This toolkit was made possible through financial contributions from the Netherlands Ministry of Foreign Affairs.

Labor Issues in Infrastructure Reform

A Toolkit

MODULE1

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

MO

DU

LE 1

Labor Issues inInfrastructure Reform

A Toolkit

© 2004 The International Bank for Reconstruction and Development / The World Bank

All rights reserved.

1 2 3 4 07 06 05 04

The findings, interpretations, and conclusions expressed herein are those of the author(s) and do not neces-sarily reflect the views of the Public-Private Infrastructure Advisory Facility (PPIAF), or the Board ofExecutive Directors of the World Bank, or the governments they represent.

Neither PPIAF nor the World Bank guarantees the accuracy of the data included in this work. The bound-aries, colors, denominations, and other information shown on any map in this work do not imply any judg-ment on the part of PPIAF or the World Bank concerning the legal status of any territory or the endorse-ment or acceptance of such boundaries.

Rights and Permissions

The material in this work is copyrighted. Copyright is held by the World Bank on behalf of both the WorldBank and PPIAF. No part of this work may be reproduced or transmitted in any form or by any means,electronic or mechanical, including copying, recording, or inclusion in any information storage and retrievalsystem, without the prior written permission of the World Bank. The World Bank encourages disseminationof its work and will normally grant permission promptly.

For permission to photocopy or reprint any part of this work, please send a request with complete informa-tion to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; telephone978-750-8400; fax 978-750-4470; www.copyright.com.

All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of thePublisher, World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax 202-522-2422; e-mail pubrights@worldbank.org.

ISBN 0-8213-5470-1

Library of Congress Cataloging-in-Publication Data

Labor issues in infrastructure reform : a toolkit / Public-Private Infrastructure Advisory Facility (PPIAF).p. cm.

Includes bibliographical references.ISBN 0-8213-5470-1

1. Public works—Management—Handbooks, manuals, etc. 2. Public contracts—Handbooks, manuals, etc. 3. Labor contract—Handbooks, manuals, etc.4. Public works—Personnel management. 5. Public works—Employees. 6. Public works—Costeffectiveness. 7. Contracting out. 8. Infrastructure (Economics) I. Public-Private Infrastructure Advisory Facility. II. World Bank.

HD3850.L23 2003352.7'7268—dc22 2003061161

Contents

ACKNOWLEDGMENTS v

HOW TO USE THE CD-ROM vi

MODULE 1OVERVIEW OF PPIAF LABOR TOOLKIT 1

Introduction to the Toolkit 1Objectives 1Structure of the Toolkit 2Support for and Contributors to the Toolkit 3

Labor Issues In PPI: An Overview 4Labor Issues in Infrastructure Enterprises 4Labor Impacts of PPI 8Dealing with Labor Issues In PPI 10

Defining Objectives 12Assessing the Size and Scope of Labor Restructuring 14

Staff Audits 14Benchmarking 15Work Force Analysis 16

Developing Strategies and Options for Labor Restructuring 17Timing and Sequencing Issues 17Restructuring Options 19

Developing Key Elements of a Labor Program 20Severance Payments 20Pension Arrangements 24Redeployment Programs 27Employee Share Ownership Plans 29

Managing the Restructuring Process 30Engaging with Stakeholders 30Implementation Arrangements 34

Monitoring and Evaluating Labor Programs 37Objectives 37Concepts and Approach 38

Integrating Labor Programs in the PPI Process: A Road Map 41

Phase 1: Initial Assessment 42Phase 2: Design of the Labor Program 44Phase 3: Implementation 44Phase 4: Monitoring and Evaluation 44Additional Material (on the CD-ROM) 45Background and Overview Articles 45PPIAF Case Studies of Labor Issues in PPI 45Web Sites 45References 45

FIGURES1.1 Organization Chart for a PPI Team (Sample) 351.2 Road Map for Labor Adjustment 42

BOXES1.1 Technology and Reform in Ports 51.2 Argentina Rail: Crisis and Reform 61.3 Evidence of a Public Sector Wage Premium 81.4 Labor Market Flexibility and Work Force Adjustment—A Snapshot in Estonia and Slovenia 111.5 Generic Labor Benchmarks 141.6 Brazil—Work Force Analysis in Rail Privatization 161.7 World Bank Support for Severance 231.8 South Africa—Pensions at Johannesburg Water Company 251.9 Bolivia’s Capitalization Program 261.10 Brazil Rail—Pension Reform and Labor Adjustment 261.11 Active and Passive Labor Market Programs 271.12 Malawi—Experience of Consulting with Labor in Privatization 301.13 Key Points for Stakeholder Analysis 321.14 Côte d’Ivoire Railways—Participatory Processes 341.15 Port of Santos, Brazil—The Special Labor Fund 361.16 British Coal Enterprise—Privatizing Redeployment 371.17 Monitoring vs. Evaluation 401.18 Guidelines for Submissions to Decisionmakers 43

TABLES1.1 Standard Severance Formulas—Advantages and Disadvantages 211.2 Benefits and Costs of Participatory Engagement Processes 311.3 Understanding Stakeholder Interests 331.4 Analysis, Monitoring, and Evaluation in Labor Adjustment Programs 38

Contents

iv

Acknowledgments

v

This toolkit was funded by the Public-PrivateInfrastructure Advisory Facility (PPIAF), a multi-donor technical assistance facility aimed at help-ing eliminate poverty and achieve sustainabledevelopment through private involvement ininfrastructure. Funding was also provided by theWorld Bank and the Netherlands ConsultantTrust Fund.

The toolkit was managed and led by SunitaKikeri of the World Bank’s Investment ClimateDepartment. Olaf Smulders, a consultant fundedby the Netherlands Consultant Trust Fund, pro-vided research and analytical support. The toolk-it was prepared by the Adam Smith Institute(ASI) of the United Kingdom. The team com-prised Terry Green (team leader), BrendanMartin, and Roberto Battista, and was super-vised by Roger Usher, ASI Director.

The toolkit benefited from the inputs of representa-tives from international development institutions,the labor movement, and colleagues in the WorldBank Group. They included Alberto Chong (Inter-American Development Bank); Antero Vahapassi(Asian Development Bank); Cleopatra Doumbia-Henry (International Labor Organization); MikeWaghorne (Public Services International); PeterBakvis (International Confederation of Free TradeUnions); David Cockroft and Stuart Howard(International Transport Workers’ Federation); andAmit Dar, Amy Luinstra, Anita Schwarz, BillKingdom, David Fretwell, Gordon Betcherman,John Speakman, Jordan Schwartz, Lou Thompson,Marc Juhel, Martin Rama, Omer Karasapan,Richard Hinz, Robert Mertz, Warrick Smith, andYash Pal Kedia (the World Bank). Maria Honradoand Rosario Bartolome provided logistical support.

The toolkit’s CD-ROM is designed to be intuitiveand easy to use. It works like a Web site and can beviewed through a Web browser, but it doesn’trequire you to be connected to the Internet, exceptto access external web sites.

To access the toolkit, insert the CD-ROM in yourcomputer. Choose the CD-ROM drive from MyComputer and open the “toolkit.html” file locatedin the toolkit directory. You can then navigatethrough the CD-ROM as you do Web pages.

On the left-hand side of the screen you can see themain navigation panel.

• The Home link takes you back to the begin-ning.

• How to Use the Toolkit takes you to thispage on the CD-ROM.

• The seven modules are the body of thetoolkit. Click on each header to open the listof contents in each module, then click oneach subheading to go to its pages. To col-lapse the subheadings, go back to the Homepage.

• Below the modules you will find a site mapthat gives a graphic overview of the wholetoolkit and how its elements relate to eachother. The map requires the Flash Playerplug-in. This may already be installed inyour browser software, but if you don’t have

it, you can download it free fromhttp://www.macromedia.com/downloads/

• At the bottom there is a search option. Typea keyword relating to the subject for whichyou are looking in the search box and clickon Go. The search will look at the content ofthe toolkit and point you to the documentsand pages that deal with that subject.

The main screen area in the center displays thepages of the module or submodule you haveselected from the left-hand menu. Within the textyou will find links to additional documents andtools as well as links to external Web sites. Thelinks will only work if you are connected to theInternet.

On the right-hand side you will find a link to apage where you can download each module of thetoolkit as a PDF document (which can be printed).To view the PDF documents you need to have theAdobe Acrobat Reader installed on your computer.If you don’t have it, you can download it free fromhttp://www.adobe.com/products/acrobat/readstep2.html

The right-hand menu also has links to the addition-al areas of the toolkit, including references, casestudies, tools, additional materials, and documents.These come in a variety of formats, mainly PDF,but also documents in such Microsoft Office for-mats as Word, Excel, and PowerPoint.

vi

How to Use the CD-ROM

Windows, Word, Excel, PowerPoint, and Internet Explorer are trademarks of Microsoft Corporation. Flash Player is a trademark of Macromedia Inc.

Acrobat Reader is a trademark of Adobe Systems Inc.

MO

DU

LE 1

FRAMEWORKWhere labor adjustments are required it is impor-tant to ensure that labor programs are fullyplanned for in the PPI reform process. Designingand implementing labor strategies are difficult andsensitive tasks, and the challenges facing govern-ments are many: labor opposition, lack of socialsafety nets, and lack of functioning labor marketsamong them. But experience shows that PPI canproceed smoothly if efforts are made early in theprocess to deal with labor issues. Labor programsare most effective when efforts are made to devel-op a strategy that balances the interests of con-sumers in receiving better and more efficient servic-es with measures that provide fair and equitabletreatment for workers, develop a mix of restructur-ing options, compensate surplus employees, helpworkers reintegrate into the labor market, and

inform and involve workers and labor unions inthe reform process.

ObjectivesThe primary objective of the Toolkit is to providepractical tools and information to help policy-makers handle labor issues in PPI. (The Toolkitdoes not address the policy decision to undertakePPI or the other challenges involved, such asintroduction of competitive markets, developmentof regulatory frameworks, and access to servicesfor the poor. Those materials can be found inother toolkits and documents listed in the bibliog-raphy of this Toolkit.) Drawing from availableinformation, practical experiences, and globalbest practices, the Toolkit provides a practicalguide to assist practitioners in designing, imple-

1

The Toolkit providespractical tools andinformation to helppolicymakers handlesensitive laborissues in PPI.

1Labor Toolkit:Framework and Overview

Auniversal concern in infrastructure reforms is the effect such reformshave on labor. State-owned infrastructure firms often employ morepeople than required for efficiency, and often under favorable terms

and conditions of service, leading to lower labor productivity and higherlabor costs than private employers would bear. Some reform, in particularthose involving private participation in infrastructure (PPI), may thusprompt surplus labor and changes in working conditions as governmentsadjust the work force to prepare for PPI, or as new owners or operatorsintroduce efficiency improvements and expose enterprises to greater man-agement discipline, new technologies, and increasing competition.

MODULE

menting, and monitoring labor programs in PPIreforms, thus also helping build capacity in thischallenging area.

Although broad lessons on labor adjustment in PPIare emerging, experience shows that no one strate-gy is universally applicable and the choice of meas-ures depends on country and enterprise circum-stances. Experience also shows that the interests ofall parties need to be carefully balanced to ensureboth good processes and good outcomes. Efforts toprovide fair and equitable treatment for workersmust be economically and financially feasible forthe government and must give private operatorsthe needed flexibility in making employment deci-sions. The Toolkit thus provides a wide range offrameworks, concepts, checklists, model docu-ments, and case examples that together aim to helpgovernment officials make the appropriate choicesfor their circumstances.

The Toolkit focuses on labor issues in PPI, but itapplies equally to restructuring of state-ownedinfrastructure enterprises without private participa-tion arrangements. Such reforms often involve sim-ilar labor issues and many of the approaches andlessons are applicable. Similarly, whereas theToolkit focuses on infrastructure enterprises, it isalso applicable to state-owned enterprises in othersectors of the economy where labor issues are amajor source of concern.

The primary audience for the Toolkit is governmentofficials responsible for preparing and implementingPPI and enterprise reforms. The term “implementingagency” is used to represent this audience. Theagency may be a unit within the ministry of finance,the ministry of economy, the privatization agency,the relevant sector ministry, or the enterprise itself.Wherever the Toolkit is used, it provides guidanceon the policy and implementation challenges thatgovernments face in dealing with labor issues. TheToolkit may also be a reference point for otherstakeholders, including labor, the private sector, andconsultants engaged in this area.

Users of the Toolkit should be better prepared to:

• Understand the benefits, risks, challenges,and key issues related to designing and

implementing labor approaches in PPIreforms

• Choose among available approaches andanalyze their political, social, financial, andeconomic implications

• Develop and implement appropriate laborstrategies that balance the interests of thevarious stakeholders

• Formulate procedures for developing, imple-menting, and evaluating appropriate com-pensation packages and labor redeploymentprograms

• Manage labor issues during the transitionprocess, including determining the respectiveroles of government, the private sector, andlabor in the restructuring process

• Establish dialogue and communicationswith key stakeholders.

Structure of the ToolkitThe Toolkit consists of seven modules as describedbelow.

Module 1—Framework and Overview: This mod-ule provides a summary of the entire Toolkit andsets out the framework for the more detailed techni-cal modules that follow. It provides a decisionmak-ing framework and road map that policymakerscan use to guide them through the process of laborrestructuring. The overview module consists of thisintroduction to the Toolkit and eight other sections:

• Section 2 provides an overview of the mainlabor issues and concerns in PPI, brieflyexamines the impact of PPI on labor, high-lights the broad lessons of experience in thisarea, outlines the broader policy reformsthat can facilitate labor adjustment, and laysout the practical steps for designing andimplementing labor programs that are cov-ered in the rest of the module.

• Section 3 discusses the objectives in dealingwith labor issues in PPI.

• Section 4 examines how to assess the sizeand scope of labor restructuring by conduct-

Labor Issues in Infrastructure Reform: A Toolkit

2

Users of the Toolkitwill be betterprepared to designand implement alabor program.

MO

DU

LE 1

ing enterprise-level staffing analyses basedon functional analyses and staffing norms.

• Section 5 examines strategies and options fordealing with labor restructuring where need-ed, in particular whether to carry out laborrestructuring before or after PPI and therange of options for restructuring labor andthe conditions under which they can be used.

• Section 6 discusses the key elements of alabor program, including severance pay-ments, pension payments, retraining andredeployment support, and employee shareownership plans.

• Section 7 focuses on managing the process oflabor restructuring, particularly issues relatedto stakeholder participation and implementa-tion arrangements for the program.

• Section 8 covers the monitoring and evalua-tion aspects of labor programs.

• Section 9 concludes with a road map for abroader integration of labor issues in the PPIprocess.

Module 1 focuses on the key issues and lessonslearned in each of those areas. It can be read as astand-alone piece that provides an overview of theissues, or the various sections can be read inde-pendently as summaries of the subsequent modulesthat cover in depth the technical aspects and stepsinvolved in designing and implementing variousaspects of the labor program. These modules fol-low the same flow as module 1 and consist of thefollowing:

Module 2—Labor Impacts of PPI: This moduleprovides an analysis of the impact of PPI onemployment, wages, labor contracts, and unionparticipation.

Module 3—Assessing the Size and Scope of LaborRestructuring: This module describes the tools(such as staff audits, benchmarks, and work forceanalysis) available to determine the nature andlevel of labor restructuring requirements.

Module 4—Strategies and Options: This modulereviews issues of timing and sequencing, and the vari-

ous restructuring approaches that the implementingagency can take when designing labor programs.

Module 5—Key Elements of a Labor Program:This module considers the key issues in planningand implementing severance payments, pensionpayments, redeployment programs, and employeeownership share arrangements.

Module 6—Engaging with Stakeholders: Thismodule discusses strategies and tools for involvingworkers, unions, and other stakeholders, as well asarrangements for implementing the labor program.

Module 7—Monitoring and Evaluation of LaborPrograms: This module provides a framework forestablishing monitoring and evaluation systems forlabor programs.

The Toolkit also contains a Web-based CD-ROMwith searchable documents, spreadsheets, sampleterms of reference for obtaining the needed expert-ise to carry out various labor-related tasks, casestudies, and other relevant data. The CD-ROMfollows the module structure described above andprovides a gateway into and an outline of eachmodule’s content, with such supporting materialsas links to the full-print text modules of theToolkit, and checklists, articles, case examples,tools, and links to other relevant Web sites.

Each module of the Toolkit is designed so that itcan be read as a stand-alone piece in its own right,with cross-references highlighted where appropri-ate. Each module ends with suggestions for findingadditional material on the CD-ROM and else-where. Icons in the Toolkit highlight:

• Tools available on the CD-ROM

• Additional material and documents on theCD-ROM

• Links to Web sites

• Other materials and sources of information.

Support for and Contributors to theToolkitThe Toolkit draws on a wide range of materialsand experiences from around the world. The aim is

Labor Toolkit: Framework and Overview

3

Module 1 providespolicymakers with asummary and aframework to helpguide theirdecisionmakingthrough the laborrestructuringprocess.

Modules 2 through 7detail the technicalaspects and steps inthe design andimplementation of alabor program.

The Toolkit iscomplemented by aCD-ROM thatprovidessupplementary toolsand documents.

MO

DU

LE 1

to provide policymakers with practical guidance onhow to deal with labor issues in PPI reforms. Thefollowing organizations’ financial contributionsmade the development of this Toolkit possible:

• Public-Private Infrastructure AdvisoryFacility (PPIAF), a multidonor technicalassistance facility that helps developing-country governments improve the quality oftheir infrastructure through private sectorinvolvement (see www.ppiaf.org/)

• Netherlands Consultant Trust Fund

• World Bank.

The Toolkit was prepared by the Adam SmithInstitute of the United Kingdom, under the man-agement and supervision of the World Bank’sInvestment Climate Department. Representativesfrom the following organizations reviewed theToolkit and provided comments:

• Asian Development Bank

• Inter-American Development Bank

• International Confederation of Free TradeUnions

• International Labour Organisation

• International Transport Workers’ Federation

• Public Services International

• World Bank.

We thank them for their contributions and recog-nize that those contributions do not necessarilyimply endorsement of the final product.

LABOR ISSUES IN PPI: AN OVERVIEWOften protected from competition and subsidizedby their public sector owners, state-owned infra-structure enterprises frequently employ more peo-ple than required for efficiency, pay wages andbenefits that are higher than their counterparts inthe private sector, and have large, unfunded pen-sion liabilities. These factors have led to lowerlabor productivity and higher labor costs than pri-

vate investors could accept. As a result, thoseaffected by PPI often fear that PPI and the associat-ed efficiency improvements will require substantiallabor restructuring, both before privatization asgovernments cut the work force to prepare forreforms and afterward when privatized firms con-tinue to improve productivity. Indeed, PPI—andenterprise reform in general—has often requiredsignificant labor adjustments. But workers havealso gained in some situations as new investmentsand dynamic expansion resulted in the creation ofnew jobs and as productivity improvements led tosimilar or better terms and conditions of service.

Labor Issues in InfrastructureEnterprisesInfrastructure firms vary greatly between countriesand within a single country, but the introduction ofprivate participation generally produces a numberof changes:

• The objectives of the enterprise change, andbroad social and political objectives of pub-lic sector ownership—such as full employ-ment—become less important than efficien-cy and improved service delivery. At thatpoint, governments look to other policies tosecure employment, and the costs of suchpolicies are made transparent or managedthrough the use of explicit budget subsidies.

• New competitive pressures are brought tobear, and private operators can no longerafford to maintain surplus employment or awork force with relatively low productivity.The threat of new entrants forces operatorsto adjust all operational aspects, includingthe work force.

• Contractual and regulatory obligations to pro-vide cheaper, more accessible, and more reli-able infrastructure services, often linked topenalty and incentive clauses, become a majorfeature and a source of pressure on the firm. Iflabor costs are high relative to comparablenorms or if surplus labor exists on the payroll,private operators will seek to cut costs.

Labor Issues in Infrastructure Reform: A Toolkit

4

MO

DU

LE 1

• New financial disciplines are imposed both bycapital markets and by sector regulators, andmanagers must respond to these disciplines.

When firms remain under state ownership thesechanges and pressures also often arise as a result ofsome of the following factors:

• New technologies: New technologies are rap-idly evolving in many of the traditional infra-structure sectors, such as fixed line telephonyand ports (see box 1.1), and they requireadjustments irrespective of ownership.

• Sector reforms: Many infrastructure sectors arefacing profound shifts. Vertically integratedpower companies are being restructured intoseparate companies for generation, supply, anddistribution; national post and telephone com-panies that once offered the full range of serv-ices are being separated. Infrastructure enter-prises must adapt to these changes.

• Increasing competition: Former monopolis-tic providers are being exposed to competi-tion and challenged by new market entrantsor regulatory regimes. These processes arelikely to place pressures on firms in terms ofjob numbers, technologies, working prac-tices, and skills.

• Structural changes in the wider economy:The adjustments that are being driven bychanges within sectors (for example, technol-ogy) or at the micro level within companies(for example, changing work practices) alsotake place against a backdrop of macro-levelchanges within the market for labor, thecomposition of the labor force, and the econ-omy as a whole. These overall structuralchanges in the economy—for example, mov-ing from agriculture to services—can haveprofound effects on the demand for services,the structure of firms, and the work force.

In most enterprises these forces for change shouldresult in a continuous process of work forcerestructuring, but for social and political reasonsthe adjustment process in state-owned firms maybe delayed. The introduction of PPI, which oftenreflects the acute need for reform, tends to serve as

Labor Toolkit: Framework and Overview

5

Longer-termstructural changesin the economy willalso drive workforce restructuring. M

OD

ULE

1

Box 1.1: Technology and Reform in Ports

Containerization, a technological improve-ment in ocean shipping, has revolution-ized maritime transportation. By handling

individual pieces of general cargo loads onlytwice—at loading and at unloading from a con-tainer—less port labor and ship capacity arerequired to transport the same amount offreight.

In a review of the impacts of technology onocean shipping, Talley (1999) reported that theincreased use of containers, coupled with newcargo-handling techniques and work practices,have led to a significant decrease in the demandfor port labor. Huge job losses have resulted,ranging from 40 percent to 60 percent in manycountries. In the United Kingdom dock jobs fellfrom 80,000 in 1967 to 11,400 in 1986 and byanother 44 percent between 1989 and 1992. InFrance work rule reforms introduced in 1992 ledto employment declines of 66 percent at sixmajor ports. In Australia waterfront reforms intro-duced in 1989 led to a 40 percent reduction instevedore labor over a two-year period.

Although dock jobs dramatically declined, long-shoremen unions were reluctant to acceptchanges and negotiated arrangements to pre-serve work. In some cases, work rules, gangsizes, and compensation patterns remained thesame for containerized cargo as for break-bulkcargo. On the Atlantic and Gulf coasts of theUnited States, labor–management negotiationsled to a “50-mile rule” (that reserved for union-ized longshoremen all “stuffing and stripping” ofcontainers in or near ports), guaranteed annualincomes regardless of hours actually worked,and produced agreements that required shippinglines to use union labor for their vessel calls.

In the port of Buenos Aires the combination ofderegulation, competition, and privatization hasled to dramatic reductions in port charges:charges for shipping containers betweenArgentina and northern Europe declined by 30percent to 70 percent in less than two years. Mostof the savings have come from improved laborproductivity. At the port of Buenos Aires totalemployment fell from about 8,000 just before thereforms to 2,500 in 1994 and has remainedaround that level. The liberalization of operatingrules drastically reduced the requirements forstevedores, which also led to higher labor produc-tivity; the weight of cargo per nonadministrativeworker rose from 800 tons in 1991 to 3,000 tonsin 1995 (see Estache and Carbajo 1996).

a catalyst for reform. The case of Argentina’s rail-ways, where employment fell from 92,000 to18,500 after privatization, is an example of dra-matic employment changes arising from the intro-duction of PPI (box 1.2).

There is significant variation among countries andenterprises but generally speaking three labor issuesneed to be tackled in the course of PPI:

1. Employment levels

2. Labor contracts

3. Pension liabilities.

Those three factors are described below.

Employment Levels

Historically, many infrastructure companies haveemployed more workers than they needed to deliv-

er services efficiently and effectively. This wasbecause the public sector often was seen as a vehi-cle for creating jobs in the absence of a private sec-tor, partly for reasons of patronage and partly formeeting developmental or social objectives. Subjectto weak performance incentives and to “soft”budget constraints, public sector managers oftenwere also able to avoid dealing with the difficultrestructuring and adjustment issues that privatesector managers most likely would have beenforced to tackle.

Kikeri 1998

As a result many infrastructure firms have excessmanpower. For example, many African water utili-ties employ more than 10 employees per 1,000connections, compared with a typical 2.5 to 5employees per 1,000 elsewhere in the world

Labor Issues in Infrastructure Reform: A Toolkit

6

Overstaffing hasseveral causes,which oftencombine to ensureits persistence.

MO

DU

LE 1

Box 1.2: Argentina Rail: Crisis and Reform

With 30,000 kilometers of track, theArgentine railroad enterprise,Ferrocarriles Argentines (FA), was the

largest railroad in Latin America and the sixthlargest in the world (after those in China, France,India, the former U.S.S.R., and the United States)at the time of its privatization in 1990. With 92,000employees in 1990, FA was one of the largestemployers in Argentina. FA’s employees were notonly unionized but also very powerful. Over timethe unions intervened in all aspects of manage-ment, including staffing, internal organization, andstrategy. For instance, because passenger servic-es were more labor intensive than was freighttransportation, the unions got FA to pay moreattention to the former than to the latter, eventhough freight was relatively more profitable. Theysuccessfully resisted efforts to streamline FA’soperations through consolidation and rationaliza-tion because it would make some employeesredundant. From time to time the unions broughtBuenos Aires to a halt by going on strike and par-alyzing suburban rail service in the capital. FA’sunions opposed privatization and there was noreason to think they could not veto its implemen-tation by the Argentine government.

In most countries railroads would not haveappeared on the first list of candidates for privati-zation, but the sector did so in Argentinabecause of the heavy demands it was placing onthe government’s out-of-control budget.

Subsidies and grants received by FA made upfully 9 percent of the government’s budget and 1percent of Argentine gross domestic product.Among Argentine state enterprises, FA was thesingle biggest recipient of federal funds. In sucha context only the privatization of giant enterpris-es could make a dent in the government’s finan-cial problems. Railway unions protested the gov-ernment’s plans and went on strike, paralyzingBuenos Aires, which coped with limited com-muter rail service for 75 days. But the govern-ment held firm until the unions cut a deal with it.That deal included an agreement that redundantFA employees could be let go in exchange forone month’s salary per year of service, with nomaximum limit. Because the average worker hadspent 20 years in FA’s employ, the deal wouldcost the government an average of US$10,000per worker. The World Bank helped finance thisprogram through its structural adjustment loan.The Bank’s backing assured workers that sever-ance payments would be prompt and paid in full,unlike previous severance programs run by theArgentine government. Initial staff reductionstook place through voluntary retirement pro-grams because many of FA’s employees wereold. Subsequent reductions resulted from layoffs,concurrent with privatization. The companyreduced its staff from 92,000 to less than 20,000in 1997.

Source: Ramamurti 1996.

(World Bank 2001). Loss-making long-haul carrierAir India had a staff-to-aircraft ratio of 663 work-ers per aircraft in 1997, compared with ratios ofbetween 170 and 340 workers in various SoutheastAsian carriers: Singapore Airlines, Thai Airways,Malaysian Airlines, and Cathay Pacific (India,Disinvestment Commission 1998). In the 1960sBrazil’s federal railways had a staff strength of160,000, which came down to 42,000 before theprivatization transaction began and was furtherreduced after privatization. In Argentina the corre-sponding figures for the railways fell from 92,000to 18,600.

Excess employment results in lower labor produc-tivity and higher labor costs than private investorsmight be willing to bear and thus are often a cen-tral and controversial issue in PPI. Excess employ-ment has led to fears of labor force reductions asgovernments cut the work force to prepare for PPIor as new owners and operators introduce efficien-cy improvements and expose enterprises to greatermanagement discipline, new technologies, andincreasing competition. The fear of job losses oftenbecomes a focus for opposition by workers andtrade unions—and sometimes for popular discon-tent with PPI as a whole.

Labor Contracts

The terms and conditions of employment are oftenstipulated in a contractual relationship between theenterprise and the employees. That relationshipmay be set out in legislation, in standard employ-ment terms for public service workers, in separatelabor contracts for each enterprise, or in individualemployment contracts. Typically, many infrastruc-ture firms are governed by well-defined collectivelabor contracts, partly because the size of the enter-prise demands them and partly because workers insuch sectors as railways, transportation, and powerwere among some of the earliest groups of workersto organize into the trade unions that helped putcontractual agreements in place.

Labor contracts help create acceptable terms andconditions of employment, including the health,safety, or social well-being of the work force, andmay have been negotiated many years previously.

Some labor practices, however, may cause an enter-prise to operate in less productive ways or at high-er cost than is required. For instance, public sectoremployees are often paid better than their privatesector counterparts, particularly at the lower skilllevels, and often receive tangible and intangiblebenefits—such as job security, seniority rights, spe-cial pension arrangements, subsidized housing,health and educational services—that are not pro-vided by private firms (Assaad 1997, Panizza1999). All of these factors have led to a public sec-tor wage premium in many countries (box 1.3).

As industries and technology change and as firmsare exposed to increased competition, change oftenbecomes necessary and greater flexibility in work-ing practices may be needed. These changes mayinvolve market-based and merit-based remunera-tion systems, greater flexibility in the use and allo-cation of labor, and more flexible hiring and firingpractices. Usually both employers and employeesrecognize that changes will be inevitable and sonegotiation and revision is an important aspect oflabor contracts. The process of gaining agreementon change may, however, be a challenge. In somecases there will be a tradition of centralized collec-tive bargaining with one or two trade unions, butin others there will be fragmentation of work forcerepresentation across a much larger number oftrade unions. In such cases the large number ofunions itself often becomes a factor for considera-tion in revising contractual agreements.

Where private infrastructure investors are enteringcompetitive or contestable markets, they may wishto renegotiate or change some of these terms andconditions of service as part of a new employmentagreement. New private owners or managers ofinfrastructure companies also may seek to renegoti-ate labor practices to meet commercial and opera-tional performance objectives, as well as torespond to changing market demands and newtechnologies.

Pension Liabilities

Many infrastructure enterprises have large accumu-lated pension liabilities that have been promised toand earned by current workers under different pen-

Labor Toolkit: Framework and Overview

7

There are severalcases where stafflevels before PPI aretwice those thatexisted after PPI.

A surplus ofemployees,accumulated overmany years, oftenmeans that joblosses becomeinevitable.

Labor contracts canbe highly effective inprotecting workers,but may make itdifficult forenterprises to adjustto newcircumstances.

Large pensionliabilities canthreaten the viabilityof a PPI transaction.

MO

DU

LE 1

sion arrangements. Many of the pension programsare operated on a largely unfunded or “pay-as-you-go” basis in which obligations are treated as acurrent operating expense rather than paid fromreserves or asset pools to which payment is madeat the time a future obligation is incurred. As aresult many firms have a large liability for futurebenefits that are not accounted for and for whichfunds have not been set aside.

Unfunded pension liabilities and other pensionissues can present a significant challenge for infra-structure privatization. Such liabilities, which arelegally enforceable obligations, can be substantialand their settlement can become a major issue dur-ing PPI because investors may be reluctant to takeover an entity until those liabilities are resolved.Voluntary departure or early retirement programscan also put a financial strain on pension plans. InMorocco, for example, the state-owned railwayprovided a generous pension plan with benefitspaid by the railway itself. The system becamefinancially unsustainable over the years and had tobe reformed, not least because a proposed down-sizing program would have made the pension planeven less sustainable (see box 5.14, module 5).There might also be legal issues involved. Attemptsto tackle high pension costs and unfunded liabili-ties in the urban water supply in South Africa, forexample, were subject to legal challenge (see box1.8 in this module).

Those labor factors have led to fears about thepotential negative effects of PPI on labor and havegenerated interest in developing labor programsaimed at mitigating the social impact of reform.The next section briefly describes the evidence con-cerning the impact of private infrastructure partici-pation on employment.

Labor Impacts of PPIA good understanding of the effects of PPI on thework force is essential because the implementingagency has to deal with a variety of interest groupswith a range of beliefs and perceptions about whatPPI will mean for them. Module 2 of this Toolkit

Labor Issues in Infrastructure Reform: A Toolkit

8

MO

DU

LE 1

Box 1.3: Evidence of a Public Sector WagePremium

There is no doubt that some groups ofpublic employees are underpaid. The factthat governments have to offer very gen-

erous terms to persuade workers to leave vol-untarily, however, suggests that the workersplace high value on continued public employ-ment. Even though salaries themselves may below, the overall employment package of benefitsas well as terms and conditions may be attrac-tive to workers—and better than they mightexpect from the private sector. Several countrystudies have revealed evidence of a public sec-tor wage premium. The following are someexamples:

• Assaad (1997) found a public sector wagepremium in Egypt, particularly for publicenterprise workers who earned an average20 percent to 28 percent more than othergovernment workers. The wage premium,plus nonwage benefits equivalent to approx-imately 85 percent of wages, explained theattractiveness of public sector employmentand the long lists of applicants for jobs inthe civil service and state enterprises.

• Bales and Rama (2002) concluded that stateenterprise workers in Vietnam were overpaidby about 20 percent relative to comparableprivate sector jobs. (Their findings werebased on measurable earnings and benefitsalone, and took no account of better jobsecurity, more generous pensions, moreflexibility in work, or lowered effort levels.)

• Analysis by Bhorat and Liou (2002) in SouthAfrica indicated that state enterprise work-ers received a wage premium of approxi-mately 21 percent compared with those inthe private sector.

• Terrell (1993) found large and statisticallysignificant public–private wage differentialsin Haiti, where public sector wage rateswere about four times the average privatesector wages. The high wages in the stateenterprise sector (telecoms and electricity)appeared to arise from an explicit rent orpremium—perhaps because of a process ofdistributing the enterprises’ “monopolyrents” to employees. (Again, the rents didnot include nonwage benefits and are there-fore likely to be conservative.)

describes in detail the evidence of the effect of privateparticipation on employment, labor productivity, payand benefits, work practices, and workers’ represen-tation and rights in infrastructure enterprises. Only abrief summary of the main findings is presented here.

The evidence shows that diverse labor impactsarise from PPI:

• Significant job losses have occurred in infra-structure enterprises, compared with lossesresulting from privatization in general, andwork force reductions of 50 percent or moreare common.

• The greatest impact has been in those sec-tors where demand is declining because ofcompetition from other modes (for example,railways), and in sectors with long traditionsof overstaffing (for example, ports).

• Job losses have been significantly lower, andin some cases negligible, in sectors wheredemand is rising (for example, telecommuni-cations) or where there is demand forexpansion of the network (for example,water and sanitation).

• Labor restructuring is often necessary toimprove the efficiency and competitivenessof state enterprises, regardless of whetherPPI is involved. Indeed, much of the laborrestructuring that has taken place to dateand has been associated with PPI occurredwell before privatization when state enter-prises themselves adjusted their labor forcesto improve performance.

• Where substantial job losses have occurredin state-owned infrastructure enterprises,they generally account for a small percent-age of the total national labor force.

• In some countries, PPI combined with liber-alization has led to net job creation in thesector as a whole (in telecommunicationsand ports, for example).

• Labor adjustments have led to improve-ments in labor productivity, particularly insuch sectors as railways where surplusemployment was high.

• These productivity gains have also resultedin wage improvements for employees whoremained with the firms after PPI, particu-larly for skilled employees. Wages have gen-erally been aligned more with market condi-tions.

• Labor contracts often have been revised inexchange for higher wages. In LatinAmerican countries, for example, work rulesand conditions of service were renegotiated toprovide managers greater flexibility withrespect to decisions on content and pace ofwork, labor allocation, and subcontracting ofsupport and administrative services tononunionized firms and subsidiaries.Although such changes tended to reduceunion influence within the workplace, theyconformed to labor laws and protected manyof the other benefits enjoyed by workers.

In sum, the evidence shows a wide range of experi-ences with respect to PPI’s impact on labor,depending on the initial conditions at the countryand enterprise levels. PPI can bring benefits toworkers through retained jobs, higher pay, andnew job creation as a result of new investmentsand dynamic expansion. But large employmentlosses can and do occur as infrastructure sectorsand enterprises are reformed, regardless of whetherPPI is involved.

Labor adjustments are one of the most sensitiveaspects of PPI and enterprise reform. The process isnot easy and there are many challenges. Becausethey perceive the threat of unemployment and lossof benefits, labor unions and state enterprise work-ers often oppose reforms and that opposition leadsto actions that delay or stop governments from tack-ling infrastructure problems. Particularly in develop-ing countries, these difficulties are compounded bythe lack of unemployment and social welfare sys-tems, as well as the lack of alternative jobs in theeconomy. In such circumstances, the costs of reformfor state enterprise workers can be high, involvingloss of income, uncertainty, and difficulties in jobrelocation. Dealing with labor issues early in thereform process can help mitigate these social costs ina manner that both protects the interests of workers

Labor Toolkit: Framework and Overview

9

New technologies,competition, andsector reform aredrivers of work forcerestructuring.

Private operatorshave strongincentives to reducecosts.

MO

DU

LE 1

and ensures that consumers gain from better deliv-ery of critical infrastructure services.

Dealing with Labor Issues In PPIWhatever type of reforms are adopted, experienceshows that PPI can proceed smoothly if efforts aremade to deal with labor issues early in the reformprocess. Early attention is needed to:

• Assure workers that measures are beingtaken to compensate them for losses and tomitigate the social impact of adjustment

• Ensure fairness and transparency in thetreatment of workers and thus help buildwider support for the process

• Clarify labor liabilities and approaches inorder to attract private investors

• Facilitate reforms and improve enterpriseperformance, including better and more reli-able services, reduced fiscal drain, invest-ment in new infrastructure, and faster eco-nomic growth.

Dealing with labor issues often requires actions ontwo parallel fronts. At a broader level, the reformof labor market and other policies may be neededto promote private sector job creation. Suchreforms and the development of the private sectorhelp facilitate labor adjustments. Although impor-tant in their own right, these broader policyreforms are covered only briefly here; a detaileddiscussion is beyond the scope of the Toolkit. ThisToolkit focuses on the parallel set of actions: thedesign and implementation of specific labor pro-grams that address labor issues in PPI and infra-structure reforms. Such programs can do much tomitigate the social impacts of reform and in somecircumstances can act as a catalyst for broaderlabor market and policy reforms.

Promoting Policies for Job Creation

The lack of alternative jobs for workers adverselyaffected by PPI or by enterprise reforms is a majorconcern in many developing countries. Manyeconomies have been dominated by public sector

enterprises and the vast resources they absorb fromlocal banking systems have often crowded the pri-vate sector out of financial markets, slowinggrowth and the creation of productive jobs by theprivate sector. Labor market regulations—includingrestrictions on hiring and firing—and payroll taxesthat raise the cost of labor have further exacerbat-ed the problems and led to private sector reluc-tance to hire permanent labor. They have also cre-ated difficulties in adjusting the labor force accord-ing to changing economic circumstances.

Labor market rigidities make it harder for workersto move into new jobs as enterprises adjust, andgenerally lead to:

• Greater difficulty in implementing PPIreforms and labor adjustments.

• Greater opposition from labor to enterpriserestructuring.

• A greater chance of limited results fromretraining and redeployment initiatives. Suchmeasures may provide a short-term pallia-tive to labor adjustment, but there needs tobe sufficient responsiveness in the economyto provide new private sector jobs intowhich workers can move.

• Higher economic cost of labor adjustment,because adjustment in the work force as awhole takes longer.

• Higher financial cost because more reliancemust be placed on compensation paymentsto workers displaced as a result of restruc-turing.

Reforms aimed at increasing labor market flexibili-ty and developing the private sector can, therefore,help make the restructuring process easier, enablingworkers to shift more easily from one sector of theeconomy to another (see box 1.4). Typically, suchreforms involve:

• Reducing hiring and firing restrictions bychanging and simplifying labor laws andrules of severance.

• Removing barriers to labor mobility bydealing with housing and other social bene-

Labor Issues in Infrastructure Reform: A Toolkit

10

Broader labormarket policies areimportant but arecovered only brieflyhere.

MO

DU

LE 1

fits, removing location rules on collection ofunemployment benefits, and improving jobregistration efforts. The availability of jobsand the ease of movement in and out of thelabor market determine how quickly enter-prises and workers can adapt following PPIand other broader economic reforms (seeBetcherman 2002 and Freije 2001).

Betcherman 2002

• Improving information dissemination on jobmarkets by making labor exchanges moreefficient and removing restrictions on pri-vate sector recruitment and placement agen-cies.

• Reducing legal, fiscal, and regulatory barri-ers and disincentives to the development ofsmall and medium-size enterprises, particu-larly disincentives to create new formal sec-tor jobs.

• Accelerating the implementation of privati-zation policies. In countries where heavystate intervention continues to stifle the pri-vate sector and crowd out finance andopportunity for private investment, PPI andprivatization, combined with liberalizationand other policies aimed at private sectordevelopment, can help generate employmentgrowth.

Whereas labor market reforms and other reformsassociated with developing the private sector facili-tate labor restructuring, they require attention at abroader level than does PPI and they are not thedirect responsibility of the PPI implementingagency. Moreover, although labor market reformsease the PPI process, they often take time. Becausegovernments usually cannot delay PPI until labormarket reforms are completed, labor restructuringstrategies need to be designed in ways that takeinto account existing labor market constraints. Insome cases these actions themselves can become acatalyst for undertaking the broader reforms.

Developing Labor Programs

In alignment with policies aimed at creating jobs inthe private sector, the development of specific labor

programs as part of the PPI effort can do much tosecure support and mitigate the social costs ofreform. Many countries around the world alreadyhave successfully managed major labor adjustmentprograms as part of the PPI process and there is asubstantial body of international experience fromwhich to draw.

There is no one approach to addressing laborissues in PPI. Every country and every transactionis different. Much depends on the political, eco-nomic, and social conditions, as well as on thenature of the enterprise and the sector in which itoperates. Nevertheless, several general lessonsemerge from the experience:

• Labor issues can and should be addressedearly in the reform process. Labor issues are

Labor Toolkit: Framework and Overview

11

Relatively rigid labormarkets will slowthe process of post-reform adjustment.

Issues of laborflexibility may ariseif the implementingagency needs tonegotiate changes inlabor contracts.

Managers inimplementingagencies can drawon the lessons ofexperience frommany sectors andcountries.

MO

DU

LE 1

Box 1.4: Labor Market Flexibility and WorkForce Adjustment—A Snapshot in Estoniaand Slovenia

The experience of two transition economiesin Central Europe, Estonia and Slovenia,illustrates the consequences of different

approaches to labor policies. In the early yearsafter the fall of Communism in Central Europe,Slovenia took a highly interventionist approachwith significant barriers to job termination, gener-ous severance payments, generous support forunemployed workers, and efforts to support realwage rates. Estonia, by contrast, took a very lib-eral approach, with few barriers to job displace-ment or to new job creation and with modestseverance payments, and gave little support tothe unemployed and no effective wage floor.Unlike Slovenia, the government of Estonia alsoremoved most barriers to foreign investment.These various measures led to markedly differentadjustment paths and labor markets. The transi-tion in Estonia led to a massive increase in work-er flows out of some jobs and into others. Jobdestruction peaked at 10 percent per year in1992 and 1993, but, with a lag of one year, jobcreation also surged to a 10 percent yearly rate.By 1994 the job creation rate exceeded the jobdestruction rate. By contrast job creation rates inSlovenia usually remained below 1 percent peryear, although job destruction rates ranged from3 percent to 8 percent.

Source: Orazem and Vodopivec 1996.

one of the more complex and politicallychallenging elements in PPI, and dealingwith them early can help secure employeesupport and provide a social safety net.

• Governments have an important role to playin the restructuring process in large troubledenterprises. In cases with high levels of over-staffing or difficult labor relations, transfer-ring infrastructure firms with the labor forceintact is not an option because privateinvestors are wary of taking on the burdenof labor adjustments and employees areconcerned that private investors may notprovide an adequate social safety net.

• A mix of options can be used to deal withlabor restructuring. Redundancy is one butnot the only option. Other measuresinclude: voluntary departures and earlyretirement, correcting payroll errors (such asidentifying ghost workers), freezing recruit-ment, enforcing retirement rules, removingbarriers to employees’ departure, and reduc-ing costs of staff substitutes such as overtimeand fee-paid workers.

• Voluntary departures are the most frequent-ly used option. Such schemes are generallyconsidered to be politically and sociallyattractive, but issues of affordability andadverse selection need to be considered intheir design and implementation.

• Redeployment programs have yielded mixedresults, but if properly targeted they canhave social and economic benefits. Betterresults can be achieved by ensuring thatretraining is driven by demand, that it is tar-geted to workers for whom it is most cost-effective, and that nongovernmental and pri-vate institutions are involved in the deliveryof services.

• Funding for meeting the costs of labor adjust-ment must be secured early. Severanceschemes often involve very substantialamounts of money and it is important to giveworkers credible assurances that the fundingfor making timely payments is in place.

• Early communication and consultation withlabor is important. Such efforts help buildunderstanding and support and they assureworkers that adequate arrangements arebeing made. Openness and transparency indecisionmaking on labor issues are impor-tant confidence-building measures. Theimplementing agency should ensure thatlabor adjustment is seen as a fair process,even if job losses are inevitable.

These lessons are discussed in detail in the subse-quent sections of this module. The module showsthat although labor issues are complex and chal-lenging, they need not be an obstacle to PPI. Withclear objectives, careful planning, and adequateresources, labor issues can be handled effectivelywhile implementing government policy on thefuture of infrastructure.

The main steps involved in developing and imple-menting labor programs are dealt with in the sub-sequent sections of this module:

• Defining objectives

• Assessing the size and scope of laborrestructuring

• Developing strategies and options for laborrestructuring

• Developing key elements of a labor pro-gram, including severance, pensions, rede-ployment support, and employee shareschemes

• Managing the restructuring process

• Putting in place monitoring and evaluationsystems.

DEFINING OBJECTIVESWork force reduction is rarely an objective in andof itself. Rather, the objectives of labor restructur-ing are better defined in terms that focus on thedevelopmental, economic, and social goals soughtfrom PPI or from enterprise reforms more broadly.In cases where stakeholder opposition is high, it isparticularly important to communicate these objec-

Labor Issues in Infrastructure Reform: A Toolkit

12

MO

DU

LE 1

tives. If the objectives are not credible, or are vagueor poorly articulated, the implementing agency willhave more difficulty securing the support andresources needed for labor adjustment.

One tool to help clarify objectives is to describecarefully the ultimate desired outcomes, and to usethose as a basis for setting objectives. Possible out-comes are:

• More efficient operations and better servicefor customers: Today’s surplus labor oftenreflects the inability or failure of past man-agers to respond to changing circumstances.These surpluses need to be addressed if futurePPI operations are to become efficient inadjusting to new markets, new technologies,and increasing competition (see module 2).

• Lower operating costs: The overall goal ofmany labor adjustment schemes is to helpmake the infrastructure enterprise or PPIscheme financially viable. Ideally this impliesthat it can reduce costs, including laborcosts, and achieve sufficient levels of opera-tional profitability to yield a return oninvestment and provide for reinvestment inmaintenance or expansion of the infrastruc-ture.

• Better skills mix: A better work force—per-haps one with fewer people who are betterpaid, better trained, and more capable—is acommon objective. The enterprise may beoverburdened with skills that have becomeredundant and may face a shortage of other,perhaps newer skills. For example, intelecommunications firms there may be acritical need to change the skills mix inresponse to new technology and changingcustomer requirements.

• A more adaptable work force: In addition toskill improvements there may be a desire forgreater labor flexibility in work hours, workpractices, or the structure of the work force.

• Better labor relations: There may be a needfor new negotiation frameworks or a newlabor contract to replace older ones andincrease flexibility.

Outcomes may be short-term and urgent (forexample, reducing operating costs), medium-term(improved services to consumers and business), orlong-term (improved international competitivenessin the sector).

In addition to outcome objectives, governmentswill also have process objectives—that is, objectivesdealing with how work force restructuring is to bedone. Such objectives generally aim to ensure that:

• The adverse social costs of labor restructur-ing are mitigated by social safety net andredeployment programs.

• Implementation of the program is fair andtransparent and balances the interests of thevarious parties involved.

• There is sufficient consultation and dialoguewith key stakeholders, particularly unions andworkers, during the restructuring process.

These objectives provide a framework for discussionand debate within government about the tradeoffsthat must be made in designing a labor program,and help identify and make mid-course correctionsduring implementation as needed.

ASSESSING THE SIZE ANDSCOPE OF LABOR RESTRUCTURINGOne of the first steps in developing a labor pro-gram is to estimate the extent of labor restructur-ing that is needed. The implementing agency mustget a clear handle on the work force to determinewhat if any changes are needed in terms of num-bers, skills, and work practices. This usuallyinvolves carrying out a systematic staffing assess-ment that enables the implementing agency to:

• Determine the size and scope of any workforce restructuring, including potentialdownsizing

• Improve targeting in any labor restructuringprogram, including identification of skilledworkers critical to the operations of theenterprise

Labor Toolkit: Framework and Overview

13

The bottom line:although laborissues are oftenchallenging,experience showsthat they can beaddressed in a fairand professionalway and need not bean obstacle to PPIsuccess.

Thinking about thedesired outcomes isone way to defineobjectives.

A reduced workforce is not usuallythe main objective ofa labor program.Rather it is a meansto an end (such asimproved efficiencyor better service).

MO

DU

LE 1

• Enter into more informed negotiations anddiscussions with trade unions and labor rep-resentatives

• Ensure fairness and transparency of theprocess.

There are three tools that the implementing agencycan use to carry out the staffing assessment anddefine the size and scope of labor restructuring:

• Staff audits or personnel inventories

• Benchmarking

• Work force analysis.

In practice these are related and often are usedtogether in a comprehensive analysis.

Staff AuditsA staff audit is an essential first step in assessinglabor issues in the enterprise. Staff audits make anup-to-date analysis of the work force and providethe basis for subsequent benchmarking and workforce analyses. They also create the necessary data-base for analyses of the costs of alternative sever-ance and pension strategies.

Staff audits help to:

• Bring personnel records up to date.

• Identify and eliminate ghost workers whoare on the payroll (that is, workers forwhom salary payments are made but whodo not in fact work in the enterprise). InArgentina’s SOMISA steel company, forexample, the introduction of a plant censuswith photo identification quickly revealedthat 17 percent of the work force wereghost employees (Hess 1997).

• Provide a basis for developing severance andearly retirement options and estimating costsif downsizing is needed.

• Improve the accuracy of subsequent workforce analysis.

• Establish effective record-keeping proce-dures to allow management to control or

regain control of the payroll, operate effec-tive human resource management practices,maintain staffing information and databas-es, and comply with labor laws and regula-tions.

• Provide an accurate basis for review of jobpositions, pay grades, and scales. In someenterprises, harmonization of staff terms is anessential prior task to PPI. This is particularlythe case where a new operating company isto be formed, perhaps drawing staff fromcivil service and public enterprise cadres.

BenchmarkingBenchmarks are fixed pieces of information thatcan be used to make comparisons with other simi-lar fixed pieces of information. The process ofbenchmarking will help identify main problemareas in terms of the competitiveness of staffinglevels and labor productivity. Labor benchmarksare used not only as a one-off activity for workforce restructuring but also as a tool for continu-ously monitoring and improving performance andcompetitiveness (see box 1.5). In practice it is theprocess of benchmarking that generates most bene-fits through the challenging of current norms.

Labor Issues in Infrastructure Reform: A Toolkit

14

Three tools, allclosely related:• Staff audits• Benchmarking• Work force

analysis.

MO

DU

LE 1

Box 1.5: Generic Labor Benchmarks• Gross or net revenue per employee

• Total payroll costs (all employment-relatedexpense) per employee

• Total/functional labor cost as a percentageof sales

• Ratios of headcount by function (manage-ment/operations, customer service/mainte-nance)

• Salary levels by function (adjusted to allowcomparisons)

• Hourly wage rate (standard and overtime)

• Average weekly hours per worker

• Units produced per work hour (unit produc-tivity)

• Product/service line revenue per staff-hour/full-time equivalent employee

• Training in person-days per year.

Benchmarks provide managers with comparativedata on performance and labor productivity.Although like-for-like comparisons are not alwayseasy, there are several sources of information, andbenchmark measures can give the implementingagency crude indicators of the scale of overstaffing.

There are three main types of benchmarks:

1. Internal benchmarks: By making compar-isons within an organization, perhapsamong different offices or time periods, itmay be possible to quickly and easily identi-fy some areas for improvement. An exampleis the approach adopted by Kenya’s electrici-ty distribution company. For each electricitydistribution district, the company identifiedits characteristics (number of consumers,area, length of overhead line, number ofsubstations, energy sales per customer) andfound weighted averages for different classesof staff (engineers, foremen, linesmen, forexample) that enabled it to easily compareareas of different labor productivity.

2. Sector benchmarks: Comparisons betweenenterprises in the same sector provide anoth-er measure. International or regional com-parisons can be used where the PPI enter-prise is a monopoly provider in the country.

3. Functional benchmarks: Organizations inother sectors that have similar functions canbe compared. For example:

• Gas, water, and power utilities mightcooperate in benchmarking of meteringor billing collection procedures.

• Airlines and railways have similarities inmanaging the turnaround and dispatch-ing of aircraft or trains.

• Administrative processes are similaracross many sectors and comparisonscan be made.

Each of the three types of benchmarks has itsplace. In all cases, however, a combination ofmeasurement and process analysis is important foreffective benchmarking. Measurement identifies thegap, but the discussion, debate, and working

through of changes provides the knowledge abouthow to close the gap.

Module 3 provides further details on benchmarksspecific to each sector, as well as guidelines onwhere to obtain benchmarking data and theprocess of carrying out benchmark analysis.

Work Force AnalysisStaff audits and benchmarking are valuable forindicating the size and scope of overall downsizingthat is likely to be needed. These tools, however,still only provide part of the information necessaryfor detailed assessments of downsizing require-ments and methods of selection.

Detailed work force analysis is often needed, asillustrated by the case of Brazil Railways in box 1.6.The purpose of such analysis is to identify staffingrequirements at the unit or operational level. Workforce analysis will help the mangers in the imple-menting agency and enterprise managers to:

• Identify the levels and types of staff neededfor future requirements

• Make more informed decisions on theorganization of severance plans

• Avoid the loss of critical skills (adverse selec-tion).

Module 3 provides a set of tools for work forceanalysis, including analysis and consolidation ofstaff audit and benchmark data, functionalreviews, ratio analyses of staffing data, productivityreviews, age profiling, and supply forecasting.

More fundamental functional reviews are valuable,too, because they challenge the existing organiza-tional structure and norms. They essentially asksome straightforward but difficult questions:

• Is this activity needed at all?

• Should the enterprise be undertaking thisactivity?

• If the activity is needed, is the enterprisereally the best provider of this function?

• If this is a critical activity, is the scale andscope of operations appropriate?

Labor Toolkit: Framework and Overview

15

Up-to-date recordsof personnel and theremoval of “ghost”workers areimmediate benefitsof a staff audit.

A “benchmark” is acomparativemeasure.“Benchmarking” isthe process ofcomparison.

MO

DU

LE 1

These tools and the process involved in implement-ing them to carry out a staffing assessment are cov-ered in detail in module 3. The assessment is gener-ally carried out at the enterprise level. Where thequality of data is poor, the process of data collec-tion and analysis can be time consuming.Moreover, the enterprise may lack the skills orresources to undertake staff assessments. In thesecircumstances the process may require a combina-tion of resources involving enterprise staff and spe-cialized consultants.

Investment in obtaining good staffing assessmentsis usually worthwhile. The assessments form partof the overall due diligence exercise in the PPItransaction and provide a tool to help the imple-menting agency negotiate with workers andunions. More important, they provide an estimateof the size and scope of labor restructuring, whichhelps identify the level and location of any excessstaff and skill deficiencies. Staff assessments, how-

ever, should not become an end in themselves,because actual staffing needs will only be knownduring the PPI process as investors are brought onboard. When the broad estimates of labor restruc-turing needs are known, the implementing agencycan move on to strategic issues of timing andsequencing, and choices about which restructuringoptions to use.

DEVELOPING STRATEGIES ANDOPTIONS FOR LABORRESTRUCTURINGWhen the size and scope of labor restructuring areknown, the next steps are to deal with strategicquestions of timing and sequencing and to chooseamong various restructuring options. Among thekey questions that practitioners usually face are:Should labor restructuring be done by governments

Labor Issues in Infrastructure Reform: A Toolkit

16

Work force analysisis a staffingplanning activitythat is focused onoperational units inthe context of workforce restructuring.

MO

DU

LE 1

In 1992 the government of Brazil included thefederal railway, Rede Ferroviária FederalSociedade Anônima (RFFSA), in the National

Privatization Program. This was the first majorprivatization of public infrastructure services inBrazil. Implementing the proposed privatizationplan required some degree of reduction inRFFSA’s labor force. Although RFFSA hadalready made significant progress in reducing itsemployee headcount, the company’s labor pro-ductivity continued to be low. RFFSA hadreduced its total staff from about 110,000 in 1975to about 42,000 in May 1995. This reduction ledto a substantial increase in labor productivity,from 250,000 to almost 1 million net ton-kilome-ters per employee. This level of labor productivitycontinued to be insufficient, however, not onlywhen compared with similar North Americancompanies but also with recently restructuredand privatized railways in Argentina and Chile.

The strategy to deal with this excess labor had tobe subtle because there were significant differ-ences in labor productivity across RFFSA’s regionsand uniform cuts across the board would notmake sense. The solution was to come up withnew cost reduction plans for each of the sixregional areas to be privatized, based on new

operational procedures, with redundant activitiesidentified by job categories. This was essentially avery meticulous job that required a detailed studybased on international practice. The redundancyestimates were to be conservative because theidea was to avoid second-guessing what the con-cessionaire would actually need, while avoidingforcing the concessionaire to have to re-recruit“fired” workers, as had been the case in Argentinaand the United Kingdom. In addition, the staffremaining at the company at the time of transfer tothe private operator had to be adequate to avoidinterruptions in service. To ensure this, a detailedanalysis was conducted by the regional managersto assess both the staffing needs for each functionand the number of excess workers. By the end ofthis analysis, RFFSA’s management had reason-able estimates of the staff reduction needs in eachregional area. In May 1995 this process led to anemployment reduction target number of 20,000workers. Between May and September 1995,1,953 workers voluntarily decided to leave thecompany so that by the time the first concessionwas announced in September 1995 the newreduction target number was 18,047.

Source: Estache, Schmitt de Azevedo, and Sydenstricker2000.

Box 1.6: Brazil—Work Force Analysis in Rail Privatization

or be left to private investors? and What is therange of available restructuring options and underwhat circumstances are the options best used?

Timing and Sequencing IssuesThe primary strategic decision is whether laborrestructuring should be carried out by the govern-ment prior to PPI, or whether such restructuringshould be left to the private sector after the PPItransaction is completed. There is no singleapproach and countries have followed differentpaths, depending on the timetable and urgency ofPPI, the nature of the labor issues at the enterpriselevel, and the existing legal framework. There arethree options in terms of timing:

1. Leaving restructuring entirely to the privatesector

2. Leaving restructuring entirely to the govern-ment

3. Adopting a mixed approach.

Restructuring by the Private Sector

At one extreme is the option of leaving laborrestructuring entirely to the private sector, on thegrounds that private investors are generally in abetter position to judge the level of employmentand kind of skills needed. This option can workfor companies in which earlier labor adjustmentshave largely tackled any problems of overstaffing,or where prior downsizing efforts have led toestablished processes and norms in place for sever-ance and redeployment, thus making any futurerestructuring by the private investors relativelyeasy. In these circumstances it is possible to transferresponsibility for restructuring to the private sector.

But in infrastructure enterprises with high levels ofoverstaffing and difficult labor relations, leavingrestructuring entirely to private investors may notbe a viable option. Attempts to do so can put thePPI transaction at risk. Where political and laboropposition to PPI are high, private investors arewary of taking on the political burden of carryingout large-scale layoffs and thus are reluctant to bid.

Moreover, when investors have to absorb largelabor liabilities they discount the sale price accord-ingly, leading to lower sale revenues and potentialpublic allegations that assets are being sold cheaply.Alternatively, they may demand government subsi-dies to cover the cost of the liabilities, thus subvert-ing one of the original goals of PPI. Leaving large-scale restructuring to the new private investors mayalso create social problems, particularly whereweak severance laws and social safety nets reducewelfare protection for workers.

Restructuring by the Government

For the above reasons, labor restructuring in large,troubled infrastructure enterprises is often seen as agovernment responsibility, on the grounds thatgovernment involvement is needed to:

• Resolve potential labor conflicts and mini-mize the burden of politically sensitiverestructuring measures on private buyers

• Ensure that the social consequences of laborforce reductions are properly addressed, par-ticularly in ensuring payment of severanceobligations and development of other socialsafety measures

• Increase the attractiveness of the enterpriseand thus the feasibility of PPI. Government-led downsizing is also one of the few priorrestructuring activities that is likely toimprove the price that investors will pay forstate-owned assets. The most comprehensivestudy of factors determining privatizationprices, based on 236 privatizations inMexico, found that a 20 percent reductionin the labor force prior to privatizationwould lead to a 24 percent increase in theprice (López-de-Silanes 1997).

The strategy of leaving restructuring to the govern-ment has been adopted in a number of cases. InArgentina, for example, where surplus staff andstrong unions were major sources of inefficiencies,the railway and energy enterprises undertookmajor employment cuts prior to PPI. The railwaycompany reduced employment by close to 80,000over several years. Similarly, in Brazil more than18,000 of the nearly 40,000 railway workers were

Labor Toolkit: Framework and Overview

17

Functional reviewscan challenge thebasic assumptionsabout theorganizationalstructure of theenterprise.

MO

DU

LE 1

retired or became redundant before the systemswere concessioned. Prior restructuring was under-taken not just to improve the prospects for sale butalso to overcome labor opposition and ensure thatthe social consequences of layoffs were properlyaddressed.

Government-led labor restructuring has its ownrisks and disadvantages, however, particularly interms of cost and adverse selection. Governmentscan be more generous than the private sector insetting compensation payments, leading to over-payment and issues of cost and sustainability ofseverance payments. Moreover, poor targetingtechniques can lead to the loss of the better, mostskilled, and most valuable workers during thecourse of labor restructuring. In the worst cases,workers who took the packages have been rehired,which has created incentives for the best and mostskilled workers to accept severance (knowing thatthey can be employed or rehired easily) and led tothe inefficient use of scarce public funds. The sec-tion on severance in module 5 treats these issues ingreater detail.

A Mixed Approach to Labor Restructuring

To minimize the risks of overpayment and adverseselection, some governments have stayed awayfrom a direct role in restructuring and have adopt-ed more of a mixed approach. In some cases theyhave made the policy decision to grant privateinvestors full flexibility to select the work forcefrom the existing pool of workers according toneed, while the government assumes responsibilityfor developing the labor program beforehand andfor dealing with residual workers. Such anapproach was used in Argentina’s gas company,where employee restructuring was left to the newprivate investors who were allowed to selectemployees and the government provided incentivesand a social safety net for displaced workers.

In other cases both government and investors haveplayed a role in implementation: part of therestructuring has been done by the governmentprior to PPI, targeted at obvious areas of surplus,and part of the restructuring has been done by thePPI investor who is given full flexibility to further

adjust staffing levels after assuming managementcontrol. Such an approach was adopted in the pri-vatization of Argentina’s telecommunications andenergy companies, and in the case of Manila Waterin the Philippines. In these cases prior to privatiza-tion the government established the severance andsocial safety net program, which was used for bothphases of restructuring, and the cost of downsizingwas shared by government and the private investor.

Whichever approach is used, the key is for govern-ment to clearly define the labor program beforePPI so as to assure workers that their interests aretaken into account and to clarify for investors thelabor liabilities involved. Developing the labor pro-gram in advance and letting the new managershandle the actual restructuring process helps withthe political viability of the process, while it avoidsproblems of adverse selection and sustainability ofseverance packages.

Table 4.1 in module 4 summarizes the pros andcons of the three approaches.

When issues of timing and sequencing are dealtwith, the next step is for governments to examinethe range of restructuring options and to decidewhich ones can be suitably applied under existingcircumstances.

Restructuring OptionsA range of options is available for dealing withlarge-scale labor restructuring. The options can bedivided into three broad groups:

1. “Soft” options, which do not introduce ele-ments of incentive or compulsion but relyon the application and enforcement of exist-ing, and therefore less controversial, work-place regulations. Not simply natural attri-tion, these measures include hiring freezes,payroll management, and the transfer ofstaff to other government departments.

2. Restructuring of the workplace: Options inthis group generally fall short of voluntaryor involuntary departure and include meas-ures such as administrative leave, job-shar-ing, part-time work, and, in some cases, the

Labor Issues in Infrastructure Reform: A Toolkit

18

Work forcedownsizing and CEOreplacement aremeasures that tendto raise privatizationprices.

MO

DU

LE 1

shedding of noncore businesses. Some ofthese will be voluntary and others, such asclosure of noncore units, may be seen byworkers as involuntary change, especially ifthere has been little consultation.

3. Retirement and redundancy, which caninvolve:

• Voluntary departure options that provideincentives for people to leave voluntarily,either through an early retirement pro-gram or the provision of generous sever-ance packages. Acceptance of theseoptions is not forced (although in caseswhere workers’ future prospects are verypoor, workers may feel that they havehad little choice).

• Compulsory redundancy options, whereworkers are required to leave employ-ment without their consent.

Several factors influence the choice of options,including the extent and level of labor surplus,existing labor legislation, and the role of unions.Moreover, these restructuring options are notmutually exclusive; in practice they can be used insequence or combined with one another:

• Soft options and workplace restructuringare generally most appropriate for a man-aged process of work force restructuringand downsizing, or in special circumstancessuch as transition economies where rapid,large-scale downsizing is not politically orsocially feasible. In general, these approach-es may work in the adjustment period priorto PPI, but in heavily overstaffed enterprisesin urgent need of PPI such measures aloneare not likely to be sufficient.

• Where there are high levels of surplus laborand PPI reforms are urgent, more drasticmeasures have been required. The mostcommonly used options in this regard areearly retirement and voluntary departureprograms. Particularly in countries withstrong labor unions and weak social safetynets, as well as those where labor legislationprohibits outright layoffs, governments have

resorted to voluntary departure programs byproviding severance pay packages that havegenerally exceeded legally mandated require-ments. The size of the payments has variedwidely among countries and within coun-tries among enterprises, depending on legaland contractual obligations and the strengthof labor unions. Severance payments arecovered in greater depth below.

• Compulsory retrenchment is used when vol-untary departure programs are not sufficient.At Brazil Railways, for example, the restruc-turing program allowed for a two-phasedapproach. In the first phase, redundant staffwere given the choice of voluntary separa-tion with an enhanced severance package.Employees who did not accept the voluntaryplan were laid off with legal entitlementsplus an incentive package equivalent to 80percent of the incentive offered under thevoluntary program. Because of the reducedincentive, most of the employees took thevoluntary program and only a few wentthrough the compulsory retrenchment route.In other cases, compulsory retrenchment hasbeen used because of difficulties in targetingseverance offers or cost concerns associatedwith voluntary departure programs, or whencertain units are closed or spun off.

The choice of strategies and options open to theimplementing agency is often constrained by exist-ing legislation. An early task is to review the legalframework for dealing with labor issues to seewhat is possible and what is not. Sometimeschanges in the legal framework may be needed.Where such changes are difficult to carry out, theimplementing agency may select options (such assoft measures or voluntary departures) that complywith existing legal requirements and so avoid therisk of legal challenges in court.

Thus, the choice of which approaches to use willdepend on the existing circumstances at the enter-prise and country levels, in particular:

• The size of accumulated surplus labor

• The legal framework

Labor Toolkit: Framework and Overview

19

MO

DU

LE 1

• The nature of labor relations

• The presence of established procedures andpractices for handling restructuring.

DEVELOPING KEY ELEMENTSOF A LABOR PROGRAMOnce the broad strategy and options are deter-mined, the next step is to develop the main ele-ments of the labor program. Specific approaches tolabor restructuring are bound to vary from onecountry and enterprise to the next, depending onlocal circumstances. But labor programs typicallyinclude four main components: severance pay-ments, pension payments, retraining and redeploy-ment support, and employee share ownership plans.

Severance PaymentsIn the absence of unemployment insurance andother social security arrangements in many develop-ing countries, severance packages are typically theprimary source of income support during the transi-tion period to alternative employment. As such,they are a central element in any labor program.

Severance packages typically include some or all ofthe following components, each of which is dis-cussed in detail in module 4 of the Toolkit:

• Statutory end-of-service payments, the levelsof which are usually set out in legislation.Statutory payments can include notice period,or payments in lieu thereof; termination bene-fits; gratuity or pension benefits; earned leave;and payment of salary or wage arrears.

• Compensation for enterprise-level benefits,these being payments to retrenched workersfor benefits as set out in the rules for eachenterprise or as part of a formal collectivebargaining agreement. These benefits caninclude housing; medical, education, andother welfare services; access to loans, per-haps at preferential rates; and subsidized orfree food, equipment (for example, tele-phones), or services (such as supply of elec-tricity).

• Ex gratia severance payments beyond statu-tory requirements, which are usually a keyelement of voluntary departure programs.

In addition to providing immediate income sup-port, severance payments facilitate labor supportand allow PPI to happen, and mitigate the socialimpact of layoffs in the absence of unemploymentinsurance systems. The financial and economicreturns can also be high, with short payback peri-ods and increases in the marginal productivity ofredundant staff redeployed to productive activitieselsewhere in the economy.

But the design and implementation of severancepayments is one of the more challenging areas inlabor restructuring. Four main issues typicallyarise:

1. Setting severance levels

2. Developing targeting and selection mecha-nisms

3. Choosing between a uniform approach anda case-by-case approach

4. Financing severance packages.

Severance Levels

For the reasons outlined above, many governmentshave resorted to voluntary departure programs byproviding severance payments that exceed legallymandated requirements. The size of the paymentshas varied widely among countries and withincountries by enterprise, depending on legal andcontractual obligations, the negotiating strength oflabor unions, and prior precedents.

In some cases an established severance arrange-ment may already be in place; in others, the exist-ing scheme is not considered sufficient or theresimply is none, and a new scheme may need to bedeveloped. Usually, severance plans are based on amultiple of years of service and salary, taking intoaccount legal or contractual obligations and, insome cases, prior experiences or precedents in thestate enterprise sector more broadly. Such formulasare easy for managers to use (table 1.1) and arewidespread in both the public and private sectors.As module 5 shows, severance formulas have gen-

Labor Issues in Infrastructure Reform: A Toolkit

20

MO

DU

LE 1

erally ranged from 1 to 3 months of salary per yearof service, with a typical average payment of 1.5months per year of service.

In developing a severance scheme the main chal-lenge lies in devising severance payments that areboth attractive for workers and financially afford-able and sustainable. Severance has to be attractiveenough for workers to leave voluntarily. At thesame time, governments cannot afford to overpaybecause overpayment leads to problems of cost andfinancing and to problems of adverse selection.

In practice, overpayment has often occurredbecause generous payments are seen to be political-ly and socially attractive. In Pakistan, for example,an agreement with the unions resulted in a packageequivalent to five months’ pay for each year ofservice, which neither the government nor the firmscould afford to pay and which subsequently led todelays in implementation. If an overly generouspackage is offered to all workers, there is the addedrisk of adverse selection. In the rail sector inArgentina, for example, the across-the-board offerled to the loss of key staff, which hurt the perform-ance of the newly privatized firm. There is consid-erable evidence from evaluation of downsizing pro-grams that adverse selection can result in the rehir-ing of workers who have received compensationpayments, and that leads to problems of moralhazard and inefficient use of scarce public funds(see, for example, Haltiwanger and Singh 1999).

Haltiwanger and Singh 1999

One way to contain the risk of excessive payoutsand to minimize adverse selection is to set mini-

mum and maximum levels of payment, where aminimum floor can be seen as fair to everybodywhile a maximum cap would ensure that workersand managers with longest seniority are not over-paid. Another way is to make workers who arewithin a few years from retirement, and thus havethe least to lose in terms of future income, ineligi-ble for voluntary schemes or for their severancepayments to decline as they approach retirement.

A third approach is to better tailor severance pack-ages to workers’ characteristics through a loss-based method that takes into account factors suchas seniority, gender, and education. Unlike standardseverance formulas that compensate workers main-ly for past service—where workers with higher sen-iority benefit disproportionately compared withworkers having fewer years of past service andmore years of denied service—a tailored approachthat takes into account factors such as educationand gender can predict the welfare loss of eachworker and compensate the worker accordingly.

The loss-based severance method is more a com-plement to any existing method than a purely alter-native severance method. Its advantages are that itserves as a benchmark to assess possible overpay-ment, helps induce the right self-selection, andhelps contain costs. The approach aims to:

• Identify the factors that contribute to earn-ings loss after retrenchment. In several stud-ies, women’s earning losses have beengreater than those of men.

• Use estimations of earnings functions, basedon data from national household statistics,

Labor Toolkit: Framework and Overview

21

A severancepackage has anumber of elements.Each of these maybe determined bydifferent legislation,regulations, rules, oragreements.

MO

DU

LE 1

Advantages

• Relatively simple to understand, communicate, and implement.

• Attractive to unions because they can negotiatea formula for a class of workers.

• Attractive to government because it can set a single formula as part of a uniform approach.

Table 1.1: Standard Severance Formulas—Advantages and Disadvantages

• Can be easily imported. Managers in a hurry may simply copy formulas from another country or enterprise.

• Can substitute for analysis of actual needs.

Disadvantages

and make comparisons of workers’ existingincomes with alternative private sectorincome sources.

• Create a specific formula to calculate the wel-fare losses that then can be used to determinethe compensation each worker would receive.

The approach has been piloted in Guinea-Bissau,Madagascar, and Tanzania, and is described inChong and Rama (2000). The data requirementsand methodological issues are covered in furtherdetail in module 5.

Chong and Rama 2000

Targeting and Selection

Another way to avoid adverse selection, reduce therisk of rehiring, and contain costs is to identify thework activities and subsequently the work forcecadres to be separated, and then to target the sever-ance offer only to workers whose jobs have beenidentified as redundant (through benchmarkingstudies or functional analysis, for example), ratherthan to offer severance to all employees.

There are other ways to improve targeting andselection as well. These are discussed in greaterdetail in module 5 and include:

• Giving managers the right to refuse anapplication for voluntary departure

• Using objective and independent methods ofemployee selection (standardized servicerecords, assessment committees with inde-pendent members)

• Obtaining contractual commitments byworkers to retire from public service as partof their exit arrangements

• Instituting penalties for the public enterprisein the event of rehiring

• Strengthening human resource managementsystems to enable better monitoring of staffrecruitment and to avoid the rehiring ofworkers who were prior beneficiaries of vol-untary departure or early retirement pro-grams.

Uniform or Case-by-Case Approach?

Where governments are starting a program ofwork force restructuring across a number of infra-structure enterprises or organizations, a criticaldecision arises. Should government adopt a single,uniform approach to compensation that will applyto all enterprises, or should severance packages benegotiated on a case-by-case basis?

A uniform approach may be preferable wherethere are strong trade unions and where a series ofcase-by-case negotiations can result in very highcosts to government. Such an approach wouldavoid a situation where each new award raises theminimum severance level for the next negotiationsand creates a ratchet effect that leads to increasing-ly higher severance levels that eventually becomeunsustainable.

At the same time, some flexibility might be neededto take into account the particular circumstances ofthe enterprise (for example, based on levels of over-staffing or financial performance). One approachwould be for the government to develop severancepolicies and guidelines within which enterprises areallowed flexibility. In such an approach, the govern-ment might finance the cost of severance accordingto the guidelines, and any severance beyond theguidelines would be financed by the enterprise. Thismight allow enterprise managers in more profitablefirms to restructure their work force more quicklyby offering higher payments that can be financedfrom the enterprise’s own resources but within anoverall framework that prohibits excessively highcompensation payments that the rest of the publicsector could not afford.

Financing Severance

The cost of severance payments can be high andfunding arrangements need to be put in place earlyin the process to assure workers that timely pay-ments will be made. In the absence of such arrange-ments government credibility can be at stake.

There are several sources of finance for severanceand for labor programs more broadly:

Labor Issues in Infrastructure Reform: A Toolkit

22

Rehiring is anindicator of programfailure and poordesign.

MO

DU

LE 1

1. The government budget: Government rev-enues can be used to finance restructuringprograms, although these may be insuffi-cient when countries are faced with large-scale severance programs.

2. Disposal of assets: An enterprise that hasacquired nonrelated assets over the yearscan dispose of them through asset sales orprivatization, and the revenues generatedfrom the sales can be allocated to meetingthe costs of downsizing. Problems can arise,however, if:

• Government rules prevent disposalreceipts from being retained by the enter-prise, requiring instead that they be allo-cated to the general treasury account.This is normally the case where the enter-prise is a departmental (civil service)organization, but there may be moreautonomy where the enterprise is a pub-lic corporation or a company.

• The assets are illiquid or difficult to sell,or the market for assets is (temporarily)depressed. This will lead to a timingdelay in sales, or reduced proceeds if theimplementing agency implements a “firesale” of assets, at cheap prices.

• There are legal issues such as a priorcharge on the assets by creditors oruncertainty of title.

• PPI enterprise managers and governmentofficials are reluctant to dispose of keyassets. For example, port managers orbus company managers may resist thedisposal of potentially valuable land forcommercial development.

3. Privatization proceeds: Some countries havesequenced the sale of valuable enterprisesfirst in order to build up adequate fundsfrom initial proceeds to finance the laboradjustment and other costs of more difficulttransactions.

4. Bonds, loans, and grants: Governments canissue government bonds, and some profitableenterprises may take commercial loans to

finance the costs of work force restructuring.Many governments, however, look to loansor grants from multilateral institutions andbilateral donors as potential sources of funds(see box 1.7). For example, World Bankloans, under carefully specified criteria, pro-vided financing for severance packages forredundant workers in the restructuring of thePolish and Brazilian rail sectors and the pri-vatization of Togo telecommunications.

5. Enterprises and investors themselves,through profits from operations: Where thelevel of employment reduction is low, cur-rent operations can finance the costs ofretrenchment. For example, in restructuringthe Lesotho Electricity Corporation (LEC),40 to 50 percent of staff were consideredsurplus—around 250 people in total. Thecosts of severance were funded from LEC’songoing operations. (Other items such astraining, counseling, and communicationwere supported under a World Bank credit.)Where the new investors are responsible forfinancing retrenchment costs after assumingoperations, these costs too are effectivelyfinanced from current or future profits.

Labor Toolkit: Framework and Overview

23

A key decision iswhether to adopt auniform or a case-by-case approach toex gratia severancepackages.

Work on securingfinancialcommitments muststart early.

There are sixsources of funds forlabor programs.

Asset disposal isimportant but cantake some time toimplement. There is,however, nothing tostop theimplementingagency from startingthe process by, forexample,commissioninginitial valuations ofassets.

MO

DU

LE 1

Box 1.7: World Bank Support for Severance

In the past the World Bank was not allowed todirectly finance severance pay because it wasnot considered a productive investment. There

were also concerns about the effectiveness ofretrenchment schemes and the Bank’s vulnera-bility to accusations of supporting and financingunemployment. But a number of factors led theBank to decide in February 1996 to allow directBank financing for severance pay as part ofinvestment operations. These included theimportance of large-scale restructuring and pri-vatization, the potential obstacles arising fromlack of financing for labor shedding prior to sale,the growing evidence on the economic andfinancial returns to severance pay, and the limi-tations of adjustment lending. As a result, sever-ance pay financing can now be provided forindividual state enterprises or groups for enter-prises throughout the reform process; that is,from corporatization to restructuring prior to pri-vatization” (Kikeri 1998, p. 35).

“

6. Creditors: Where the enterprise is liquidat-ed, creditors may ultimately fund the costsof (statutory) redundancy, as was the case inthe liquidation of Aeromexico (see box 4.9,module 4).

Pension ArrangementsSome of the more complex issues in PPI arise fromthe way pension arrangements are addressed. Apoorly conceived pension strategy can make an oth-erwise viable transaction effort untenable. The cost ofpast obligations or future commitments of an exist-ing pension scheme are commonly among the moresignificant considerations in determining potentialinvestors’ willingness to participate in a PPI initiative.

Public sector employees often benefit from gener-ous pension schemes that may far exceed thoseavailable to private sector workers, especially indeveloping and transition economies where pen-sions of any type are often unavailable to themajority of the work force. The value of pensionbenefits can be the largest single component of thecompensation package for many workers and theonly form of “savings” accessible to them. Thecurrent and future costs of sustaining thesearrangements, however, are often so great that theway in which they are structured and financedbecomes a key factor in whether an enterprise isfinancially viable over the long term.

Pensions also represent a potentially powerful toolfor the restructuring of labor. They may be aneffective means to lower labor costs through earlyretirements or voluntary separations and a power-ful incentive to attract and retain highly skilled oressential workers.

The implementing agency must address three close-ly related but distinct matters in dealing with pen-sion issues in PPI:

1. The measuring and resolution of existingpension commitments that have beenaccrued to date in a manner perceived to bereasonable, fair, and financially viable

2. The utilization of pensions in the process oflabor restructuring through limited windowsfor early retirement or voluntary departure

3. The restructuring of pension arrangementsin a way that makes them consistent withthe future requirements and financial sus-tainability of the enterprise.

Each of those issues is complex and its resolutiondepends on the specific type of pension scheme inplace. The issues must also be addressed in waysthat satisfy the differing interests of employees, theprospective PPI investor, and government. Module5 presents detailed guidance on each of these tasksfor different types of pension schemes.

Engaging with Stakeholders

Given the importance of pension provision toworkers’ welfare, it will be essential to communi-cate, consult, and often negotiate with employees,trade unions, pension plan trustees, government(ministry of finance, labor, or social protection),and other stakeholders on changes in pensionarrangements.

Where national or multiemployer pension plansmay be required to commit additional resources tomeeting the future obligations for pensions, it isessential that the relevant government ministry orthe government actuary is consulted before com-municating with unions and workers. Especiallywhen these plans are actuarially bankrupt, orwhere government is about to reform the pensionsystem generally, the implementing agency willneed to be very careful about raising the expecta-tion of workers’ representatives about what is—and is not—possible.

Pension plan trustees or supervisory board mem-bers are important actors and stakeholders in pen-sion issues in PPI. They will be able to inform theimplementing agency about the structure of theplan, past precedents in interpretation of the plan,plan assets, and financial circumstances. As box1.8 illustrates, pension plan arrangements may beboth technically and politically difficult, and theimplementing agency and trustees may find them-selves on different sides of the negotiating table. Itis therefore essential to consult with pensiontrustees. Their primary duty, however, is to thebeneficiaries of the plan, and they may well havetheir own source of independent professional

Labor Issues in Infrastructure Reform: A Toolkit

24

MO

DU

LE 1

Labor Toolkit: Framework and Overview

25

Pension liabilitiesare increasinglyrecognized as apotential risk toenterpriserestructuring.

Pensions issues willbe a criticalcomponent of anycommunicationsplan.

The implementingagency will usuallyneed to enlistspecialist advice onpensions.

If pension systemsare in the midst ofreform, this maywell affect the waythe implementingagency will packageor restructurepensions within thePPI transaction.

MO

DU

LE 1

advice. The implementing agency may also identifya need to recruit specialist professional advice onbehalf of government.

Managing Links with Pension System Reform

For some PPI implementing agencies national pen-sion reform will be of little relevance, eitherbecause the reform has already happened orbecause pension reform is not in the pipeline.

Where government is in the midst of systemreform, however, strategic decisions may need to bemade. Wider pension reforms can affect specificdecisions on pension issues for a power, telecom-

munications, rail, or port PPI, or may set prece-dents for other state-owned enterprises. Until 1981most public pension plans were defined-benefit,pay-as-you-go arrangements, potentially unsustain-able financially. The most well-known earlyreforms were those of Chile in 1981, which intro-duced a system based on individual accounts withfully funded, fully vestable, and fully portable ben-efits, plus minimum pension guarantees for work-ers on low wages and interrupted contributions.Chile’s reform was followed by reforms in severalother Latin American countries. In Bolivia’s capi-talization program (box 1.9), pension reform andprivatization of state enterprises (including utility

The reform of Johannesburg’s water supplyincluded efforts to implement more equi-table and more financially sustainable pen-

sions. The Johannesburg Water Company(JWater) was formed on November 21, 1999, andit inherited staff who were contributing to 12 dif-ferent (defined-benefit) pension plans originallyset up by the City of Johannesburg. Not onlywas this arrangement complex to administer, butthe plans were not equitable, and generallyfavored white employees and senior managers.Moreover, the unfunded liabilities of the planswere large and growing, a situation exacerbatedby high administration costs.

Restructuring of the company’s pensions wasessential. However, union representatives on thepension plans’ boards of trustees had been gen-erally opposed to the reforms of the city’s watersupply. Trustees had not formally recognizedJWater as an employer because the rules of theplans did not generally provide for admission ofnonlocal government employees. Moreover,trustees of the two largest (City of Johannesburg)pension plans were reluctant to make rulechanges, and they used the need for suchchanges as a bargaining tool to obtain other con-cessions from the city.

These problems were not unique to the watersector; other city pension schemes had similardifficulties. In December 2001, in an attempt toresolve citywide pension problems, the cityadministration unilaterally closed all pensionplans to further contributions. A new providentfund was created. This was a defined-contribu-tion (accumulation) plan with a 15 percent contri-

bution by employers and a 7.5 percent contribu-tion by employees, plus life insurance and dis-ability benefits. The city’s unilateral action waslegally challenged in the courts, and an interimorder granted relief to the two largest plans inMarch 2002. Other schemes subsequentlylaunched similar applications. If all are success-ful, then pension provision will essentially beunchanged and the implementation of pensionrestructuring will be dealt a severe blow or atleast delayed.

With hindsight, there had been too little consulta-tion with plan trustees, unions, and workers. Thecourt challenge to the city was based on thegrounds that (a) the interests of some pensionplan members had been prejudiced, (b) the con-sultation process had been inadequate, and (c)earlier guarantees given by provincial legislationand the employer during the reorganization hadbeen infringed. Those pensions issues are notyet resolved, but key lessons for labor adjust-ment in PPI are:

• Pension issues are complex.

• It is important to have early and effectiveconsultation on pensions with workers,unions, and trustees.

• The allocation of pension liabilities must beclearly defined. (In this case the City ofJohannesburg will meet unfunded pensionliabilities up to the time of staff transfer, butthereafter responsibility rests with JWater.)

Source: Personal communication from staff in JohannesburgWater, 2002.

Box 1.8: South Africa—Pensions at Johannesburg Water Company

Labor Issues in Infrastructure Reform: A Toolkit

26

Pension systemreform canencourage moreworkers to opt forearly retirement.

MO

DU

LE 1

and infrastructure companies) were intimatelylinked because the state’s shareholdings were usedto establish new pension schemes.

PPI work force restructuring may also be affectedby the timing of systemwide pension reform. Forexample, labor adjustment in Mexico rail tookplace as pensions were being reformed, whichmeant that workers knew that the new privatecompanies would provide pensions under newmandatory, fully funded arrangements (López-Calva 2001). In Brazil many workers opted forearly retirement rather than severance because ofpension reform (box 1.10). This response is notunusual; several countries have seen surges inapplications for retirement when pension systemsare under review and workers are uncertain of theoutcomes. Depending on circumstances, therefore,PPI labor adjustment may be helped or hinderedby pension system reform.

Consultation with government is important toensure consistency of approach and policy coher-

ence between pension system reform and PPIimplementation, particularly where:

• Government is proposing other changes thatare directionally opposite to those that thePPI implementing agency might propose.For example, pension system reforms caninclude increasing the statutory age andsometimes removing any difference in retire-ment age between men and women (theimplementing agency may in contrast pro-pose early retirement); raising the minimumyears of service for eligibility or changing thebasis of eligibility from years of service toage (although the implementing agency maywant to relax these eligibility criteria inorder to encourage early retirement); orremoving special and privileged pensionrights (although the implementing agencymay need to recognize these rights explicitlyin negotiations with unions).

• Pension systems are moving from pay-as-you-go plans to ones that are at least partial-ly funded to ensure that the systems as awhole are financially sustainable and willnot collapse in 10, 20, or 30 years.

Box 1.9: Bolivia’s Capitalization Program

Bolivia’s 1994 capitalization law is a uniqueexample of a combined privatization andpension system reform program. Major

state enterprises, including gas, telecommunica-tions, railways, airlines, and electricity genera-tion and transmission, were capitalized througha capital increase by private investors of up to50 percent of the companies’ capital. During1995 and 1996 international and domesticinvestors acquired shares in these enterprisesthrough competitive bidding processes.Government transferred the remaining 50 per-cent of the shares in the enterprises into twonew privately managed pension plans that werealso mandated by the capitalization law.

The Bolivian approach meant that there were noprivatization proceeds to the state budget, butthat the investment went directly into theseinfrastructure assets. At the same time, the uti-lization of pension plans helped provide fordevelopment of the local capital market andmeet the social protection needs of the popula-tion.

Source: Guislain 1997.

Box 1.10: Brazil Rail—Pension Reform andLabor Adjustment

In planning for labor adjustment in Brazil rail, itwas expected that about 5,000 workers wouldtake early retirement and 13,000 would

choose voluntary separation. In fact, almost thereverse occurred: 11,771 opted for early retire-ment and 5,886 for separation. The reason wasa proposal in Brazil’s Congress for radicalreform of the pension system, which impliedthat eligibility to retire no longer would be basedon the number of years worked but on age.Fear that changes in the social security lawwould mean that workers would have to workmany more years to receive similar benefits orthat the changes would jeopardize the retire-ment income of older workers led to an unex-pected increase in the number of applicants forearly retirement.

Source: Estache, de Azevedo, and Sydenstricker 2000.

• Governments are moving away fromdefined-benefit plans to accumulationschemes.

• Negotiations on pension reforms are inprogress between government and unions.

Redeployment ProgramsMany governments have supplemented severanceand pension packages with redeployment supportto help workers regain productive incomes—whether through formal employment, self-employ-ment, or informal livelihood activities. Such pro-grams also offer an additional incentive to encour-age voluntary departures and help win support forpolitically difficult restructuring. They are aimed atfacilitating the shift of economically unproductiveworkers from infrastructure sectors to more pro-ductive sectors of the economy. One by-product ofretraining and redeployment programs may be ageneral shift in attitude among the work force atlarge, away from a perceived reliance on publicsector employment toward private sector jobs andself-employment.

In some cases redeployment programs have beendeveloped explicitly for PPI employees; in others,redeployment support is provided as part of broad-er active labor market programs for the unem-ployed. Such programs differ from passive labormarket programs that act more as a safety net forthe most vulnerable people (box 1.11).

The main types of redeployment support are:

1. Counseling, which might include elementsof trauma, financial, and life counseling, inaddition to advice on services and supportopen to the displaced worker.

2. Job search assistance, which can includeplacement assistance (employment interme-diation) to match workers with opportuni-ties in the job market, time off for jobsearch prior to termination of employment,and help in building skills and confidence tofind a new job (through personal skillassessments, coaching, and job clubs).

3. Training, which may have different areas offocus, such as retraining and developingnew skills in workers so that they can findnew paid employment elsewhere; or trainingin small business, microenterprise, or liveli-hoods to help displaced workers find self-employment and incomes.

4. Employee enterprise, whereby opportunitiesand facilities are provided by the governmentor the PPI enterprise to enable employees toset up their own businesses. These facilitiesand supports include contracting out of servic-es by the enterprise to newly separated work-ers; simple workspace facilities (sheds,garages, and small offices); and business incu-bators where workspace facilities are support-ed by business advice, shared facilities (faxand photocopier), and a degree of mentoring.

5. Community-based approaches, which look tolocal government, nongovernmental organi-zations, and community self-help groups,alone or in coalition, to develop employmentopportunities at a local level. These caninclude public works programs that providetemporary employment opportunitiesthrough large-scale, labor-intensive projects.

Labor Toolkit: Framework and Overview

27

The PPIimplementingagency’s plans fortackling pensionsmay run counter togovernment’s overallpolicy.

Redeployment is atool of active labormarket policy.

MO

DU

LE 1

Box 1.11: Active and Passive Labor MarketPrograms

Redeployment support services linked toenterprise restructuring programs are aparticular subset of “active” labor market

policy. “Active” labor market policies or pro-grams are those that directly prepare or assistthe reintegration of workers in the labor marketthrough activities such as job search assistance,job placement plans, training programs, andemployment subsidies. These can be contrast-ed with “passive” labor market policies, whichsupport incomes, usually through financialtransfers. Examples of passive policies areunemployment insurance, worker disability pay-ments, and—relevant to the context of PPI laboradjustments—severance pay. Some have usedthe metaphor of a trampoline to characterize theconcept behind active labor programs (to liftworkers back into work) compared with thesafety-net concept of passive programs.

Counseling, training, and job search lie at the coreof most redeployment programs (see figure 5.2 inmodule 5). Community-based approaches, publicworks, and employee enterprise are supplementaryelements that can be appropriate in some circum-stances. In designing programs, there is oftenopportunity to involve unions, local governmentand, where relevant, wider community and civilsociety groups in consultation processes.

On the whole, redeployment programs have hadmixed results. Most evaluations have focused onthe experience of industrialized countries. Theseexperiences generally show that retraining pro-grams resulted in modest gains in reemploymentprobabilities, but wage changes were negligible ornegative. The same evaluations found that thecosts of retraining are two to four times higherthan job search assistance but are no more effec-tive. By contrast, placement and counseling effortstend to show positive results and are generallymore cost-effective. Systematic evaluations of theimpact of such programs in developing countrieshave been few, although anecdotal evidence sug-gests that retraining programs often fail because oftiming delays, weak institutional capacity, loweducation levels, and the lack of employmentopportunities for retrained workers. The programsfrequently have been more driven by supply thanby demand.

Nonetheless, emerging experience from a widerrange of circumstances can inform the design andimplementation of redeployment programs andhelp workers with several remaining years of pro-ductive life to acquire gainful new livelihoods.Better results are more likely if wider economic andlabor market policies aimed at creating sustainableemployment growth are already in place (as out-lined in the second section of this module). At theprogram level, effectiveness can be improved by:

• Ensuring that redeployment services aredriven by demand rather than by supply.This can be done by giving workers a choicebetween training and severance, and bybuilding in a cost-sharing element throughthe use of vouchers and other instruments.

• Targeting services to workers for whomsuch services are most cost-effective—in par-ticular, younger workers with basic educa-tional levels and skills who are most likelyto improve their labor market outcomesfrom retraining. Effective targeting requiresthat a survey of workers be undertaken as apreparatory step to obtain a clear profile ofworker characteristics and needs.

• Developing a good understanding of thelabor market that workers will be entering.This requires a labor market survey, done asearly as possible. Proper labor market infor-mation is critical for setting up training andemployment support programs that are rele-vant to workers’ opportunities and needs.

• Getting a good understanding of the typesof services, programs, and institutions thatare in place to deliver services. This requiresa survey of existing training and other labormarket infrastructure to determine thecapacities of service providers.

• Developing good counseling and advisoryservices to help match workers to appropriateand relevant retraining and other programs.

• Bringing in a wide range of institutions,including nongovernmental and privateinstitutions, to foster competition and effi-ciency in the delivery of services, and usingperformance-based contracting arrange-ments where possible to improve incentivesand efficiency.

• Giving workers access to information on theperformance of training and other serviceproviders so that they can better selectcourses and services (and at the same timemake the program more demand-driven).

• Providing redeployment support to helpworkers set up their own small businessesinstead of merely equipping them with skillsfor jobs in the formal sector—jobs that tendto be limited.

• Engaging effectively with stakeholders, partic-ularly through consultation (see module 6).

Labor Issues in Infrastructure Reform: A Toolkit

28

There are five maintypes ofredeploymentsupport.

MO

DU

LE 1

Module 5 discusses in detail the different types ofprograms and the key steps involved in designingand implementing redeployment programs. Theseprograms and steps include the use of surveys asbuilding blocks for the program, designing incen-tives to ensure that programs are demand driven,costing out the services and securing fundingarrangements, setting up labor funds or otherimplementation arrangements, identifying andcommissioning service providers, and developingsystems for monitoring and evaluating results.

Employee Share Ownership PlansIn work force restructuring, employee share owner-ship plans can be used in three ways:

1. As a form of compensation to displacedworkers

2. As part of an incentive or reward packageto workers who remain

3. As the basis for management employee buy-outs or employee buyouts of units of theenterprise.

In such arrangements, governments have reservedshares, ranging anywhere from 3 to 10 percentdepending on the size of the transaction, foremployees in the PPI process, often at discountedprices and with special financing arrangements.Financing share ownership schemes is the majorconcern in developing countries. Some countries(for example, Bolivia and Chile) overcame this con-cern by allowing workers to use their end-of-serv-ice benefits to invest in the share program, with theguarantee that the value of the shares would notfall below their entitled benefits at the time ofretirement. As a result, in many of the enterprisesmore than 80 percent of employees participated inthe program. In addition to financial gains fromsuch arrangements, ownership programs can helpgive employees a direct stake in the performance ofthe company and thus help improve labor relationsand labor productivity.

Module 5 summarizes experiences in the use ofemployee share ownership mechanisms, and con-cludes that share transfers are probably best

viewed as a supplement to rather than a substitutefor severance payments.

MANAGING THE RESTRUCTURING PROCESSExperience shows that labor restructuring can pro-ceed smoothly if stakeholders are involved andthere are good communication mechanisms inplace. The fears and concerns of workers can besignificantly reduced when explicit efforts are madeto inform them about the objectives, timing, andmethods of PPI, as well as the packages and incen-tives that will be developed to minimize the socialimpact. To ensure a smooth process, governmentsalso need to define implementation arrangementsand create the capacity to carry out labor pro-grams.

Engaging with StakeholdersPrivatization is almost everywhere a highly contro-versial process opposed by various stakeholders,including labor. For the following reasons laboropposition can be greater for infrastructure enter-prises than for other state-owned enterprises:

• High levels of downsizing are often needed.

• High levels of investment are required, andthe involvement of foreign investors is likely.

• PPI may coincide with tariff increases andrebalancing, particularly in sectors wherethe service is seen by some stakeholders as a“right” (for example, water).

• Infrastructure services are essential, and organ-ized labor has some degree of power to disruptthose services through industrial action.

The lack of labor involvement in the process hasbeen a further contributing factor. Although thereis a trend toward more openness, some govern-ments are still reluctant to engage earlier and moreopenly with stakeholders. They recognize both theadvantages and disadvantages of engagement (box1.12) and will weigh the benefits and costs ofengagement as summarized in table 1.2 below.

Labor Toolkit: Framework and Overview

29

The implementingagency can benefitfrom experienceelsewhere whenplanningredeployment.

Share transfers arebest seen as anaddition to ratherthan a substitute forvoluntary departureor early retirement.

MO

DU

LE 1

Three of the most common concerns for govern-ments are that:

1. Stakeholder participation might delay PPI.Just as workers may be fearful of job loss,implementing agency officials, particularlyat the start of the process where there are noready answers, may fear that engaging withstockholders can lead to delays.

2. The very process of engaging stakeholderscan raise workers’ expectations, which offi-cials may not be able to meet (for example,on the scope of consultation or on levels ofseverance). If, for example, the implement-ing agency consults with trade unions onthe process of labor adjustment in PPI, thenthe implementing agency may fear thattrade unions will reject that consultationand demand that the policy of PPI itself bechallenged.

3. There is a lack of specialist skills, tools, andexperience within government to engagewith confidence on labor issues.

Those concerns have some validity in experience.But experience also shows that failure to involvestakeholders can have significant costs, can fuelconflict and suspicion, and may further delay PPIor lead to problems down the line. Instead, a well-managed process involving stakeholders can facili-tate PPI. Equally important, it can encourage adop-tion of labor and working practices suited to localcircumstances and thus improve outcomes. If prop-erly planned and managed, the involvement ofstakeholders can play an important role instrengthening the fairness, transparency, andaccountability of the PPI process.

Key Stakeholders

There are many different stakeholders, defined asgroups or institutions that may be affected by ormay influence the design, implementation, and out-comes of labor restructuring and PPI more broadly.A good stakeholder analysis can identify thesegroups and provide information about and under-standing of their interests and concerns. It is also atool to inform the design of labor approaches andthe process of engagement with workers.

Further information on stake-holder analysis is set out inmodule 6, and stakeholderanalysis worksheets are

included on the accompanying CD-ROM. Somekey points are outlined in box 1.13.

For any PPI scheme, there are five principal groupsof stakeholders:

1. Employees: Employees can be broken downinto several categories by, for example, sta-tus (temporary, permanent, and contract);age; skill; cadre (management, skilled, ormanual); gender; and ethnicity. The con-cerns of different groups and the likelyimpact of work force restructuring on themmay vary (see, for example, the case of SriLanka Telecom in box 5.7, module 5), anda better understanding of their concerns cantherefore influence and improve not only thedesign and content of communication pro-grams for workers but also the design of thelabor programs themselves.

Labor Issues in Infrastructure Reform: A Toolkit

30

Stakeholders havemore concerns overthe involvement ofthe private sector ininfrastructure thanin other state-ownedenterprises.

MO

DU

LE 1

Box 1.12: Malawi—Experience of Consultingwith Labor in Privatization

Malawi’s experience is that labor opposi-tion to privatization escalates if the labormovement is not involved early in the

process:

In many African countries labor unrestoccurred because of lack of consultations.Our experience is that an explanation of thegovernment’s plans assures labor unionsthat sacrifices will be balanced with meas-ures to allow the employees to share in thebenefits of privatization….Two problems[however] emerge with regards to consulta-tion with labor unions in general. The first isthat most employees consider privatizationas a means of forcing redundancyoffers….There is yet another dimension toinvolvement of unions in the process. Aswe discovered in Malawi it could—andoften does—lead to a significant complica-tion and lengthening of the transaction. Inaddition, employees who have access toprivileged and sensitive information tend touse it to their advantage in the biddingprocess, thereby undermining the integrityof the transaction (Sauti-Phiri 2002).

Stakeholder

analysis

worksheets.

2. Unions: Trade unions are important stake-holders and can influence the restructuringprocess, but implementing agencies may facechallenges in engaging with them. Like gov-ernments, unions may lack the capacity toengage effectively. Some unions do not wantto be seen as cooperating with employers,whereas others may oppose governmentpolicy on PPI as a whole. Consultation maybe made difficult by the number of unionsinvolved. For example, discussions on portreform in Sri Lanka involved 19 differentunions, and in Orissa’s power sector, forexample, employees were represented by 43trade unions and federations. In some cir-cumstances, national and international fed-erations are important parties, both in termsof capacity building for local unions indeveloping countries and of entering intogeneral framework agreements at nationalor even international levels.

3. Government: As is true among employeesand unions, there are different groups with-in government itself. The key ministriesinvolved are likely to be the relevant sectorministry, the ministry of finance or econom-ic planning, and the ministry or agencyresponsible for privatization. Other min-istries, including those of social protection,labor, justice, and local government, alsohave a role. Understanding the differentroles—and frequently competing interests—of these different factions within govern-ment is important in making sure that laborprograms are effectively carried out.

4. Investors: Investors, whether foreign ordomestic, often only enter into the discus-sions at the transaction stage. As a resulttheir specific concerns may not be heardduring much of the restructuring process.Earlier involvement of investors allows their

Labor Toolkit: Framework and Overview

31

The implementingagency may firstneed to convincecolleagues of theneed to engage withlabor and otherstakeholders.

An inclusive processis more likely to leadto approaches thatare tailored to localcircumstances.

Key message: thechallenges ofengagement can besuccessfullymanaged.

Stakeholder analysisinforms laborapproaches.

MO

DU

LE 1

Potential Benefits

• Helps build consensus about and “ownership”of PPI policies and projects

• Saves time in the medium term by avoidingmisunderstandings, disputes. and mistakes

• Brings in the skills, experience, and knowledgeof workers, trade unions. and other stakehold-ers, and thus helps identify and reduce techni-cal risks to the PPI scheme

• Enables more informed and inclusive decision-making

• Fosters public debate and discussion

• Encourages the adaptation of approaches tothe particular local circumstances of each PPI

• Develops awareness of other stakeholders’points of view, concerns, and aspirations

• Reinforces the legitimacy and transparency ofgovernment’s policies and tactics

• Reduces political risks by bringing all interestsinto the process and demonstrating govern-ment commitment

• Improves transparency and accountability toshow that there is no hidden agenda, favoritism,or corruption associated with PPI

Table 1.2: Benefits and Costs of Participatory Engagement Processes

• Debate and discussion take the place ofaction; they delay implementation of the PPIplan

• Wastes time by allowing the engagement tobecome an end in itself (“analysis paralysis”)

• Raises political awareness of (and oppositionto) PPI proposals before government itselfhas had time to fully consider the options

• Diverts scarce implementation capacity intomanagement of the engagement process

• Selection of those to be involved in consul-tation and dialogue reinforces the suspicionsof those excluded (that is, participation isseen as a vehicle for exclusion of somegroups)

• Conflicts of interest rather than shared inter-ests shape (that is, distort) project designand implementation

• If not fully transparent, the process can beabused or captured by vested interests

Potential Costs

interests to be taken into account and bal-anced with those of other stakeholders. Suchinvolvement can take place through consul-tation meetings with investors, better com-munication of relevant policy papers, andearly scheduling of prequalification. It mayalso be helpful to the ultimate PPI outcomesto delay the renegotiation of labor contractsuntil short-listed investors are consulted.

5. Customers: Consumers and other users ofinfrastructure services have a direct stake inthe broader PPI process. Opinion polls andattitude surveys are valuable sources for

developing an understanding of what peoplethink. In general, however, customers wantreliable services they can afford, whichmeans they have an indirect interest in thecapacity and productivity of the work forceproducing them. Conversely, workers arealso consumers and so they have a broaderstake in the process as well.

Table 1.3 summarizes the stereotypical concerns ofthese stakeholders. The stereotypes are only a start-ing point, however. Stakeholder analysis will mostlikely reveal other interests as well as differenceswithin these groups, under the unique circum-stances of each PPI transaction.

Forms of Engagement

Stakeholders can be engaged at different levels inthe labor restructuring process through four mainforms of engagement: communication, consulta-tion, negotiation, and cooperation.

Communication is mainly a one-way transfer ofinformation from government, the implementingagency, or redeployment counselors to the stake-holder audience.

Consultation and negotiation are both two-wayprocesses, but the expectations of outcomes arevery different. Participants in consultation expecttheir views to be heard and taken into account,while those in negotiations expect that mutuallybinding results will be the outcome.

Negotiation is a distinct form of engagement thatarises from the contractual employer–employeerelationship between government and work force.

Cooperation can be seen as a more mature form ofengagement where both sides expect to participateactively and are committed to win–win outcomes.This distinguishes it from negotiation, which canbe adversarial. Cooperative approaches often havelonger-term and broader perspectives than negotia-tion.

A key question is which type of engagement shouldbe used, and when. It is rarely the case that allstakeholders need to be communicated with at thesame time. Engagement activities are therefore

Labor Issues in Infrastructure Reform: A Toolkit

32

MO

DU

LE 1

1. Stakeholder analysis can be based on newresearch (such as attitudinal surveys), avail-able secondary data (such as publications,reports, or press clippings), simple struc-tured interviews with individuals, focusgroups, and small group meetings withstakeholders themselves and with informedpeople.

2. Based on the analysis, stakeholders can bemapped in terms of their influence andimportance. Influence is the power thatstakeholders have to affect the outcome ofwork force restructuring. Importance refersto the extent to which a successful outcomedepends on involving those stakeholders.

3. The analysis may first be done as a draftexercise. The draft will reveal gaps in under-standing of stakeholders’ interests anduncertainties over appropriate engagementstrategies, which then can be developedand refined through subsequent interviews,focus groups, or targeted opinion polls.

4. Interviews, small workshops, or a series offocus groups led by the senior manager inthe implementing agency or a delegated offi-cer ideally will involve representatives fromstakeholder groups.

5. Making the process as objective as possiblehelps reinforce the credibility of govern-ment’s commitment to a fair process, andcan improve the quality of the analysis.Using an independent analysis, perhapsfacilitated by a commissioned consultant,and drawing on prior interviews with stake-holders helps ensure objectivity.

Box 1.13: Key Points for Stakeholder Analysis

often sequential and, depending on the circum-stances, usually include steps to:

• Inform stakeholders about PPI.

• Communicate with employees and unionson the need for work force restructuring.

• Consult with employees, unions, andinvestors on restructuring approaches,including severance packages and proce-dures.

• Carry out negotiations among government,workers, and investors on issues such aslabor contracts, pensions, and working prac-tices. If there is an economic regulator forthe sector, it may be involved as well.

One particular risk is premature activities. Hastyand ill-prepared communications can damage thecredibility of government and delay PPI if theyexpose uncertainty in the government’s approach.The implementing agency should be able to advisegovernment officials and politicians about how and

when to sequence engagement events and aboutthe key messages to be conveyed. The basic ration-ale for why work force restructuring is essential forthe PPI plan must be clearly articulated and under-stood within government before any efforts aremade to communicate it.

Module 6 outlines how todevelop an engagement strate-gy and an engagement plan,with detailed guidance oncommunication, consultation,

negotiation, and cooperation, and the CD-ROMincludes guidelines on communicating with work-ers and guidance on the use of tools such as focusgroups and videos.

The actual process of engagement is likely to havestarts and stops, periods of progress and setback. Itmay not always be possible to follow a precise,neatly sequenced plan. As the discussion of Côted’Ivoire Railways illustrates (box 1.14), a commit-ment to engage on work force restructuring issues

Labor Toolkit: Framework and Overview

33

Government is not amonolith—theremay be pro- andanti-PPI factions.

Investors’ viewsmay not be hearduntil late in the PPItransaction.

MO

DU

LE 1

Whatcustomers

want

• Dependableservice

• Affordable tariffs

Whatinvestors

want

• Tailored servicefor the poor

• Steady, long-term returns

• Market share,reputation,geographicpresence

• Mitigation ofrisks not undertheir control,or profits com-mensuratewith risks

• Clear regulato-ry frameworksand adequatefreedom tomanage thebusiness

Whatgovernment

wants

• Budget sav-ings, reducedliabilities forthe state

• Tax, fee, orsale revenues

• Happy cus-tomers andconsumers

• Fast environ-mentalcleanup

• Jobs fordomestic firms

Whatworkers

want

• Security ofemployment,livelihood, andremuneration

• Satisfactorypay, workingconditions,and workpractices

• Training,enhancedskills

Whatunionswant

• Retained bar-gaining posi-tion and status

• Involvement inconsultation

• Maintainedmembership

• No shift to“atypical”work

• No loss ofworker rights

• Evidence offuture jobs

Table 1.3: Understanding Stakeholder Interests

Source: Adapted from Saghir and Taylor 1999.

Hints and tips

on communi-

cating with

workers.

can lead to mutually acceptable solutions andimproved outcomes for the implementing agency,the workers, and other stakeholders.

Implementation ArrangementsTwo major requirements for a successful labor pro-gram are clear implementation arrangements andthe capacity to undertake the various aspects of theprogram. For major PPI transactions this usuallymeans the creation of a labor unit within the PPIimplementing agency. Arrangements vary among

countries but common institutional homes for theimplementing agency are the national- or state-levelprivatization agency; the relevant sectoral or lineministry; the PPI enterprise itself; the central min-istry of finance, economics, or planning; or specialunits set up to support sector or individual enter-prise reforms (examples are British Coal Enterpriseand Brazil’s Port of Santos Labor Fund).

Once the agency is created, the government willneed to assemble a team for implementation. Asample organizational structure for a PPI team is

Labor Issues in Infrastructure Reform: A Toolkit

34

Consultation andnegotiation are verydifferent processes.

MO

DU

LE 1

The importance of the way in which laborissues are handled in the context ofrestructuring and privatization was demon-

strated in the case of Côte d’Ivoire Railways. Inthe early 1990s the railway was restructuredwithout union consultation. The union was unpre-pared for the challenge of restructuring andlacked the know-how and experience to effec-tively respond to it. The work force was reducedby about a third, but it soon became clear thatthere had been too many redundancies in somekey areas—notably in signaling and security—and this caused operational difficulties. As aresult of this and other deficiencies, the companyfound itself having to pay some of its remainingwork force overtime, so that much of the poten-tial of the process to put the railway on a moresound financial footing was undermined.

By contrast, a second phase of restructuringfrom 1993 onward, with which the union wasmore effectively engaged, produced betterresults. The company’s 1993 activity report high-lighted changes in commercial attitudes, reduc-tion of fraud by 30 percent, and a 60 percentincrease in availability of locomotives—a goodsign of maintenance efficiency. The secondphase also created a labor relations climate moreconducive to the effective implementation of theprivatization process, which began in earnest in1995.

Subsequently, however, a breakdown in commu-nications and consultation among government,management, and labor again produced avoid-able problems. When the union asked for infor-mation about how further labor restructuringwould be carried out, the request was ignored.This led to a strike and to direct actions, such as

blocking the center of Abidjan with a locomotive.Details of the redundancies envisaged were pro-vided eventually, but the union had managed tonegotiate severance terms equal to 14 months ofwages, double the initial offer. In addition, thenumber of years of contributions required to enti-tle an employee to an early retirement packagewas reduced from 20 to 15.

During the negotiations the union also proposeda plan to enable workers to establish their ownbusinesses after retrenchment. Sitarail, the con-cession company that took over operationalresponsibility for the railways, agreed in principleto favor companies created by former workerswhen looking for subcontractors. Since then,track maintenance, company car fleet manage-ment, and printing of timetables and tickets havebeen contracted out to firms set up by formerworkers. Sitarail has also agreed to give prefer-ence to workers made redundant in 1995 whenrecruiting new staff, albeit on terms different fromthose that obtained before privatization.

These measures have helped mitigate theadverse effects of retrenchments. At the sametime, workers who remained with Sitarail havegained from privatization. For example, althoughon paper the terms of the company’s healthinsurance program is worse than before privati-zation, in practice employees have more confi-dence in it because the company now meets itsreduced obligations, whereas before the largercommitments were not honored. In addition, therail workers’ union reports that working condi-tions have improved since privatization becauseof investment in workshops, equipment, and uni-forms.

Source: Martin and Micoud 1997.

Box 1.14: Côte d’Ivoire Railways—Participatory Processes

management contractors have been engaged to imple-ment enterprise or work force restructuring as part ofthe overall preparation process for PPI. Such arrange-ments enable the implementing agency to bring inadditional capacity to deal with the short- and medi-um-term workload associated with the restructuring.And in yet other cases, special units or funds havebeen set up at the enterprise level to implement thelabor program. Some are wholly financed by the cen-tral government budget, others by levies or with fund-ing from local governments. One example is the Portof Santos Labor Fund (box 1.15).

The labor unit (or individual expert) is usuallyresponsible for overall implementation, and the rel-evant tasks cover the full range of activities in thelabor program:

• Completing the establishment of the laborunit, including the training and capacitybuilding of any in-house staff.

Labor Toolkit: Framework and Overview

35

MO

DU

LE 1

shown in figure 1.1. The implementing agency willneed to include labor experts as part of the teamfor the labor restructuring program. The structureof the labor team will vary depending on the sizeand scope of the labor program.

In large PPI programs with a major labor compo-nent, and where there is little capacity elsewhere inthe government to deal with labor, a small laborunit may have to be created within the PPI agency.In smaller programs, and where other groups inthe government such as the ministry of labor havethe required capability, it may be necessary torecruit an individual labor expert in the agency tocoordinate implementation.

If there is an enterprise with a large labor restructur-ing program (as in many railways enterprises), laborexperts may be recruited at the enterprise level (ratherthan at the implementing agency) to help carry outthe restructuring process. In other cases, private sector

PPI Supervising Committee(senior civil servants, political

representation)

Implementing agency(manager)

Transactionmanager(s)

Laborspecialist(s)

Legalexpert(s)

Communicationsmanager(s)

Officesupport

Severance(finance) officer

Redundancycounselors(in-house orcontracted)

Council of Ministers/Cabinet

Redeployment /training manager

Contractedtraining

providers

Contractedauditors/

accountants

Figure 1.1: Organization Chart for a PPI Team (Sample)

• Securing funding (if this has not alreadybeen done).

• Commissioning early initial reviews (forexample, legal reviews and assessments ofthe need for labor restructuring).

• Designing and enacting procedures for theimplementation of voluntary redundancy,early retirement, and other actions.

• Ensuring that stakeholder engagement(consultation, communication, negotiation,and cooperation) takes place sufficientlyearly for all key stakeholders so that theprocess can be seen as both fair and legiti-mate.

• Coordinating with other existing govern-ment organizations that may already beinvolved in redeployment support and that

Labor Issues in Infrastructure Reform: A Toolkit

36

MO

DU

LE 1

Box 1.15: Port of Santos, Brazil—TheSpecial Labor Fund

Aspecial port workers’ fund is being set upin Santos that should resolve years ofbitter confrontations between stevedores

and port operators at Brazil’s leading port. Thereal 80 million (US$47.73 million) fund will beused to soften the impact of cutting the laborpool in Santos to about 4,500 dock workersfrom a current total of 11,500 employees.Money from the fund will be used to retrain portlaborers employed by the administrator of thecasual labor pool for alternative work within newhigh-tech and light industries that will beencouraged to locate to Santos. The project isalso backed by the São Paulo State Federationof Industry, the Santos Port Council, localimporters/exporters, state and municipal gov-ernments, and national governmental bodiesdealing with dock labor. Rules for dismissalsand claims are to be worked out. The local andcentral governments are expected to help bringhigh-tech and small businesses to Santos withinthree years or so. The technical plans are beingpresented to the unions for discussions, and fol-lowing their agreement the plans will be pre-sented to the national body coordinating themodernization and privatization of Brazil’s portsand to the government in Brasilia.

Source: World Bank 2000 (Port Reform Tool Kit, Module 7,p. 16).

must be brought into the process. Theseinclude ministries of labor, trade, and indus-try and commerce; public employmentoffices; and finance authorities.

• Designing procedures for the approval, dis-bursement, monitoring, and auditing of sev-erance funds.

• Designing and coordinating implementationof the redeployment (counseling and retrain-ing) program for workers.

• Monitoring specialized contractors hired forthe labor restructuring process.

• Commissioning monitoring and evaluationactivities.

In order to build capacityand support for developinglabor programs, the imple-menting agency can seektechnical and financial sup-port for organizing studytours and workshops in theearly stages of the reform

program. As with other aspects of PPI, internationaldonors are one source of such support. One inter-national facility is the Public-Private InfrastructureAdvisory Facility (www.ppiaf.org), a multidonorfacility that offers rapid disbursing of grants for spe-cific inputs by PPI, sector, and labor specialists.

Analytical and technical capacity is needed to carryout initial staffing and feasibility assessments, todesign and implement the various aspects of thelabor restructuring program, to analyze the finan-cial and economic viability of the program, and tomonitor and evaluate the net impact of the laborprogram.

When the labor restructuring program is complet-ed, the labor teams or labor funds will be wounddown, and this needs to be taken into account dur-ing project design. The case of British CoalEnterprise is relevant at this point because it showshow a significant redeployment support service canbe closed down and privatized (see box 1.16).

PPIAF provides a

quick-mobilizing

grant facility for

small inputs by

PPI and sector

specialists (see

www.ppiaf.org).

MONITORING AND EVALUATINGLABOR PROGRAMSDuring and after the labor restructuring program itis important to carry out a cost-benefit analysis ofthe various elements of the labor program so it ispossible to decide which approach is the most suit-able and beneficial to the situation at hand and tomake midcourse corrections as needed in thedesign and implementation of the program.Although critical to success, monitoring and evalu-ation activities are often a neglected aspect of thelabor restructuring process. Capacity in this areaneeds to be built and systems put in place to evalu-ate the individual elements and the program as awhole.

ObjectivesGovernments often invest too little in monitoringand evaluation even though there are severalpotential benefits:

• Better understanding of the financial andeconomic case for labor programs

• Reduced costs to government (mainlythrough better design)

• Greater benefits for workers, through betterprogram design and through fine-tuning ofthe program

• The capture of data and experiences so thatsubsequent phases of the labor program canlearn from past experiences.

For these reasons, monitoring and evaluation activ-ities are important and explicit efforts are requiredto include the activities in the early stages of pro-gram design and after implementation to evaluatethe net impact of the program.

Monitoring and evaluation tools can be incorpo-rated in all four phases of the PPI process, as table1.4 illustrates. Among the tools are:

• Undertaking the initial feasibility assessmentsduring phase 1: Examples of such assess-ments are an analysis of labor redundancy inVietnam in preparation for future downsiz-

Labor Toolkit: Framework and Overview

37

Governmentscommonly fail tomonitor laborprograms effectively.

MO

DU

LE 1

Box 1.16: British Coal Enterprise—Privatizing Redeployment

One of the potential concerns about set-ting up large sector- or enterprise-specif-ic redeployment units is that their servic-

es are only needed for a short time. How will theclosure of those redeployment units be han-dled? One answer is to privatize them, too. Thiswas the case with British Coal Enterprise (BCE).The fact that it could be privatized indicates thecommercial worth of the assets and skills thatBCE developed in support of former miningcommunities. Each of BCE’s main areas ofactivity was privatized separately.

• Outsourcing: The expertise developed infour years of running what was initiallyknown as British Coal’s Job and CareerChange Scheme led to the creation of thecommercial outplacement division of BCE,known as Grosvenor Career Services.Grosvenor specialized in the delivery ofredeployment training and career adviceand was successful enough that Grosvenorwas privatized through a management buy-out in 1996. The Grosvenor team providedoutplacement services to British Coal aswell as other large private and public sectorcompanies undergoing restructuring in theUnited Kingdom and Europe. Grosvenoralso provided outplacement and laborrestructuring technical assistance in coalfield regions in countries of Eastern Europeunder the European Union’s Phare program.Grosvenor was subsequently acquired byCapita IRG plc.

• Managed workspaces: These units werethe principal physical asset at the time ofBCE privatization. BCE’s experiences in pro-viding serviced units in mining communitiesenabled the privatization of this division ofBCE through a trade sale to Birkby plc in1995. Birkby later merged with MentmoreAbbey in 1999 so that BCE’s former work-space and business incubator division isnow a part of a leading European officespace management company.

• Funding operations: The business-fundingdivision of BCE was sold to a managementbuyout in 1995, and the new company,Coalfield Investments Ltd., continues toinvest in small and medium-size businessesin the coal-mining areas.

Source: British Coal Enterprise, Tawney and Levitsky 1997.

Labor Issues in Infrastructure Reform: A Toolkit

38

MO

DU

LE 1

Activity

Orientation

Focus ofwork

Timing

Typicalissues

Typical tools

Table 1.4: Analysis, Monitoring, and Evaluation in Labor Adjustment Programs

Source: Adapted from Owen and Rogers 1999.

Initialassessment

(Phase 1)

Relevance andfeasibility

Program contextand goals

Before implemen-tation

• What is thenature andextent of anyoverstaffingproblem?

• What doinvestors want?

• Do we need agovernment-funded laborprogram at all?

• Roughly howmuch will itcost to restruc-ture the workforce?

• Financial analy-sis

• Staffing auditsand bench-marking (seemodule 3)

• Stakeholderanalysis andconsultationwith investorsand labor (seemodule 6)

Designclarification

(Phase 2)

Clarification andimprovement

All elements; spe-cific objectives

Before and duringimplementation

• Which sever-ance or earlyretirementpackages arethe most cost-effective?

• Which rede-ployment serv-ices should beoffered?

• Financial analy-sis

• Economicanalysis

• Stakeholderanalysis

Performance monitoring(Phase 3)

Justification andfine-tuning

Delivery and out-puts

During implemen-tation

• Are workersapplying forvoluntarydeparture in thenumbersexpected?

• Are severancepayments andredeploymentservices beingdelivered ontime?

• Financial analy-sis

• Monitoring ofstaffing num-bers

• Monitoring ofparticipationand dropoutsfrom counsel-ing, trainingcourses, and soforth

Impactmonitoring and

evaluation(Phase 4)

Justification,accountability, andreview

Outcomes

After implementa-tion

• What was theimpact on thePPI transaction(delays, price)?

• What were thefiscal costs andbenefits forgovernment?

• What havebeen the wel-fare impacts onworkers?

• Financial analy-sis

• Economicanalysis

• Socioeconomictracer studies

ing (Belser and Rama 2001), an assessmentof alternatives for a new restructuring pro-gram in Sri Lanka (Fiszbein 1992), and anassessment of the likelyimpact of a national civilservice downsizing pro-gram in Ghana (Gregory1994).

• Clarifying the design of the labor restructur-ing program during phase 2: For example,an assessment of the financial and economicviability of the Algerianretrenchment program(Ruppert 1999).

Belser and

Rama 2001

Ruppert

1999

Labor Toolkit: Framework and Overview

39

Most evaluations ofwork forcerestructuringprograms revealvery fast paybackperiods.

MO

DU

LE 1

• Monitoring the performance of the programduring implementation in phase 3: Forexample, monitoring theprogress of work forcerestructuring in Brazil’sfederal railway (Estache,Schmitt de Azevedo, andSydenstricker 2000).

• Monitoring outcomes and evaluating the netimpacts of labor restructuring during phase 4:Examples include an evaluation of activelabor market programs in Hungary (O’Leary1997) and a tracer study of the consequencesof retrenchment for civil servants in Ghana(Alderman, Canagarajah, and Younger 1994).

Concepts and ApproachMonitoring and evaluation early in the design ofthe program are critical in helping governmentscompare alternative options for severance and rede-ployment and in determining the cost and financingneeds. Evaluation has to take into account thefinancial and economic returns of both the variouscomponents of the labor program and the programas a whole and the outcomes of the program itselfin terms of effects on workers and on the enterprise.Periodic monitoring is important in keeping trackof the program and learning from experience.

Financial vs. Economic Returns

A cost-benefit analysis has two main dimensions:financial returns and economic returns. From apublic finance perspective, the financial returns of alabor program will be central to decisionmakers.This is particularly the case where the enterprisetakes a loss and completing the labor restructuringwill bring an improvement in public finances byreducing government transfers and subsidies andincreasing tax payments from PPI firms. Providedthat no rehiring takes place, labor programs can becost effective with a payback period ranging fromtwo to six years. Labor programs thus appear tooffer good rates of financial return that few publicinvestment projects would be able to match.

However, economic returns also need to be calcu-lated and taken into account for several reasons:

• First, economic analysis assesses the impacton aggregate output or welfare. Financialreturns do not indicate whether displacedworkers are, in aggregate, more or less pro-ductive following the labor program. It isquite possible that a proposed labor pro-gram can be attractive from a financial per-spective but can fail when subjected to eco-nomic analysis.

• Second, such analysis provides an answer toopponents of labor restructuring in PPI whomay argue that government is making a baddecision by ignoring the wider economiccosts and benefits (see, for example, theSouth African case in box 6.8, module 6).

• Third, the analysis may be a requirement ofinternational funding agencies, whose lend-ing or funding procedures need economic aswell as financial analyses.

The economic rate of return from a labor programcan be defined as the net increase in marginal pro-ductivity of surplus staff redeployed to anotherproductive activity elsewhere in the economy, andthe marginal gain from avoided labor-related costs;available evidence shows that it can be high.Returns are particularly high if there are otheropportunities for employment and if redundantworkers are likely to find jobs in the labor market.

Module 7 examines in greater detail the approach,methodology, and findings on both financial andeconomic returns in labor programs. It also pro-vides the necessary tools to carry out cost-benefitanalyses of labor programs.

Evaluating Effects on Workers and Firms

Evaluating the impact of the labor restructuringprogram on workers is another aspect of evalua-tion. Such evaluations typically cover (a) theimpact on employment because job losses areimportant “headline” figures in the media and else-where and understanding what happens to jobs isimportant economically and politically; (b) theimpact on wages and benefits; (c) the broaderimpact on workers’ welfare in terms of theirsocioeconomic profile; and (d) the follow-up situa-tion of workers, including duration of unemploy-

Estache,

Schmitt de

Azevedo, and

Sydenstricker

2000

ment and types of jobs found. Item (d) is particu-larly relevant in the evaluation of redeploymentand training schemes, where common objectivesare to reduce the period of unemployment and toprevent displaced employees from entering a poolof long-term unemployed workers. An equallyimportant question for evaluation is the impact oflabor restructuring on enterprise efficiency, includ-ing improvements in labor productivity, and onfinancial and operational performance.

These effects are often assessed through ex postworker and firm surveys. Such surveys are beingcarried out in a number of countries (for example,Brazil, India, and Vietnam). Module 7 highlightsthe main types of data that are usually collected inthe surveys and the methodology for carrying outsuch assessments.

A central requirement of any evaluation is that itseparate the effects that would have happened any-way from those that resulted from the specificintervention. Before and after comparisons aloneare not sufficient. If earnings rise after training, forexample, that may not result from the training butfrom changes in the macroeconomy, local changesin labor demand, or such worker-specific attributesas lifecycle earnings changes.

Evaluation therefore requires a counterfactualalternative, which is normally provided by a con-trol or comparison group of workers who did notparticipate in the severance or redeployment pro-gram. Counterfactual analysis can use either of thefollowing groups:

• Control groups, which consist of participantsselected at random from within a well-defined population from which the membersof the treatment group are also selected

• Comparison groups, which consist of partic-ipants that are purposely matched to theparticipants of the treatment group.

Selecting appropriate evaluation techniques andcounterfactuals is particularly important whenexamining the impact of redeployment programs.Policymakers need to know whether the resourcesthey are spending are being efficiently used and

whether programs are having the intended effect.Such analysis also helps make informed decisionson whether to expand, better target, or scale downprograms. A range of evaluation techniques isavailable and module 7 discusses these in detail.

Periodic Monitoring

Monitoring differs from evaluation in that it isprincipally a management function aimed primarilyat keeping track of implementation and makingperiodic assessments of the performance of the pro-gram (see box 1.17).

In potentially large-scale labor restructuring pro-grams involving great numbers of workers and sig-nificant financial resources, it is important to moni-tor the implementation of the program to ensurethat funds are being properly used to help workersand that all workers are being helped. At a mini-mum this requires the development of a systemthat tracks the numbers of workers leaving theenterprise and expenditures on severance and rede-ployment at the enterprise level and for the PPIprogram as a whole.

Periodic assessments or monitoring the perform-ance of the various aspects of the labor program—for example, severance payments and redeploy-ment support—also help policymakers and practi-tioners learn from experience and undertake mid-course corrections as needed. The potential benefitsof such assessments are large: one study inTanzania found that the information from moni-toring studies could have saved government up toUS$7 million during the course of retrenchment ofaround 5,000 state enterprise workers.

Monitoring of redeployment programs can be par-ticularly valuable in that it can help identify weak-

Labor Issues in Infrastructure Reform: A Toolkit

40

Economic analysisprovides aperspective thatfinancial analysisalone cannotprovide.

MO

DU

LE 1

Box 1.17: Monitoring vs. EvaluationMonitoring is the continuous assessment ofprogram implementation in relation to agreedschedules and the use of program outputs bybeneficiaries.

Evaluation is the periodic assessment of therelevance, performance, efficiency, and impactof the program in relation to stated goals.

nesses in the implementation of the program (forexample, late disbursement of funds, lack of infor-mation on redeployment services) and readjust pro-grams to deliver more appropriate training, coun-seling, and redeployment services to workers.

INTEGRATING LABOR PROGRAMS IN THE PPIPROCESS: A ROAD MAPThis section presents an illustrative road map forintegrating the labor program in the broader PPIprocess. The road map will help the implementingagency put the various tasks and activities of alabor adjustment program within the context andphasing of the overall PPI transaction. The firststeps for the implementing agency, however, are toestablish a labor unit and to secure funding for thelabor program.

There are four main phases involved in developinga labor program for PPI:

• Phase 1: Initial assessment or diagnosticphase

• Phase 2: Design of the labor program

• Phase 3: Implementation of the program

• Phase 4: Monitoring and evaluation activities.

Figure 1.2 presents a road map or overview of thelabor adjustment process. It illustrates the integra-tion of labor program planning with the wider PPIprocess. For example, the initial assessment of thelabor issue should follow or take place parallelwith a wider assessment of the health of the enter-prise and in the context of the overall economic,social, and development objectives sought fromenterprise restructuring or PPI.

The main boxes in the road map link to key mod-ules in this Toolkit, and the oval shapes representspecific tasks or tools in the Toolkit.

Editable copy of the road map in

Microsoft Power Point format

Phase 1: Initial AssessmentThe first phase—initial assessment—is criticalbecause it identifies the nature and scope of thelabor program and provides the planning frame-work for the program as a whole. It includes:

• Clarifying the overall goals and objectives oflabor adjustment (see the third section ofthis module)

• Assessing staff levels, skills, and the extentof overstaffing (module 3)

• Reviewing the external stakeholder environ-ment (module 6), the legal circumstances,and likely costs including those related toseverance and to pensions (module 5)

• Beginning the implementation of “soft”options.

In most cases, phase 1 will be an initial assessmentof labor issues in PPI; the analysis, however, willcontinue through phase 2 (design) where detailedstrategies and options are refined. In phase 1 therewill not only be assessments regarding labor issues,but also assessments of the operational, regulatory,legal, and financial aspects of the PPI transactionmore broadly.

The various phase 1 assessments can provide thebasic information needed to prepare the initial sub-missions for decisionmaking by senior ministers orthe cabinet. Box 1.18 sets out a generic checklistfor such submissions.

Phase 2: Design of the LaborProgramPhase 2 builds on the assessments of phase 1 andallows the implementing agency to make specificproposals to decisionmakers. More specifically,phase 2 involves:

• Developing an overall strategy and restructur-ing options (module 4) for dealing with laborissues as part of the overall PPI strategy

• Developing concrete pension arrangements,severance packages, and retraining/redeploy-ment programs, as well as financingarrangements (module 5)

Labor Toolkit: Framework and Overview

41

MO

DU

LE 1

Labor Issues in Infrastructure Reform: A Toolkit

42

MO

DU

LE 1

Iden

tify

wor

k fo

rce

chan

geop

tions

(Mod

ule

4)

Iden

tify

wor

k fo

rce

chan

gest

rate

gies

(Mod

ule

3)

Avo

idin

gad

vers

ese

lect

ion

Cap

acity

bui

ldin

g:

Fund

ing,

res

ourc

es, s

taff

, con

sulta

nts,

lega

l cha

nges

(mod

ule

1)

Fina

ncia

l and

oper

atio

nal

rest

ruct

urin

g

Par

ticip

atio

n an

d c

omm

unic

atio

n w

ith a

ll st

akeh

old

ers

(mod

ule

6)(k

eep

ing

peo

ple

on

boa

rd t

hrou

gh im

ple

men

tatio

n)

PP

I bid

din

gan

d t

rans

actio

np

roce

ss

Pos

t-P

PI

mon

itorin

g an

dre

gula

tion

Cou

nsel

ing

Imp

lem

ent

red

eplo

ymen

t/so

cial

miti

gatio

n p

rogr

am (m

odul

e 5)

Imp

lem

ent

seve

ranc

e (m

odul

e 5)

Pre

- or

p

ost-

PP

I(W

hen?

)

Red

eplo

ymen

tm

onito

ring

syst

ems

Mon

itor

effe

ctiv

enes

s of

red

eplo

ymen

t/so

cial

miti

gatio

n(m

odul

es 5

, 7)

Ret

rain

ing

Con

trac

ting

out

Po

licy

dec

isio

nfo

r P

PI

Ap

pro

val

of

PP

Ist

rate

gy

Lab

orm

arke

ts

Seq

uenc

e a

ndtim

ing

Sha

resc

hem

es

Trac

erst

udie

s

Res

truc

turin

g

Sof

t op

tions R

edun

dan

cy

Ear

lyre

tirem

ent

PP

Ip

lan

app

rove

d

PP

Itr

ansa

ctio

nco

mp

lete

d

Ong

oing

lab

or-r

elat

ed t

asks

of

var

iab

le d

urat

ion

(Too

lkit

mod

ules

)D

Iscr

ete,

tim

e-b

ound

,la

bor

-rel

ated

tas

ks (T

oolk

it m

odul

es)

Maj

or

po

licy

dec

isio

np

oin

ts

Dec

idin

gke

y el

emen

tsof

the

P

PI

tran

sact

ion

pla

n

Dec

idin

ga

PP

Ist

rate

gy

Key

:

Pha

se 1

: In

itia

l ass

essm

ent

Pha

se 2

: D

esig

nP

hase

3: I

mp

lem

enta

tio

nP

hase

4:

Mo

nito

ring

and

Eva

luat

ion

Ass

essi

ng t

he e

nter

pris

e or

PP

I sch

eme

Mec

hani

sms

(How

?)

Res

pon

sib

ility

(Who

?)

Com

mun

icat

ion,

con

sulta

tion,

neg

otia

tion

with

sta

keho

lder

s (m

odul

e 6)

Ass

ess

leve

l of

surp

lus

lab

or (m

odul

e 3)

Ben

chm

arks

Sta

ff a

udits

Ski

ll au

dits

Wor

k fo

rce

anal

ysis

Ass

ess

exte

rnal

envi

ronm

ent

(mod

ule

6)

Urg

ency

of P

PI

Sta

keho

lder

anal

ysis

Pol

itica

lvi

abili

tyLe

vel o

f sup

por

t

Ass

ess

lega

l iss

ues

(mod

ules

1, 3

, 4)

Lab

or la

ws

Lab

or c

ontr

acts

Fina

ncia

l and

eco

nom

ican

alys

is (m

odul

e 7)

Fina

ncia

l and

eco

nom

icco

st-b

enef

its

Fund

ing

sour

ces

Saf

ety

nets

Job

sea

rch

assi

stan

ce

Pre

par

eim

ple

men

tatio

np

lan

(Mod

ule

1)

Mon

itorin

g of

wor

k fo

rce

chan

ges

(mod

ule

7)

Fina

ncia

l au

dits

Mon

itorin

g of

sta

ffnu

mb

ers

Dis

bur

sem

ent

cont

rol

Ele

men

ts o

f the

To

olki

t(s

ectio

ns, t

ools

)

Pro

ject

man

agem

ent

of im

ple

men

tatio

n (m

odul

e 1)

Key

ste

ps

in a

PP

I tra

nsac

tion

Imp

lem

ent

chan

ges

in

emp

loym

ent

stat

us (m

odul

es 4

, 5)

Rev

iew

pen

sion

s (m

odul

e 5)

Rev

iew

ove

rall

goal

s of

wor

kfor

ce r

estr

uctu

ring

(mod

ule

1)

Org

aniz

atio

n r

evie

w

Def

ine

obje

ctiv

es(m

odul

e 6)

Sev

eran

ce c

osts

Figu

re 1

.2: R

oad

Map

for

Labo

r Ad

just

men

t

Labor Toolkit: Framework and Overview

43

Planning of a laborprogram needs to beintegrated with thewider enterpriserestructuringprogram. M

OD

ULE

1

• Obtaining a legal mandate to deal withlabor adjustment, including approval ofpolicies to undertake work force restructur-ing supported by relevant governmentorders or regulations

• Defining implementation arrangements anddeveloping implementation capacity (see theseventh module)

• Engaging with stakeholders through com-munication, consultation, negotiation, andcooperation (see module 6)

• Developing broader public information pro-grams to ensure that the government’s earlymessages are communicated (module 6).

The above tasks typically result in the developmentof an overall plan for dealing with labor restructur-ing that is similar to the road map in figure 1.2. Amanager in the implementing agency might edit theroad map in this Toolkit (which is a MicrosoftPower Point diagram) and adapt it to his or herlocal circumstances, or use the Microsoft ProjectPlan template (see the CD-ROM).

Outline project plan for a labor program in

Microsoft Project format.

Phase 3: ImplementationPhase 3 involves implementing the actual laboradjustment program, in particular:

• Implementing restructuring options (module4) and completing any soft options startedduring phase 1

• Implementing severance and pension pack-ages (module 5)

• Implementing redeployment programs(module 5)

• Engaging with stakeholders (module 6).

Phase 4: Monitoring and EvaluationAlthough important, the final phase of any laboradjustment program—monitoring and evalua-tion—is often the most neglected aspect. When thedifficult and painful task of adjustment is over,many government officials and implementingagency managers wish to move on and overlookthis phase.

There are good reasons, however, to take monitor-ing and evaluation seriously:

Box 1.18: Guidelines for Submissions toDecisionmakers

What are the minimum requirements fora paper submitted to the decisionmak-ers for decision? Although there may

be detailed government procedures set out, inprinciple any proposal for a labor adjustment orwork force restructuring program sent to thedecisionmakers (be it a committee of seniorministers, the cabinet, or the council of minis-ters) should fulfill a few simple, straightforwardtests:

1. There should be a one-page executive sum-mary for busy ministers.

2. There should an adequate explanation ofthe objectives of the proposal (so that minis-ters do not have to try to guess the purposeof the proposal from the details).

3. There should be a range of options present-ed (typically three or four), including a basecase “do-nothing” option where appropri-ate.

4. The authors should have identified the costand revenue effects of the options on thegovernment budget.

5. Subject to confidentiality and sensitivityconcerns, the authors should have dis-cussed the proposal with other ministriesand agencies that will be most affected by,or most instrumental in, successful imple-mentation of the program.

6. There should be some comment on theplans for implementation to demonstratethat these are practical and workable.Implementation risks should be identified.

7. There should be some attempt to identifyand comment on the likely effects of eachoption. If these can be quantified, so muchthe better, but at least there will be quantita-tive assessment.

• It demonstrates that the implementingagency is accountable. Given the very highcosts of many labor programs, most imple-menting agencies will want to be able toshow that they have managed the programeffectively.

• It reduces risk. Effective monitoringallows problems to be recognized earlyand action taken to deal with those prob-lems.

• It allows lesson-learning. Many imple-menting agencies recognize that theyknow very little about what really hashappened to workers who were displaced.This puts the manager of the implement-ing agency at a disadvantage, facing ques-tions from his or her boss, the press, ortrade unions. It also allows subsequentprograms to be implemented much moreeffectively.

The details of monitoring and evaluation systemsare covered in detail in module 7.

Additional Material (on the CD-ROM)

Background and OverviewArticles

Betcherman, Gordon. 2002. “An Overview of LaborMarkets World-Wide: Key Trends and MajorPolicy Issues.” Social Protection Discussion Paper0205. World Bank, Washington, D.C.

Fretwell, David. 2002. “Mitigating the Social Impact ofPrivatization and Enterprise Restructuring.” WorkingPaper. World Bank, Human Development SectorUnit, Europe and Central Asia, Washington, D.C.

Haltiwanger, John, and Manisha Singh. 1999.“Cross-Country Evidence on Public SectorRetrenchment.” The World Bank EconomicReview 13(1):67–88.

Kikeri, Sunita. 1998. “Privatization and Labor: WhatHappens to Workers When Governments Divest?”World Bank Technical Paper 396. Washington,D.C.

Rama, Martin. 1999. “Public Sector Downsizing: AnIntroduction.” The World Bank Economic Review13(1):1–22.

World Bank. 2002. Public Communications Programsfor Privatization: A Tool Kit for Task TeamLeaders and Clients. Washington, D.C.

PPIAF Case Studies of Labor Issuesin PPI (found on the PPIAF Web site, www.ppiaf.org). Cruz, Wilfred. 2001. “Addressing Labor Concerns

during Privatization: Lessons from theMetropolitan Waterworks and Sewerage System(MWSS), Manila, Philippines.”

López-Calva, Luis. 2001. “Private Participation inInfrastructure and Labor Issues: The Privatizationof Mexican Railroads.”

Ray, Pranabesh. 2001. “HR Issues in PrivateParticipation in Infrastructure: A Case Study ofOrissa Power Reforms.”

Valdez, Jose. 2002. “Case Studies on HumanResource Issues in Private Participation inInfrastructure in Bolivia.”

Web SitesPPIAF: www.ppiaf.org. (Site provides access to other

PPIAF tool kits, as well as information aboutPPIAF and about how governments and otheragencies can access PPIAF resources to accelerateinfrastructure development.)

Rapid Response Unit: http://rru.worldbank.org. (Thissite is a gateway to a range of information on infra-structure, privatization, and private sector develop-ment policies.)

World Bank “Shrinking Smartly”:www.worldbank.org/research/projects/downsize/.(Site is a clearinghouse for researchers, developmentpractitioners, and government officials concernedabout the difficulties encountered in downsizing alarge public sector.)

REFERENCESAlderman, Harold, Sudharshan Canagarajah, and

Stephen D. Younger. 1994. “Consequences ofPermanent Layoff from the Civil Service: Resultsfrom a Survey of Retrenched Workers in Ghana.”

Labor Issues in Infrastructure Reform: A Toolkit

44

MO

DU

LE 1

In David L. Lindauer and Barbara Nunberg, eds.,Rehabilitating Government. Washington D.C.:World Bank.

Assaad, Ragui. 1997. “The Effects of Public SectorHiring and Compensation Policies on the EgyptianLabor Market” The World Bank Economic Review11(1):85–118.

Bales, Sarah, and Martin Rama. 2002. “Are PublicSector Workers Underpaid?” Working Paper 2747.World Bank, Washington, D.C.

Belser, Patrick, and Martin Rama. 2001. “StateOwnership and Labor Redundancy: EstimatesBased on Enterprise-Level Data from Vietnam.”Policy Research Paper 2599. World Bank,Washington, D.C.

Betcherman, Gordon. 2002. “An Overview of LaborMarkets World-Wide: Key Trends and MajorPolicy Issues.” Social Protection Discussion Paper0205. World Bank, Washington, D.C.

Bhorat, Haroon, and Mei-Chi Liou. 2002. “LabourConsequences of SOE Restructuring in SouthAfrica.” Paper presented at the Second AnnualConference on Labour Markets and Poverty inSouth Africa, October 22–24. Development PolicyResearch Unit, University of Cape Town/FriedrichEbert Stiftung, Johannesburg.

Chong, Alberto, and Martin Rama. 2000. “DoSeparation Packages Need to Be That Generous?Simulations for Government Employees in Guinea-Bissau.” Paper prepared for the Sub-Saharan AfricaRegion of the World Bank (AFTM5) supported bythe Research Project on Efficient Public SectorDownsizing (RF–P036874). Washington, D.C.

Cruz, Wilfrid. 2001. “Addressing Labor Concernsduring Privatization: Lessons from theMetropolitan Waterworks and Sewerage System(MWSS), Manila, Philippines.” PPIAF Case Study.World Bank, Washington, D.C.

Estache, Antonio, and José Carbajo. 1996.“Competing Private Ports—Lessons fromArgentina. Public Policy for the Private Sector.”Note 100. World Bank, Washington, D.C.

Estache, Antonio, Jose Antonio Schmitt de Azevedo,and Evelyn Sydenstricker. 2000. “LaborRedundancy, Retraining, and Outplacement duringPrivatization: The Experience of Brazil’s FederalRailway.” Policy Research Working PaperWPS2460. World Bank, Washington, D.C.

Fiszbein, Ariel. 1992. “Labor Retrenchment andRedundancy Compensation in State OwnedEnterprises: The Case of Sri Lanka.” InternalDiscussion Paper 121. World Bank, South AsiaRegion, Washington, D.C.

Freije, Samuel. 2001. “Informal employment in LatinAmerica and the Caribbean: Causes, Consequencesand Policy Recommendations.” Paper presented atRegional Consultation on Labor Issues, PanamaCity, November 26–27. República Bolivariana deVenezuela: Instituto de Estudios Superiores deAdministracion (IESA).

Fretwell, David H. 2002. “Mitigating the SocialImpact of Privatization and EnterpriseRestructuring.” Working Paper. World Bank,Human Development Sector Unit, Europe andCentral Asia. Washington, D.C.

Gregory, Peter. 1994. “Dealing with Redundancies inGovernment Employment in Ghana.” In David L.Lindauer and Barbara Nunberg, eds.,Rehabilitating Government. Washington D.C.:World Bank.

Guislain, Pierre. 1997. The Privatization Challenge: AStrategic, Legal, and Institutional Analysis ofInternational Experience. Washington, D.C.: WorldBank.

Haltiwanger, John, and Manisha Singh. 1999.“Cross-Country Evidence on Public SectorRetrenchment.” The World Bank EconomicReview 13(1):67–88.

Hess, Jolanta. 1997. “Labor Issues Associated withState Enterprise Restructuring—Handbook.”Unpublished draft. World Bank, Washington, D.C.

India, Disinvestment Commission. 1998. EighthReport. New Delhi: Government of India.

Kikeri, Sunita. 1998. “Privatization and Labor: WhatHappens to Workers When Governments Divest?”Technical Paper 396. World Bank, Washington,D.C.

López-Calva, Luis F. 2001. “Private Participation inInfrastructure and Labor Issues: The Privatizationof Mexican Railroads.” PPIAF Case Study. WorldBank, Washington, D.C.

López-de-Silanes, Florencio. 1997. “Determinants ofPrivatization Prices.” Quarterly Journal ofEconomics 112(4):966–1025.

Martin, Brendan, and Marc Micould. 1997.“Structural Adjustment and Railways

Labor Toolkit: Framework and Overview

45

MO

DU

LE 1

Privatization—World Bank Policy and GovernmentPractice in Ivory Coast and Ghana.” A report joint-ly commissioned by the World Bank and theInternational Transport Workers Federation (ITF).Public World, London.

O’Leary, Christopher J. 1997. “A Net ImpactAnalysis of Active Labour Programmes inHungary.” Economics of Transition 5(2):453–84.

Orazem, Peter F., and Milan Vodopivec. 1996.“Male-Female Differences in Labor MarketOutcomes during the Early Transition to Market:The Case of Estonia and Slovenia.” Working Paper2087. World Bank, Washington, D.C.

Owen, John M., and Patricia J. Rogers. 1999.Program Evaluation: Forms and Approaches.London: Sage Publications.

Panizza, Ugo. 1999. “Why Do Lazy People MakeMore Money? The Strange Case of the PublicSector Wage Premium.” Working Paper 403. Inter-American Development Bank, Washington, D.C.

Rama, Martin. 1999. “Public Sector Downsizing: AnIntroduction.” The World Bank Economic Review13(1):1–22.

Ramamurti, Ravi. 1997. “Testing the Limits ofPrivatization: Argentine Railroads.” WorldDevelopment 25(12):1973–93.

Ray, Pranabesh. 2001. “Human Resource Issues inPrivate Participation in Infrastructure: A CaseStudy of Orissa Power Reforms.” ReportCommissioned for the PPIAF. PPIAF Case Study.World Bank, Washington, D.C.

Ruppert, Elizabeth. 1999. “The AlgerianRetrenchment: A Case Study of Argentina.” TheWorld Bank Economic Review 13(1):155–83.

Saghir, Jamal, and Robert Taylor. 1999. “PrivateSector Participation in Water Supply and SanitationServices.” Presentation, Cairo, December. WorldBank and International Finance Corporation,Washington, D.C

Sauti-Phiri, Charles. 2002. “Privatisation in Malawi.”Unpublished paper. Privatisation Commission,Blantyre.

Talley, Wayne K. 1999. “Ocean Container Shipping:Impacts of a Technological Improvement.” Paperpresented at the Annual Meeting of the AmericanEconomic Association, January. New York City.

Tawney, Clare, and Jacob Levitsky. 1997. “SmallEnterprise Development as a Strategy for Reducingthe Social Cost of Restructuring and Privatization.”Working Paper IPPRED–6. International LabourOrganisation, Geneva.

Terrell, Katherine. 1993. “Public-Private WageDifferentials in Haiti. Do Public Servants Earn aRent?” Journal of Development Studies42:293–314.

Valdez, Jose A. 2002. “Dealing with Labor Issues inRail Roads in Bolivia.” PPIAF Case Study. WorldBank, Washington, D.C.

World Bank. 2000. Port Reform Toolkit—Module 7:Port Labor Reform. Washington, D.C.

———. 2001. “Enhancing Public-Private Partnershipin the Context of the Africa Vision for Water(2025).” In Political Economy of Water SectorReform, Volume 1. Summary of RegionalConference on the Reform of the Water Supply andSanitation Sector in Africa. Kampala, Uganda.

———. 2002. Public Communications Programs forPrivatization Projects: A Tool Kit for Task TeamLeaders and Clients. Washington, D.C.

Labor Issues in Infrastructure Reform: A Toolkit

46

MO

DU

LE 1

This toolkit was made possible through financial contributions from the Netherlands Ministry of Foreign Affairs.

Labor Issues in Infrastructure Reform

A Toolkit

MODULE1