents to fall back upon, which makes their task more complexand difficult. The responsibilities and functions entrusted tothe commissions are enormous and crucial. The systems areyet to be standardized. Many of the commissions are mannedby personnel, who do not have adequate knowledge of thepower sector. Balancing the interests of all stakeholders is not aneasy task. There are many constraints under which the commis-sions are functioning. The study undertaken by Prayas will go along way in improving the functioning of the commissions.

I��%�$�� .��%��

�� /��

)�� 4" ��4�� 5��

/��=�� '" 5��

+�� 0��" -��

6�� I��

HG� !!9

��A

� � I -�� 6)�� I�% +��

�� @��

This book contains a wealth of information about the conver-gence of information and communication technologies, whichit will bring to the information society in the 21st century. It isa magnificent tribute to Melody’s outstanding contribution toboth policy and regulatory research in North America. Thislegacy is brought forth in this book with contributions byMelody’s colleagues and students. The editors have chosensome excellent short contributions made by more than 50 emi-nent economists and telecommunication professionals. Fromvaried areas of telecommunication, IT (information technology),and broadcasting, the contributors have had the privilege ofundertaking policy research under William Melody’s guidance.Articles pertaining to convergence, information societies, tel-ecommunication, economy, and regulation are of significanceto policy-makers and regulators in India.

It was Melody who first challenged AT&T’s contention thatlong-distance telecommunication is a natural monopoly, andtherefore bar MCI from providing point-to-point long distancetelecommunication links to interconnect private networks inthe US. Melody, who worked with FCC (Federal Communica-tions Commissions), made original contributions related tonetwork costing and fixing of retail and wholesale prices byregulators globally.

(��$ ��%

(��$ ��%��

�� @��

The book structured is around five themes – ‘Inquiry’,‘Change’, ‘Next’, ‘Bias’ and ‘What’ – which have been chosen tocapsulate the central ideas behind Melody’s contribution to thecreation, networking, and application of knowledge about in-teractions between changing technologies and societies. As thecontributions are varied and many, only those relevant to theIndian telecommunication sector have been discussed here.

Although Melody contributed to areas beside telecommuni-cations – such as IT, media and broadcasting, etc. – a majorpart of his professional career was spent in dealing withtelecommunication reforms.

"��

The third theme contains interesting articles on telecommuni-cation reforms – for example,‘Utility deregulation in the US’,‘Telecom policy for information economies’, and ‘Toppling ofnatural monopoly doctrine policy and regulatory challenges ofaccess and affordability’ – and an important contribution enti-tled ‘A competitive market approach to interconnectionpayments’.

$��

The contribution entitled ‘Utility deregulation in the US’reviews deregulation of public utility industries. The deregula-tion movement started in the US on the assumption that theemerging competition would put the most efficient infrastruc-ture in place, market power of the incumbent would not besustainable in the long run because of the new players’ entry,and network economies could be easily realized through fullinterconnections. However, real world experience turned outto be quite different. The decline in economic regulation hasseen a parallel growth in market concentration at national andglobal levels. Market concentration in turn has created newopportunities for collusive pricing and investment strategies,which will adversely affect both the industry’s performance andsociety as whole.

Trebing in his contribution points out that due to concentra-tion of market power in telecommunications among a few play-ers with deep pockets and a capacity to cross-subsidize oneindustry segment with another will result in oligopoly.Oligopoly market structure in areas such as IXCs (interexchange carriers) in the US, and NLD (National Long

(��$ ��% ��

�� @��

Distance) carriers in India will try to gain lead in prices. Heillustrates this point with some interesting examples of collusivebehaviour in the power sector where a shortage of electricity wascreated artificially by electricity companies in California resultingin overcharging of customers. The telecom sector too raised theprices of the long-distance carrier artificially despite thesignificant fall in the access fees paid by them to local phonecompanies in the US. Trebing states that although oligopoly fa-cilitates investment, it increases market dominance of a fewplayers through mergers, acquisition and a variety of alliancesand joint ventures. There has been horizontal, vertical, andconglomerate mergers in electricity, telecom and natural gas inthe US. The IXCs in telecom, however, have merged vertically,seeking to reach their final customers directly. All these actionshave reduced the level of competition, which was the objectiveof deregulation in the first place. Based on the analysis of thederegulation movement during the past 20 years, the authorpoints to the need for critical industry analysis and new policyreforms. Here the author brings out Melody’s past contribu-tion to policy analysis and research, which will be a rich sourceof knowledge for anyone carrying on this task. According tothe author, of particular value will be Melody’s effort to createmeaningful cost of service standards to control crosssubsidization as part of the public policy reform. Trebing high-lights the urgency of both task by stating that electricity andtelecom industries are in the state of crisis in the US. He brieflytouches upon the Enron scandal, which has shown that marketfailure and remedial public intervention must be considered inholistic context.

Another interesting contribution in a similar light related tothe telecom policy for information economies is the contribu-tion by Pisciotta who points that mere deregulation is notenough. He says that the new information services provided bythe so called enhance service providers, who have been classi-fied as end users are exempt from access charges. They are per-mitted to use inter-state access services under local tariffs. Thishas skewed the interconnection service market, as some of themprovide service through the Internet and are treated differentlyfrom the basic service providers. After discussing the difficultiesfaced by the US in regulating information, telecommunicationand cable services, the author concludes that information andtelecom policy should ideally be zero-based. He explains this

(��$ ��%��

�� @��

concept by stating that the regulator should assume as if therewas no public telephone voice legacy based on this assumptionand develop an understanding on how the information net-work operates. He also makes a case incentive based on regula-tion. To spread information services, the author recommends anon-discriminatory interconnection for all technologies/serv-ices. This will require an intensive regulatory focus on stand-ards of interconnection as well as swift resolution of disputes.His conclusions are quite relevant to the Indian telecom mar-ket where interconnection usage is a current topic.

Another interesting contribution relevant to the Indian mar-ket is entitled ‘Policy and regulatory challenges of access andaffordability’ by Gillwald. Gillwald states that throughout thedeveloping world, monopolies have failed to meet the mandatesof universal and affordable service. This was due to the inabilityto provide quality service norms and incapability of productinnovation. The first round of privatization and liberalizationin many developing countries has not demonstrated significantgains related to universal access and service. Despite privatiza-tion of basic services in 1994 in the Indian telecom sector basicservice operators have made insignificant contribution for rais-ing the tele-density in rural and remote areas, as well as in pro-viding the much needed VPTs (village public telephones).BSNL continues to provide most of the basic telephones inuneconomic areas and uneconomic planes in economic areas.Majority of private basic operators have failed to fulfill eventheir contractual obligation of VPTs.

��

The contributions related to spectrum allocation – by Caveand Ure – are of particular relevance to Indian policy-makers.This is so because of the government’s decision of a unified li-censing regime for fixed services to end the impasse created bycourt cases between fixed and mobile operators. The first contri-bution entitled ‘Spectrum allocation controversies’ by Cavebrings out the pros and cons of the two methods of frequency allo-cation; one based on administrative assignment and the otherbased on ‘auctions’. Melody in his work on spectrum had identi-fied the fundamental nature of the problem that is the inefficiencyassociated with the administrative method of allocating frequen-cies. Melody recommends administrative pricing and a combina-tion of markets with increased flexibility for spectrum use.

(��$ ��% ��

�� @��

The second contribution on spectrum entitled ‘3G auctions:a change of course’ discusses the telecom industry debacle,which was the result of the European auction of 3G (third gen-eration) mobile telephone licences. The author who is a critic ofthe auction believes that managers and owners of 3G compa-nies in European countries who bid for astronomical amountsfor the spectrum were incapable of working out the real esti-mate of the licences. This was mainly because of the uncer-tainty surrounding the business of 3G. The author believes thatthough the auctions were held in 2002 no one knew what serv-ices would be available, which of them would sell, who wouldbuy or how the revenue would be collected. The author hasgiven a mathematical model to determine the correct marketprice of the spectrum. Hopefully, this will help future operators.As a policy decision on 3G services is still pending, these twocontributions on spectrum allocation and management will beof great interest to Indian policy-makers and regulators.

��

A very topical contribution is by Gabel entitled ‘A competitivemarket approach to interconnection payments’. It will be of agreat interest to TRAI (Telecom Regulatory Authority of In-dia), which is grappling with the task of setting the correct pricefor interconnection under the IUC (interconnection usercharges) regime. The author in the introduction points out thatestablishing the price for interconnection is a challenging taskfor regulatory commissions globally. He says that such com-missions have come under pressure to adopt interconnectionarrangements that set termination charges at zero. This is dueto the perception that regulators in the past were unable to esti-mate accurate costs and, therefore, chose interconnectionprices that were too high. Gabel in his contribution discussesthe concept of ‘bill-and-keep’ where the calling party’s networkdoes not have to pay for the receiving party’s network termina-tion charges.

The author also discusses some practical constraints of in-terconnection arrangements based on cost and benefit of a call,involving the calling and the called party. After examining thevarious pricing models, Gabel concludes that the interconnec-tion payment schemes should be based on market forces. Healso recommends the ‘calling party pays’ principle for all types

(��$ ��%��

�� @��

of calls instead of the even distribution between the caller andreceiver. His conclusion seems to vindicate some of the recentdecisions of TRAI to introduce the calling party regime forfixed to cellular calls, where the called party was required topay for the called mobile leg. This was an exception to the call-ing party principle, which has been applicable to all networkcalls since the advent of PSTN (public switched telephonenetwork) about 100 years ago.

2 ,�

Section 4 entitled ‘Next’ contains 11 interesting contributions.These contributions deal mainly with the advent of the infor-mation society in the 1980s. Although Melody was mainlyengaged in preparing the testimony for regulatory agenciesduring 1960–80, he was concerned with the impact of informa-tion services on the functioning of societies. Melody foresawthe profound effect availability of information throughnetworking would have on all institutions. The contributionsunder this section have been made by either Melody’scolleagues or students. They have brought to light the unfold-ing of the information society and the unequal distribution ofinformation. This brings its own risks.

Some of the contributors are from developing countries,such as Sanatan who worked in the Caribbean region and wasresponsible for telecom reforms. He had worked with Melodyto develop training programmes to tap the potential of infor-mation and communication technology. Like Melody, Sanatanwants equitable reform that will sustain the growth of econo-mies in the region. Sanatan, in his article, reveals thedifficulty of capacity building in small, low-income countries.

Sanjay, in his contribution, brings into sharp focus the risksdeveloping economies like India face if they see investment inthe new technologies as a panacea for all developmental prob-lems. His article reflects the observation made by WilliamMelody in 1985 in which he states that third worldnations might face the risk of instability associated with infor-mation technology markets. Sanjay highlights the problemsthat could arise as a result of commercialization of scientificdata. He concludes that investment in telecom infrastructurecannot serve in a straightforward way as a catalyst for develop-ment. He believes that many other measures are required.

(��$ ��% ��

�� @��

In a similar vein, Jussawalla’s contribution contrasts thepotential of new technologies to enhance global informationflows with the risk of the expanding digital divide. CitingChina’s example, she explains that the national institutions arestill able to influence investment strategies. Just as Melody, in1989, had argued that social and economic diversity wouldmean diversity in the institutions of regulation and market lib-eralization, Jussawalla shows that China and other countries inthe region are seeking distinctive ways to manage their partici-pation in the ‘new economy’. Advances in the informationinfrastructure in these countries tend to aggravate politicalconcerns regarding potential threat to sovereign decision mak-ing. Despite a growing view that nations can no longer enforcedistinctive policies in the face of globalization, Jussawalla iden-tifies an emergent and distinctive policy regime in China.

��

The fifth section entitled ‘Bias’ is inspired by Innis’ conceptwho refers to bias as an emphasis on one aspect to the neglectof another. Harold Adams Innis was a Canadian politicaleconomist, an economic historian, and a communicationscholar. The authors feel that Innis’ work has influencedMelody and many of his colleagues. Some of the contributors,like P Preston, identified Innis’ work as the starting point for theanalysis of issues on information society. Preston and Comor intheir contribution have highlighted the paradox of information‘…more information everywhere but less knowledge; more chan-nels for communication, but less interaction…’.

The contribution entitled ‘C for convergence (and commu-nication, content and competition)’ by Wyatt is of specialrelevance to Indian policy-makers with regard to the commu-nication convergence bill, which was introduced in the IndianParliament last year. The author provides details of her interac-tion with Melody whom she met in 1986 while interviewing fora programme on information and communication technology.She brings to light the contribution by Melody and how heimpressed upon the author the long-term economic impact ofthe convergence of information and communication technolo-gies. She goes on to say how Melody focused not only on thetechnical aspect of convergence but rather lay emphasis on theimportance of information and communication processes ofeconomic and social reform in a country.

(��$ ��%�

�� @��

%��

The last section entitled ‘What’ deals with the circulation ofknowledge—from production to distribution to consumption.These aspects of knowledge economy stressed upon by Melodyas early as 1977 have been highlighted by the authors in thissection. Melody’s analysis of these developments are closelyrelated to regulatory reform and to his critiques of the emerginginformation society. He examined advertising and public servicebroadcasting markets where he found evidence of market domi-nance or practices that he argued were not in the interest of thepublic, and called for public policy reforms in this sector. Thecontributors in this section have shown concern regardingpolicy objectives and rationale for regulation of media indus-tries. The contributions made by them draw upon experiencesin the national contexts of Canada, India, Israel, Switzerland,the UK, and the US as well as Latin America. Silverstone callsfor a fundamental re-examination of the rationale for mediaregulation. Smith like Silverstone emphasizes the need for amoral stance, one that acknowledges the role of the media inassisting the construction of a shared media. He points out thatwith the intensification of global distribution of media contentfacilitated by satellite and internet, media regulation and copyright protection are global and no longer the concern of natu-ral regulates alone. Melody in his contribution on media policyhad examined this issue about two decades ago and observedthat television programmes produced for the international glo-bal market must be different from those produced for the do-mestic market.

Agarwal from India, in an interesting contribution, exam-ines the issue of digital divide in the south Asian region, whichwill manifest in creating divisions in societies between thosewho have access to information and those who do not. Heargues that policy-makers are ‘indifferent’ towards measuresthat might facilitate greater equity in the distribution of poten-tial benefits of digital infrastructure. He calls for a vision of adistinctive Indian information society that can guide a coordi-nated policy action. The last contribution in this section byTurow and Ribak cautions against the tendency to treat theInternet and the World Wide Web as undifferentiated phenom-ena. They argue that many comparative studies oversimplifyand fail to acknowledge that these media are fundamentallyassociated with identity formation.

(��$ ��% �

�� @��

"��

After going through all the five themes, one can conclude thatMelody made wide-scale contributions not only to telecom re-forms but also to the rise of the information revolution andconvergence. These have established him as one of the originalcontributors to policy reforms in the area of telecom informa-tion, media, and public broadcasting in the 20th century.

The editors of the book need to be congratulated on theirchoice of excellent contributions, which are short and interest-ing and provide an insight in to many global policy andregulatory issues.

mr ian alexanderbookstore.teri.res.in/docs/journals/regu-prelims 3-1 june...private bag rondebosch...

Documents