the postal sector in developing and transition countries

The Postal Sector in Developingand Transition Countries

Contributions to a Reform Agenda

Edited by Pierre Guislain

September 2004

The World Bank GroupGlobal Information and Communication

Technologies Department, Policy Division

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

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

Administrator

31664

i

The World Bank Group

Global Information andCommunication TechnologiesDepartment, Policy Division

Contributions to aReform Agenda

The Postal Sectorin Developing andTransition Countries

Edited by Pierre Guislain

September 2004

ii

THE POSTAL SECTOR IN DEVELOPING AND TRANSITION COUNTRIES

Disclaimer

This report was prepared by staff from the World Bank Group’s Global Information and Communi-cation Technologies Department. The findings, interpretations, and conclusions expressed herein donot necessarily reflect the views of the Board of Executive Directors of the World Bank or the govern-ments they represent.

To request additional copies of this report please contact the Global ICT Department, World Bank,2121 Pennsylvania Ave., NW, Washington, D.C. 20433, USA; email: [email protected].

Acknowledgements

This report was prepared by a team including Boutheina Guermazi, Pierre Guislain, Juan Ianni,Charles Kenny, Graeme Lee, and Isabelle Segni. Valuable contributions and comments were receivedfrom our reviewers and colleagues in the preparation of this document. Thanks go to Anat Lewin,Gareth Locksley, Paul Noumba Um, Christine Zhen-Wei Qiang, Andrea Ruiz-Esperza, Peter Smith,and Svetoslav Tintchev (World Bank), as well as Alex Dieke (WIK), Bob Cohen (US Postal RatesCommission), and the staff at Frontier Economics. Special recognition to Mark Andrew Wahl for thecover art and interior layout.

iii

Table of Contents

..............................................................................................................iv

...........................................................................................................1

......................................................................5

................................................................................................16

................................................................................................29

.........................................................42

.........................................................................52

...................................................................................................53

...................................................................................................59.................................................................................66

..........................................................................................................72

........................................................................................................75

................................................................................................................78

Foreword

Chapter 1:Introduction

Chapter 2:Reinventing the Post OfficeBy Pierre Guislain and Graeme Lee

Chapter 3:What Drives Postal Performance – A Cross-Country AnalysisBy Charles Kenny

Chapter 4:Questioning the Monopoly-Supported Postal USO in Developing CountriesBy Charles Kenny

Chapter 5:Postal Policy and Regulatory Reform in Developing CountriesBy Boutheina Guermazi and Isabelle Segni

Chapter 6:Case StudiesBy Isabelle Segni and Juan Ianni

MoroccoTanzaniaTrinidad and Tobago

Chapter 7:Conclusion

Bibliography

Authors

iv


From the Editor

The World Bank Group has been involved inpostal sector reform for over a decade. At present,we support reform in about 25 countries in allregions of the world, with special emphasis onAfrica and Latin America. During this decade,the sector itself has changed significantly. Theerstwhile paradigm of public sector monopolyjustified by the need to deliver and finance spe-cific universal service obligations is crumblingunder the dual forces of competition (from otherproviders of postal services as well as substituteservices such as electronic communications) andfiscal restraint.

The postal sector has started a transition processwitnessed before in telecommunications and othernetwork industries, from state monopoly to pri-vate-sector-led competition. We are only in theearly stages of this process and no clear modelsor consensus have yet emerged as to the best routeto follow, in particular in developing and transi-tion countries. While sector reform in telecom-munications started over 20 years ago and is stillin process, time is running out for the postal sec-tor and reforms will need to be implemented at afaster pace if the sector is to be saved from irrel-evance.

This compilation was prepared as a contributionto the Universal Postal Union’s 23rd Congress(Bucharest, September-October 2004) and itsWorld Postal Strategy Forum. It includes sev-eral articles addressing key issues postalpolicymakers, regulators and operators are fac-ing today. Each chapter looks at the reform pro-cess from a different angle, and the views of thevarious authors do not always coincide.

Several challenges exist to rigorous research andpolicy analysis on the postal sector. These in-clude the confusion surrounding the use and scopeof the terms postal sector or services, as well asthe paucity of data and benchmarks on sectordevelopment. While information on the perfor-mance of postal incumbents or operators is al-ready limited and uneven, information on othersectoral actors is quasi non-existent in any stan-dardized or consolidated format. This is an areawhere the UPU and its membership could pro-vide an increased and much needed contribution.

We hope that this compilation will provide a use-ful contribution to postal sector reform in devel-oping and transition countries and welcome com-ments and suggestions.

Pierre GuislainManager, Policy Division

Global Information and Communication Technologies DepartmentThe World Bank Group

1

Chapter 1

Introduction

The scope of the postal sector broadly cov-ers pick-up, transport and delivery ofletters and parcels, post office counter

services related to letter and parcel delivery, andother related services such as mailbox rental (seeBox 1). At the same time, post offices (the in-cumbent operator that provides postal sector ser-vices, usually under legal monopoly) often offera range of services outside of the postal sector—pensions payments, telephone or Internet access,financial services, and retail services for example.

Postal services have existed for over two thou-sand years and have historically been importantto the development of nations. The Roman postalservice, the cursus publicus, used a network ofmessengers carrying mail up to 80 kilometers perday; illegal use of the postal system was punish-able by death. The modern postal service goesback to 1840 when the stamp was introduced inGreat Britain. The Postal Sector in Developingand Transition Countries reports on several em-pirical studies that posit a link between postalsector development and broader developmentprogress.

At the same time, it is clear from the evidencepresented in this compilation that in many de-veloping countries the postal sector is far fromhealthy. The sector, and in particular the in-cumbent post office, is characterized by low vol-umes, poor quality of service, and weak finan-cial performance. In the majority of developingcountries both the postal sector and the postoffice urgently require reform—from the levelof policy, through regulation, and to corporaterestructuring.

The papers in this collection suggest the extentof the policy, regulatory and corporate reformagenda in improving the performance of thepostal sector in developing countries, as well assome approaches to such reform. At the sametime, the papers suggest that the future of thepost office in developing countries as a viablebusiness, rests in part on the provision of a rangeof services that extend far beyond the postalsector.

INTRODUCTION

2


Box 1. Defining the Postal Sector

What do the papers suggest about the nature ofthe postal sector and the incumbent postal op-erator in developing countries?

n Postal volumes in poorer countries are low.Questioning the Monopoly suggests that,on average, countries with GDPs per capitabelow $1,000 see fewer than one letter percapita, per year delivered, compared to

about 100 letters per capita, per year incountries with GDPs per capita over $5,000.

n The quality of service provided by the in-cumbent operator is often poor. Trust inthe postal sector is low and legal universalservice obligations (requiring letter deliv-ery to all citizens at a given quality) are farfrom being met by the incumbent (PostalPolicy and Regulatory Reform and Ques-tioning the Monopoly).

Webster’s provides an interesting word history forpost: “The word post, meaning “mail”, is ultimatelyderived from Latin ponere, “to place, put in posi-tion.” This meaning of the word post was a result ofthe method of delivering mail. In the 16th century,horsemen were stationed at designated places alongcertain roads to ride in relays, with royal dispatchesand other papers. These couriers were called “posts.”As the system of mail delivery expanded during thenext two centuries, post was applied to a delivery ofmail and then to the organization responsible for theentire system of delivering mail.” “In Middle En-glish times the word mail meant simply “bag,” espe-cially one used by a traveler for provisions. Suchbags were used to carry letters, and the word maileventually came to designate the contents rather thanthe container.”

The United Nations International Standard IndustrialClassification, Rev. 3.1 (ISIC), breaks the “post andcourier activities” sector (group 641) down into “na-tional post activities” and “courier activities other thannational post activities.” National post activities in-cludes “pick-up, transport and delivery (domestic orinternational) of mail and parcels; collection of mailand parcels from public letter-boxes or from postoffices; distribution and delivery of mail and parcels;mailbox rental, poste restante, etc.; mail sorting; and

sale of postage stamps. This class excludes postalgiro and postal savings activities, and other financialactivities carried out in combination with postal ac-tivities.” Courier activities other than national postactivities includes “pickup, transport and delivery ofletters and mail-type parcels and packages by firmsother than national post. This class excludes similaractivities carried out by the national postal author-ity.” The distinction between postal and courier ac-tivities is highly interesting as it is based more on theprovider of the service than the activity itself, whichis unusual for an ISIC classification.

There is a somewhat indistinct border between thepostal and logistics sectors, with the postal sector tend-ing towards ongoing collection and delivery networksproviding services covering comparatively small itemsto a wide population base while the logistics sector ischaracterized by point-to-point deliveries of biggeritems to and from a comparatively small base of largecustomers. The definition of logistics provided bythe WTO is as follows: “Logistics deal with the sup-ply chain process that plans, implements, and con-trols the efficient and effective point-to-point flow andstorage of goods, services, and related information,throughout the production, distribution and deliverystages, from the initial suppliers of inputs to finalcustomers of products.”

3

INTRODUCTION

n The efficiency of incumbent operators isweak. Questioning the Monopoly suggeststhat developing countries deliver perhapsone-tenth or fewer letters per employee thandeveloped countries.

n While most postal sectors have a significantde jure monopoly in place covering letter de-livery (frequently up to 1 kg), de facto theposition is one of competition. Postal Policyand Regulatory Reform suggests that the lowquality of incumbent service has driven manybusinesses to provide their own postal ser-vices to deliver bills, for example.

n The financial situation of the incumbentoperator is dire. Postal Policy and Regu-latory Reform suggests cumulated losses ofincumbents can run as high as 2 percent ofGDP, and that three-quarters of Africanpublic postal operators report a negative netfinancial position.

These linked problems suggest the urgent need forreform, and the conclusions drawn by the papersmight be usefully divided into those that suggest apolicy and regulatory reform agenda for the postalsector, those that look towards the restructuringof the post office to better provide postal services,and those that discuss an expansion of the postoffice into new sectors or activities.

The papers are all agreed that the broad approachto policy and regulatory reform should involvegreater competition and a reduction of the mo-nopoly as well as improved regulation and a re-definition of the universal service obligation.Reform to allow de jure competition has an im-pact even though (as we have seen) de facto com-petition is frequently widespread. Postal Policyand Regulatory Reform notes that the Chineseexpress mail segment might triple in value underan improved regulatory environment that allowed

for more open competition. The discussion ofTanzania in the Case Studies suggests that in-creased competition can be introduced in tandemwith operator reform to strengthen overall sectorperformance. It should be noted that the paperstake different positions on the speed of reformtowards competition. Questioning the Monopolysuggests an immediate opening to full competi-tion, while Postal Policy and Regulatory Reformsuggests a more gradual approach modeled onthe European Union’s experience.

Along with competition comes an increased needfor regulatory instruments to ensure that compe-tition is fair. Postal Policy and Regulatory Re-form discusses the General Agreement on Tradein Services’ Reference Paper on regulation as auseful set of guidelines for the design of a regu-latory framework in the postal sector.

As part of policy and regulatory reform, it isalso important that the universal service obli-gation is revisited. The demand for postal sec-tor services is highly correlated with income.What Drives Postal Performance suggests thatabout 73 percent of the variation in letters percapita across countries can be predicted byvariations in GDP per capita. Questioning theMonopoly notes that, even in the best of cir-cumstances, demand for postal sector servicesespecially in the rural and remote areas of de-veloping countries will remain low. The papersuggests that given the limited demand for uni-versal postal service, along with limited finan-cial capacity to support its provision, it mightbe abandoned in developing countries, to be re-placed by a government commitment to providea range of services in greater demand by poorpeople. At the least, as suggested in PostalPolicy and Regulatory Reform, the level of ser-vice demanded of the incumbent should be ad-justed to be financially feasible.

4


Regarding incumbent reform, the inefficienciesin most incumbent post offices suggest that thereis no necessary trade-off between reducing in-cumbent losses and delivering improved ser-vices. Both the Case Studies and the cross-coun-try analysis carried out in What Drives PostalPerformance suggest that reformed post officescan deliver more post with greater efficiency andreliability all while reducing the need for finan-cial transfers. Reform techniques discussed in-clude management and performance contractsas well as franchising retail outlets. Experi-ence suggests that capacity building is a keysuccess factor in restructuring the incumbent,together with improved accountability, and gov-ernance mechanisms.

Reinventing the Post Office and Questioning theMonopoly discuss the potential for the post of-fice to provide more and different services. Givena limited demand for letter services in poorer de-veloping countries, but a significant and geo-graphically widespread demand for a rangeof services from other sectors including retail,financial and government services, the future ofthe post office may rest in provision of those otherservices. A global network of 700,000 post of-fices suggests considerable reach for their deliv-ery, argues Reinventing the Post Office.

Whatever this potential role, expanding the rangeof services delivered by post offices clearly be-longs at the end of a sectoral and corporate re-form agenda which sees the emergence of a com-petitive sector and an efficient incumbent. Fur-thermore, if the service delivery involves an ex-plicit or implicit government subsidy to extendaccess, Questioning the Monopoly argues that acareful economic analysis of the benefits of theservice, as well as a transparent and nondiscrimi-natory award of the subsidy would be importantcomponents of a new universal service agenda.

In sum, the papers suggest that in developingcountries there are benefits to be reaped by re-form of the postal sector and of the post office.Given the weak state of sectors and incumbentsin many countries today, reform will frequentlybe able to deliver more widespread quality ac-cess to a greater range of services in higher de-mand by poor people in particular, while reduc-ing the drain on government revenues. All of thismight deliver considerable improvements in eco-nomic performance, income generation, and thequality of life in developing countries.

5

REINVENTING THE POST OFFICE

Chapter 2Reinventing the Post Office

Pierre GuislainGraeme Lee

Modern postal services have been pro-vided for well over a century by postoffice administrations in virtually ev-

ery country in the world. Today the future ofthese postal offices is in question in many devel-oping and transition countries.1 Traditional postalservices (mail and parcels) remain their core prod-uct, but cannot by themselves cover the costs ofpost office networks. A more dynamic post officerequires a management structure with the freedomand incentives to increase efficiency, client orien-tation and innovation. Several approaches for in-creasing the flexibility and dynamism of post of-fice management are described in this paper.Longer term sustainability may involve: in-depthrestructuring and corporate reform; a smaller postoffice-owned network supplemented by private-public partnerships and franchising schemes; di-versification of services with post offices servingas a platform of service delivery for private sectorpartners; and provision of a broader range of gov-ernment services. Such transformation will alsorequire changes in policy and regulation regard-ing the scope of the monopoly and of the universal

service obligations (USO). These issues are cov-ered at greater length in chapters 4 and 5 of thispublication.

The Postal Dilemma

Of the 1.2 billion items posted daily worldwide,the United States Postal Service (USPS) handlesalmost 50 percent with an average of close to 700items per capita, per annum. Yet 25 percent of allcountries generate less than one item per capita,per annum, and 50 percent of all countries gener-ate less than 10 items. These low volumes prevaileven in countries with considerable post office net-works, which were developed when mail was themain communication mode and maintained in anattempt to meet USOs.2 As an example, theMaldives has a post office for every 1,500 people,with each office handling an average of only twomail items per day (compared to France whichhas a post office for every 3,000 people, with eachpost office handling on average 5,500 mail itemsper day). Over 50 percent of all countries average

6


less than 200 items per post office per day.3 Postoffice networks can only be sustained on such lowvolumes with significant transfers of scarce gov-ernment resources.

Mail volumes are highly correlated with per capitaGDP (see chapter 3) and developing country postoffices may thus expect future increases as theireconomies grow, through this may arrive too latefor many of them. Indeed, global trends are notparticularly encouraging for developing country postoffices, as competition from email and the Internetis cutting down on traditional mail volumes, andcompetition from couriers and express carriers isputting pressure on their parcels business.

The desire to ensure widespread access to the postoffice network (as embodied in USOs in manycountries), combined with very low demand forthe core service provided by that network, is a di-lemma facing most of the developing world’s postoffices. The main options in the struggle to over-come this dilemma are greater efficiency in net-work and service provision, more commercial ap-proaches and innovative product development, aswell as greater use of that network. Reinventingthe post office encompasses these various dimen-sions and requires significant policy and regula-tory reform (see chapter 5 on postal policy andregulation). This paper focuses on the post officeitself, rather than on the postal sector, and looks atthe main business and organizational models thatcan structure a modern postal enterprise, and atthe scope of services and business lines that postoffices can offer or deliver.

Is there Political Will to Reformthe Post Office?

The Post Office is a large employer in most coun-tries and sometimes the sole presence of the State

in remote areas. Politicians, employees, and thepublic-at-large know reform can lead to a reduc-tion in post office jobs and closure of some postoffices. Trade-offs need to be made. Do theywant to serve rural areas with limited demandfor postal services ? If so, how will this be funded,and are government or politicians willing to allo-cate scarce budget funds to support this pres-ence? Are politicians willing to forgo the oppor-tunity for patronage, too often provided throughpostal employment?

Where the will to reform exists, governments mustbe ready to tackle resistance to postal reform.Postal reform may indeed lead to staff reductions,closure of post offices, and changes in the statusof employees. But it should also lead to improvedproductivity, performance, customer and em-ployee satisfaction, and in the longer term newjobs. Post office closures can be mitigatedthrough franchising post office services in ruralareas leading to better access to customers. Staffreductions can be managed in a gradual way,starting with early retirements and voluntary sepa-ration packages, for example. Staff may for in-stance be encouraged to leave the post office ranksto establish franchise operations, with initial sup-port from the post office.

Restructuring the PostalIncumbent

Until recently the postal sector was among thefew sectors where little attempt was made to at-tract private sector involvement. It was only to-wards the end of the 1990s that privatization wasinitiated in countries such as Argentina. Marketliberalization has been undertaken at an evenslower pace and many countries still cling to thetraditional public sector monopoly model.

7

Figure 1. Legal structure of incumbent operator

Globally, postal sector reform and post officecorporate reform tend to be in their early stages;the record of reform attempts to date is some-what mixed and best practices have not yet fullyemerged. Nonetheless, reform experience tendsto suggest that, in common with other sectors,greater private participation in the postal sectorcan improve the quality and efficiency of servicedelivery. Commercialization is a necessary partof this process involving greater customer orien-tation, introduction of commercial and account-ing practices, enhanced professionalism of staff,a focus on profitability, and more generally thedevelopment of a business culture.

Today many incumbent post offices in the devel-oping world remain postal administrations, oftendepartments of a ministry that may also includecommunications, telecommunications or trans-port. They lack appropriate accounting proce-dures, operate from a government-defined bud-get and are usually loss making, although profitsand losses are often difficult to determine due tothe lack of commercial accounts. Political inter-ference tends to be high with political appointeesmanaging the organizations. There is frequentlya lack of management accountability or objec-

tives, over-employment, inefficient operations,and tariffs (at best) cover only operating costsleading to no investment. The network is typi-cally fully owned and operated by the post of-fice, poorly equipped and under capitalized, andretail outlets lack technology. Marketing andcustomer service are not emphasized leading topoor brand image and falling volumes.

Effective Corporatization andCommercialization

In order to increase efficiency of the postal ad-ministrations, corporate reform usually startswith a change in the legal structure moving frompublic administration to statutory corporation,and subsequently to a publicly or privately ownedlimited liability company (Figure 1).

While the legal transformation from a govern-ment bureaucracy to a company operating on acommercial basis is necessary for successful postoffice restructuring, it is not sufficient and needsto be accompanied by a strong commercializa-tion effort by introducing commercial objectivesand practices into management and operations


Public Administration Statutory Corporation Public Company Private Company

Post office as a government department, entity or public agency

Post Office organized as a statutory

corporation under public law

Post Office organized under company law, typically as a limited

liability company

Post Office organized as a company, fully or

partially privatized; may be publicly quoted on a

stock exchange

Public Law Private Law

Public Ownership Private Ownership

8


Box 1. Correos de Chile

In Chile, postal reform has been ongoing since 1982, when Correos de Chile became a statutory corporationgoverned by a board of five non-executive directors designated by the government through the board of theSistema de Empresas Estatales. The Correos board meets twice a month, selects the CEO, establishes thestrategic plan and supervises the administration of the enterprise. Employment contracts are governed by thelabor code.

Between 1982 and 1989 the postal network and staffing were restructured to improve productivity and tomore closely respond to market demand. From 1989, the reform of the postal operator has focused on creatinga market and customer driven culture, which has included the forming of commercial alliances with the privatesector and significant product and service diversification. In October 2002, over 300 rural postal outlets werefranchised to reduce costs and extend service hours. In 2003, Correos de Chile delivered a return on equity of11 percent after paying 57 percent tax on profits (17 percent general tax and 40 percent tax applicable to stateowned enterprises).

and establishing a business culture in the postalenterprise. This may also include greater pri-vate sector involvement through outsourcing, sub-contracting and franchising activities to privatecompanies.

While reform is seen as a necessity to improve theviability of the postal incumbents, there is no simpleuniversal blueprint. Indeed, one of the most suc-cessful postal operators in the world, the UnitedStates Postal Service (USPS), remains a publicadministration but operates in a commercial man-ner. In the case of USPS, governance reform isnot likely to take place in the short term but thatdoes not stop them from outsourcing and sub-con-tracting key elements of their work to the privatesector. Correos de Chile, which was transformedfrom a public administration to a statutory corpo-ration in 1982, is another example of a postal op-erator that has applied the principles of commer-cialization (see Box 1).

Moving from a postal administration to a com-mercial organization and transforming the postoffice into a statutory corporation or limited li-

ability company is an essential step in reformingthe incumbent. This includes among others, sepa-rate budgets and accounts, creation of a board ofdirectors, application of commercial and laborlaws, and so on.

A useful tool to monitor the performance of thestate owned postal operator is the performancecontract by which the government and the postaloperator negotiate and agree on objectives andthe means to reach them (see Box 2). Typicallyperformance contracts deal with: (a) legal andinstitutional changes; (b) economic regulationsuch as tariffs, price caps, salary increases, andemployment targets; (c) scope of universal ser-vice, and subsidies to compensate for their cost;(d) financial commitments to increase capital orprovide loan guarantees; and (e) financial andoperational performance targets (revenues, costreductions, productivity, quality of service andcustomer satisfaction, etc.). They tend to be anexcellent tool to track success in efforts to im-prove public sector performance, but typicallylack the incentives needed to make significantimprovements or sustain them.

9

Box 2. Performance Contracts

In Tanzania, reform of the postal operator began in 1994 with the creation of Tanzania Posts Corporation(TPC), Tanzania Communications Commission (the regulator), and a clear definition of responsibilities be-tween government, regulator and operator. TPC operates under a performance contract with incentives andpenalties for achieving and failing targets. The reform has resulted in a viable postal service in one of thepoorest countries in the world, with increased access and diversification of services. Market liberalization hastaken place and new licensed competitors are now operating within Tanzania.

In Morocco, Barid Al-Maghrib (BAM) was established as a state-owned company in 1998, and has since beenconstantly profitable, funding its own retrenchment plan and investments. A performance contract was signedwith the State. BAM has entered a number of partnerships with private sector firms.

See chapter 6 for more details.

Establishing the post office as a limited liabilitycompany also gives the government the option ofselling shares to private investors, since the com-pany is already established under company law(see section on privatization below).

Private sector participation can also be devel-oped through outsourcing non-core activities,franchising retail outlets and providing third partyservices through postal outlets. Non-core activi-ties such as staff restaurants and cleaning ser-vices can easily be outsourced. Building owner-ship and maintenance are another example.Postal office buildings can even be sold to an-other company and leased back (be it in part) tothe core business. Even core functions such astransport services can be provided by third par-ties who can share logistics costs and reducepostal fleet costs. In a number of countries, fran-chising has led to a significant increase in accesspoints in rural areas. Provision of non-postalservices on behalf of other companies is a usefulway of generating additional income and expand-ing the scope of the postal retail service, as fur-ther discussed below in the section on diversifi-cation.

Management Contracts andConcessions

A number of governments have opted for a man-agement contract to rapidly (and drastically) im-prove the operations and the financial perfor-mance of the postal incumbent. Trinidad andTobago and Guatemala offer good examples ofsuch a scheme (Box 3).

Although management contracts do not involveas strong incentives than would be the case withprivate ownership and investment, they can be agood way to achieve a quick transformation ofthe postal operator through the introduction ofexperienced and focused management. The twomanagement contract examples shown in Box 3were very different in concept but resulted in simi-lar transformations. The key to the success inboth was the incentives given to the managementcontractors. In the case of Trinidad and Tobago,the contractor was paid for performance in fivekey areas, whilst in Guatemala the contractor re-ceived payment for every universal service letterposted, a clear incentive to increase mail volumes.


10


Box 3. Management Contracts

Trinidad and Tobago outsourced the management of TT Post to help transform the company into a commer-cially focused business which the government could later privatize. In a two-phase reform process the initialfocus was on the rehabilitation of the incumbent postal operator through a delegated management arrangement(DMA), a five-year management contract. Payment of the fees of the DMA was related to the achievement offive key performance targets in the areas of: customer satisfaction, universal delivery, quality of service, totalrevenue, and net income. Four of the five targets have been met with substantial increases in volumes,revenues, and delivery, significant improvements in customer satisfaction and quality, but TT Post has not yetachieved a positive net income. See “Case Studies” for more on Trinidad and Tobago’s reform.

In Guatemala a management contract was awarded in 1998 to take over the management of the postal service.The contractor had an obligation to provide a basic service for letters weighing up to 20 grams. In return thecontractor received a government subsidy per item, and had exclusive rights to the postal network which wasused as a platform to provide other services. The management contract resulted in increased volumes, improvedquality and efficiency, and a reduction in subsidies paid by government. The success of the management contractled to the signing of a 10-year concession agreement in May 2004. The agreement changes the method of payingfor the USO, and a regulatory function is being established to regulate all operators in the market.

Concessions are another approach to introduceprivate participation by providing a private com-pany with the right to operate the postal enter-prise at its own risk for a fixed period. In returnthe concessionaire commits to investing in thecompany and to the provision of specified ser-vices. The most cited example of concessions indeveloping countries is Argentina (Box 4).

Despite the recent failure of the Argentine con-cession, the concept remains an attractive wayof introducing private sector participation in thepostal sector. The key ingredients to successfulconcessions are to ensure that the government,the concessionaire and the customers all benefitfrom the concession agreement. The recent Gua-temala concession is a good example. Govern-ment pays the concessionaire for the provision ofuniversal service (per item) and in return receivesa percentage of all revenues. It also receives afixed rental payment for postal outlets and build-ings and the rights to certain services. Other key

ingredients to a concession agreement are clearservice targets and access goals.

Privatization

The final step of the reform of a state-owned en-terprise may lead to privatization. This gener-ally includes the transfer of ownership from thepublic to the private sector through the sale ofshares. The German and Dutch post offices areamong the few examples of privatized postaloperators.

An initial 29 percent of Deutsche Post shares weresold in November 2000 through an initial publicoffering, allowing the government to maintaincontrol over the public operator, while introduc-ing a more commercially minded management struc-ture. The diversification, growth and profitabilityof Deutsche Post since its IPO illustrate the oppor-tunities created by privatization. The German gov-

11

Box 4. Concession Contract in Argentina

In Argentina, a 30-year concession was awarded in 1997 to operate the ailing and loss making postal operatorfor a concession fee of approximately $100 million per annum. Correo Argentino SA transformed the organi-zation, gave it a new corporate image, increased quality of service, diversified products and services, reducedemployee numbers by 40 percent and invested over $350 million in the company. However, the major reces-sion that hit Argentina, the high price of the concession fee, the failure to properly assess the cost of theuniversal service requirement, the failure to regulate illegal competition, and difficulties in dealing with laboragreements ultimately led to the concession being revoked in November 2003 due to nonpayment of the fee.

ernment still owns about 62% of the company (di-rectly and through KfW), with over 37 percent freelytraded on the stock exchange.

TPG, the Dutch postal operator was privatized in aprocess that began in 1994 when the governmentsold 30 percent of its share in KPN (the combinedpost and telecommunication holding company at thetime), followed in 1996 by a second tranche of 25percent. In 1998, post and telecommunicationsactivities were split and TPG demerged from KPN.

In Malaysia, Pos Malaysia was corporatized in1992. Shares in the company have since been soldwith 30 percent owned by government, 15 percentby employees, and the remainder owned by a finan-cial services company. The Ministry of Energy,Telecommunications and Multimedia regulates tar-iffs and international mail treaties, and the Malay-sian Communication Multimedia Commission regu-lates Pos Malaysia and courier companies. PosMalaysia has a special license (through 2034) tomaintain the role of the universal service provider.The partial privatization has contributed to improvedfinancial and operational effectiveness and the di-versification of products and services.

However, privatization may not be a feasible op-tion in many countries for political and opera-tional reasons. The privatization of the BritishPost Office was prevented in the mid 1990’s due

to political pressure. Implementing a successfuldivestiture will generally be more difficult forcountries with limited postal markets and a weakincumbent. In developing countries, privatizationmay not be attractive to the private sector with-out significant prior restructuring of the businessand the environment it operates in. In such cases,postal strategy should aim at optimizing the postalsystem performance through management reformand efficiency gains.

Sustainability of Restructuring

Concessions, management contracts, and exter-nal consultants have been used to introduce man-agement expertise and commercial practices intopostal operators. They have led to significantoperational improvements in most cases. Whilstsome may consider Argentina’s reform a failureafter the concession was revoked, the governmenthas taken over a much improved operation whichis more customer focused and quality managed.The Trinidad and Tobago operator is still notprofitable, but 97 percent of the population nowreceives daily delivery to their homes, compared to50 percent in 1998. Both are in a far better situa-tion than they were prior to the process of reform,which is an important measure of success. Tanza-nia and Brazil perhaps best demonstrate the keyingredients of successful reform (Boxes 2 and 5).


12


Box 5. Postal Reform in Brazil

Correios de Brazil is undergoing a commercialization process based on strengthening the postal operator’s ca-pacity and expanding the product offering through a wide range of joint ventures and partnership agreements.Correios de Brazil has a strong social obligation and has successfully introduced a range of financial servicesreaching a sector of society that previously did not have access to such services. Correios de Brazil is financiallyviable and has the highest rating of public services in the country with a 98 percent satisfaction measure.

Their reform contains the elements of private sec-tor participation, independent regulation, manage-ment accountability, and political commitment allof which are key to a successful reform process.

Financial sustainability requires additional changes.Significant improvements in efficiency often involvea reduction in staff numbers. Labor costs accountfor as much as 80 percent of total post office costsand yet, in a retail context, staff numbers cannoteasily be changed to reflect volumes of transactions.In France, 33 percent of post office outlets oper-ated by La Poste have daily activity levels of lessthan 2.5 hours resulting in a high cost per transac-tion. La Poste’s solution is to maintain the numberof access points through agency agreements withlocal government offices, and retailers. But thechallenge is even greater for an incumbent opera-tor in a developing country, where mail volumesare significantly lower and the scope of servicesnarrower.

In addition, to the extent governments wish to re-tain universal service obligations in a competitivemarket, they will need to be carefully costed andrescaled to a financially sustainable level (see chap-ters 4 and 5). Some post offices may be closed,deliveries may be cut, and/or customers asked tocollect their mail from post offices, for example.The case of Tanzania, where a reformulation of theUSO has accompanied broader reform, is furtherdiscussed in chapter 6.

Diversification

As emphasized above, in most developing and tran-sition countries, postal services cannot by themselvessustain existing post office networks. Volumes tendto be so low, and demand for postal services so lim-ited (in particular among the poor) that the survivalof the post office network may depend not only onhigher quality services and proactive marketing andpostal product development, but also on the abilityto provide other services for which there is a higherdemand in the communities served. With a net-work of almost 700,000 postal outlets, post officesare indeed well placed to provide a whole variety ofservices to those communities. Diversification en-tails significant risks however, which should not beunderestimated. Many post offices worldwide al-ready provide services in multiple markets or seg-ments in which competition exists (see Table 1).

Postal Financial Services

Financial services may account for a significantproportion of postal service revenues, up to 90percent in the cases of Algeria and Azerbaijan.Post offices play a significant and, often, unrecog-nized role in access to basic financial services (pay-ments, remittances, savings, giro). Of the 750 mil-lion accounts in social financial institutions world-wide, 56 percent are believed to be postal accounts.4

Institutional frameworks vary from region to re-gion. Postal financial services are usually man-

13

Table 1. Segmentation of Services Offered by Post Offices

aged and operated by the state-owned postal opera-tor, often under a post office savings bank agencycontract with the ministry of finance (Algeria, Mo-rocco). In some countries, the postal savings aremanaged by separate (State-owned) banks (as inIndonesia, Kenya, Sri Lanka, and Uganda), througha subsidiary or as a separate entity, under the su-pervision of the central bank or banking regulator.In many cases, the post office enters into partner-ships agreements with financial services firms toprovide their services on an agency basis; typicalexamples include Western Union and Moneygramservices for the transfer of money. Third-partyagreements can also cover real estate services, as isthe case in Malawi where the Post Office rents outspace in its main offices to a local bank.

There is a widespread consensus that postal net-works could play a much stronger role in providingaccess to financial services, especially to poor andrural communities with limited access to such ser-

vices. There is also scope to expand postal financialservices from fragmented mono-products to integratedproduct packages including payment cards, savings,deposits, insurance, and eventually credit.

However it is difficult to translate new ideas intopractice. In several countries discussions on therepositioning of the postal bank have taken placefor more than 10 years, without conclusion. Inother countries steps have been taken to separatethe postal bank into a company operating indepen-dently from the post office. However, one of theissues that may arise following such separation isthe weak performance of the postal network. Thelack of control over quality, cost and performancemay lead the separated postal bank to look for al-ternative networks to conduct its business. Thisillustrates the need for postal operator restructur-ing to establish a strong retail platform from whichpostal financial services can be securely provided.


Core Postal Services Service Product Competitors

Letters Letters, registered mail, telegrams, newspapers

Internal/private distribution, fax, email; often still postal monopoly

Parcels Parcels and registered parcels to 20Kg

local couriers and express carriers, such as DHL, FedEx, UPS, TNT

Express Mail Letters and Parcels - Express same as above Other Services

Service Product Competitors Mailroom management managing and sorting mail for large

mailers facilities companies; direct mail companies

Hybrid mail email delivered by hand telecommunications Electronic Internet, SMS, Postal Telecenters ISPs, cyber cafés Logistics warehousing, picking and packing,

distribution transportation companies, freight forwarders

Retail phones, phone cards, stationary items etc

Shops, kiosks, news agents

Rural access access to people (for delivery of variety of services)

schools, community centers, religious centers

Government licenses, taxation, pensions, benefits, information

local government, health services, schools

Financial savings, giro, payments, remittances banks, Moneygram, Western Union

CORE POSTAL SERVICES

OTHER SERVICES

14


Care should also be given to ensure that postal fi-nancial services are well integrated in the country’soverall financial sector and that they do not competeunfairly with other financing institutions. Establish-ing such a level playing field may require subjectingthe postal financial services to the oversight of thefinancial sector regulator.

Postal Telecenters

Several post offices have already tried to develop anInternet access service, with few reports of success.In many cases, pilots have been discontinued due toshortage of electricity, lack of proper maintenance,slow access, limited volume of customers, competi-tion from cyber cafes, and lack of training of thepostal staff, among other reasons. Overall, projectsare frequently poorly managed, with no market analy-sis (demand, state of the competition, tariffs), infra-structure constraints (electricity, telephone connec-tivity), little understanding of the legal and regula-tory requirements, lack of a business plan, and notenough planning for capacity building.

A recent study of telecenters in Southern Africa foundthat under a number of assumptions (five comput-ers, 850 clients per months, VSAT access of128kbps, 100 percent subsidized training and assis-tance and VSAT equipment assistance), Internetaccess in post offices in towns (50,000 to 250,000inhabitants) might be sustainable. Sustainability inrural areas might also be reached (with, say, threecomputers, 650 customers, 56kbps access) but inremote areas, additional subsidies would be required.5

Prerequisites for Diversification

Diversification requires substantial capacity build-ing, a strategy compatible with the policy, legal andregulatory framework, market analysis, tariff set-ting, preparation of business plans, and personneltraining.

Diversification may focus on provision of commer-cially competitive services or of government ser-vices. The two approaches are not exclusive, butthey do suggest different priorities in terms of broaderreform. If the post office is to compete with privatesector providers, such competition should occur ona level playing field. Dominant operators (includ-ing the post office) should not be able to provideanti-competitive cross-subsidies (hence the impor-tance of open competition and, if necessary, regula-tion) and should not enjoy customs or tax exemp-tions, government guarantees, reserved procurement,or other privileges not enjoyed by private competi-tors. It also requires that post offices are freed fromthe shackles of bureaucracy, uncompensated univer-sal access requirements, and public sector-specificlegal provisions covering areas such as employmentor procurement. In other words, diversification intoprovision of competitive services requires broaderliberalization of the postal sector (and developmentof an effective sector regulator) as a whole to occurin tandem with corporate reform. On a corporatelevel, it requires transparent cost accounting, ad-equate incentive schemes, and capacity building.

If the post office diversifies into government ser-vice provision on a noncompetitive basis (provid-ing licenses, e-government services, and outreach,for example), the urgency for corporate reform is toensure that needed services are provided as efficientlyand effectively as possible. This is likely to involveconsiderable cross-government coordination and citi-zen consultation regarding the services to be pro-vided. Government agencies will need to better ar-ticulate how they could use postal networks for bet-ter delivery of their services to citizens. Again, thiswill have to be combined with the implementationof corporate restructuring strategies such as thoseoutlined above.

Whichever diversification option is chosen (commer-cial or government services) one should bear in mind

15

that the track record of diversification by state ownedenterprises, outside of their core business, has byand large been disappointing in developing and tran-sition countries. Any diversification strategy shouldbe based on a solid assessment of the incumbent’sstrengths and weaknesses, as well as the opportuni-ties and threats it may face in its core business andin the new business being considered.

Conclusion

Post offices throughout the world were establishedto facilitate the conveyance of mail. To fulfill thecollection and distribution of mail, governments builta network of post offices to provide the entire popu-lation access to postal services. Many countries re-alized the potential of such post offices in the provi-sion of other services, thus, post offices diversifiedinto providing financial and government services.But the core product still remained mail services.

What exactly is the role of post offices in today’sworld? Governments need to define the purpose oftheir post office: is it a mail distribution company;is it an access point to government products and ser-vices; is it an access point to financial services; andis it a commercial self sustaining business?

How should it be run? If it is operated on a com-mercial basis, nonprofitable outlets will need to beclosed and service levels reduced in areas of lowdemand. If this is not acceptable to governments,they must define adequate schemes to maintain andfinance the desired service levels and implement themin a transparent way. For example, the United King-dom government pays Royal Mail a subsidy spe-cifically to maintain the rural network of post of-fices, which would otherwise be closed.

Governments should consider divesting themselves ofmanagement control of their post office and put inplace commercial practices and management. Social

or public service obligations must be clearly deter-mined and their net cost calculated and compensated.

Post offices are still contributing to economic growthin many countries, but in others, antiquated, under-funded networks, and lack of competition limit orreverse that contribution. Governments must reformtheir post offices to ensure they, once again, becomean enabler of economic development for the countryas a whole, potentially through leveraging their ex-tensive retail network. As this increasingly involvesthe provision of services other than traditional postalservices, governments will need to ensure that thisdiversification is carried out in a pro-competitivemanner. This means no favored treatment of postoffices relative to their competitors, but also freeingpost offices from the bureaucratic shackles that con-strain them.

Notes

1 The term “post office” as used in this paper refers tothe incumbent postal operator, whether it is a publicadministration, public enterprise or private company.

2 The Universal Postal Union (UPU) enjoins its mem-bers to support the concept of universal service, whichshould provide all citizens “quality basic postal ser-vices at all points in their territory, at affordableprices”.

3 Even this number is inflated as the real number ofitems passing through each post office will actuallybe lower since business and government mail typi-cally accounts for about 80 percent of mail flows andoften bypasses retail outlets.

4 Consultive Group to Assist the Poor. “OccassionalPaper.” July 2004. p.5.

5 See “Postal Telecenters – Rural Developmentthrough Internet access in the Southern Africa De-velopment Community: the Cases of Malawi, Tan-zania and Zambia”. World Bank, July 2004.


16


Chapter 3What Drives Postal Performance?A Cross-Country Analysis

Charles Kenny

Based on data from the Universal PostalUnion (UPU) and the World Bank, thispaper develops indices of postal perfor-

mance and efficiency and analyzes the factorsbehind them. It takes a preliminary first cut atexamining the statistical relationship betweenpostal performance (as measured by letters/capita), trust in the postal service, and postalefficiency, and finds a significant link betweenthe three. The results suggest that reforms thatimprove postal efficiency and trust in the postalnetwork will improve the performance of thepostal network. However, our available mea-sures of reform progress are not strong enoughto evaluate the scale of this impact. Externalfactors including urbanization and literacy donot appear to contribute significantly to perfor-mance. At the same time, it does not appearthat more heavily subsidized post offices aredelivering better services.1

Introduction

In comparative terms, there has been little focuson the relationship between postal performanceand economic development. This is especiallytrue of econometric work linking the two, where,compared to a plethora of studies looking at therelationship between telecommunications rolloutand economic growth (see Forestier et. al. [2002]for a nonexhaustive list of 15 such studies), wecould find none exploring the statistical relation-ship between postal networks and economicgrowth.

Nonetheless, there are a number of potential linksbetween postal networks and development. Likeother communications infrastructures postal net-works allow for the flow of goods, services, andpayments between economic agents. An efficientpostal network can therefore significantly reducetransaction costs across an economy (particularlyfor billing and advertising services), forge mar-kets where they did not exist before, and supportknowledge transfer between agents that allow

17

innovation and growth. In developing countriesin particular, post offices are one of the most far-reaching information and communication net-works, and provide a widespread base for inter-action with the formal financial system. By con-necting people to both markets for goods and ser-vices and the financial systems that underpinthem, postal networks can be a potentially pow-erful tool for poverty reduction.

The theoretical links have been given some em-pirical backing in historical studies such asAhmed (1981), who suggests that postal networkexpansion was significantly linked to the growthof a national economy in India. A number ofrecent studies have added microeconomic andanecdotal support to the case for a link (Walsh,2001, 2002, and Ranganathan, 1998).

Given that postal networks might well play animportant role in the development process, it isimportant to understand what might help postaloperations provide better and more widespreadservices in developing countries. This paper takesa preliminary first cut at examining the factorsbehind postal performance. Based on data fromthe UPU and the World Bank’s World Develop-ment Indicators, it develops an indicator of postalperformance measured as letters per capita de-livered compared to the number of deliveries ex-pected given income levels.

The paper then goes on to discuss possible fac-tors behind varying scores on the postal perfor-mance indicator. Examining a correlation ma-trix of variables, it appears that postal efficiency(letters delivered per postal employee and postalexpenses per letter) are correlated with improvedpostal performance. Given the scale economiesthat characterize the sector, it may be that thereis a causal relationship flowing both ways. Re-forms that improve postal efficiency (and per-

haps more importantly the quality of services)will increase use of the postal network while,conversely, more use of the system itself will al-low greater efficiency. A factor linked to effi-ciency, promoting the number of letters deliveredat a given income, is trust in the postal network(defined below). The percentage of offices thatare run by nonpostal administration staff appearsto have a marginal impact on performance. Asecond measure of liberalization, as well as mea-sures of the use of postal financial services andexternal factors including urbanization and lit-eracy, do not appear to be significantly related toperformance. The Annex discusses a regressionanalysis that confirms these findings. The con-cluding section suggests policy recommendationsand avenues for further research.

Postal Networks in Development

There is a close relationship between postal per-formance and economic development. For tele-communications, the “Jipp curve” linking GDPper capita with telephone lines per capita canpredict about 80 percent of the variation in tele-phone lines per capita across countries at a singletime (see Forestier et. al., 2002). A “postal Jippcurve” can be constructed linking letters postedper capita with GDP per capita (data from UPUpostal database and World Bank World Devel-opment Indicators).2 The relationship is almostas strong as the original. About 73 percent ofthe variation in the number of letters per capitacan be predicted by GDP per capita (Figure 1).This suggests that there is a significant relation-ship between economic development and expan-sion of postal networks.

WHAT DRIVES POSTAL PERFORMANCE?

18


Developing the Postal Network

If a strong postal network has a relationship withdevelopment, it is important to find evidence onwhat might drive the development of a strongpostal network (here defined as one which deliv-ers more mail per capita). Supply and demandfactors will both play a role in determining thenumber of letters sent per capita.

On the demand side, income is clearly an impor-tant factor, as already suggested by the “postalJipp curve.” Perhaps literacy levels might alsoplay a role. On the supply side, one might plau-sibly expect price, extent, efficiency, and qualityof service to have an impact—indeed, such rela-tionships have been found in developed countrypostal services (Nankervis et. al.,1999). It is alsopossible that a post office network offering a rangeof ancillary services and/or services aimed at theneeds of businesses would attract a larger clien-tele that might post more letters.

At this point, as we have chosen to use a datasetthat covers as many countries as possible (as weshall see), data limitations force the use of proxyor imperfect indicators. Furthermore, we do nothave cross-country comparable data on price. Wedo have various measures of extent (measures ofpopulation covered and post offices per squarekilometer), efficiency, and quality of services,however.

Efficiency variables include ratio of revenues toexpenditures, postal revenues per staff, and let-ters delivered per employee and post office. Look-ing at potential determinants of efficiency, the gen-eral level of development may allow for econo-mies of scale or more efficient postal institutions.Private competitive involvement has been foundto correlate with improved service delivery (in-cluding reduced prices) in a number of sectors,and this includes studies of the postal sector indeveloped markets (Frontier Economics, 2002).Urban postal delivery is more straightforward

Figure 1. The Postal Jipp Curve

0.01

0.1

1

10

100

1000

100 1000 10000 100000

GDP per Capita (USD)

Let

ters

/Cap

ita/

Yea

r

GDP per capita

19

than rural delivery, as is common with most net-worked services. All of these factors may im-pact overall efficiency.

“Quality” is, at the outset, a comparatively ar-bitrary term. In previous postal studies it hasbeen measured using indicators such as numberof letters delivered on time. It may also encom-pass the linked concept of reliability—are lettersdelivered at all? Again, we do not have such dataat the cross-country level. Instead, we use ameasure of trust in the postal system (surveyanswers to a question regarding willingness tosend an object worth $100 through the posts) asa proxy for quality. This trust measure mightplausibly be related to measures of efficiency—bloated public monopolies frequently provide alower quality of service than efficient ones (World

Bank, 1995). Trust, and a broad definition of“quality” might also be related to the extent ofservice provision.

In order to empirically examine our posited rela-tionships, a range of relevant variables were se-lected from the UPU database as well as infor-mation from Kirkman et. al. (2002), and theWorld Bank’s World Development Indicatorsdatabase. These are described in Table 1, withdescriptive statistics in Table 2.3 There is con-siderable variation across countries in most ofthe indicators—for example, the number of let-ters delivered per year per employee varies from196 to 174,930, receipts as a percentage of postoffice expenditures varies from 11 percent to over400 percent, privately operated post offices as apercentage of all offices varies from none to 95

Table 1. Indicators for analysis


Postal index Index of postal success calculated as (log) actual minus (log) predicted letters/capita from regression of log letters/capita against log GDP/capita (PPP). Letters/Capita from UPU, GDP/Capita from WDI.

GIT Postal trust Score on GIT index of ‘trust on postal services’ (survey response to question: ‘Do you trust your country’s postal system sufficiently to have a friend mail a small package worth $100 to you (1=not at all 7= yes)’). Source Kirkman et. al. (2002)

Illiteracy Illiteracy rate, adult total (percentage of people ages 15 and above). Source: World Bank, 2002.

Urban population Urban population (percentage of total). Source: WDI. Savings and giro Number of post office savings and giro accounts per 1,000 people. Source:

constructed from data in UPU. Money orders Number of money orders, domestic service per 1000 people. Source:

constructed from data in UPU. Letters/staff Letters/staff. Source: constructed from data in UPU. Receipts/Expenditure Postal receipts/Postal operating expenditure. Source: constructed from data

in UPU. KM/Office Average area covered by a permanent post office (km2). Source: UPU. % of Offices staffed private Secondary offices staffed by nonadministration as a percentage of total

offices. Source: constructed from data in UPU. Expenses/Letter Postal operating expenses/letter (SDRs). Source: constructed from data in

UPU. Receipts/Staff Postal receipts/staff (SDRs). Source: constructed from data in UPU. Offices/Population Post offices/population (1,000). Source: constructed from data in UPU. Staff/Office Total postal staff/post offices. Source: constructed from data in UPU. Population unserved Percentage of the population without postal delivery, 2000. Source: UPU. GDP/Capita GDP/Capita, PPP. Source: WDI.

20


percent and the average geographic coverage ofpost offices varies from less than one square ki-lometer to nearly 30,000 km2.

Table 3 reports regional average statistics, sug-gesting considerable variation even at this aggre-gate level. Letters per capita at this level varybetween nine and 224, expenses per letter from$0.47 to $1.92, and the trust score from three tosix. Much of this variation would be predicted

based purely on income differences between re-gions, as we have seen. The remainder of thispaper discusses if other factors are also at work.

A first stage in the empirical analysis is to lookat simple correlations between variables. To dothis, we construct a relative measure of postalperformance based on the “postal Jipp curve.”This is the (log) actual letters per capita postedin a country minus the (log) predicted number of

Table 2. Descriptive statistics

Table 3. Regional statistics of selected variables

Number of Observations

Average Standard Deviation

Minimum Maximum

Postal index 110 0.00 1.26 -3.39 3.22GIT Postal trust 51 4.36 1.50 1.80 6.70Illiteracy 87 20.97 20.93 0.20 84.07Urban population 110 54.30 22.68 9 100Savings and giro 41 177 286 0.20 1574Money orders 88 347 1096 0.02 7322Letters/Staff 105 31,830 38532 196 174930Receipts/Expenditure 103 1.07 0.50 0.11 4.73KM/Office 106 1,491 4358 0.80 29182% of Offices staffed private 106 29 30 0.00 95Expenses/Letter 103 2.07 13.96 0.001 142Receipts/Staff 101 13,288 21900 58.84 132579Offices/Population 106 0.16 0.18 0.00 1.32Staff/Office 101 7.73 7.60 0.54 57.44Population Unserved 94 7.74 18.89 0.00 97.00GDP/Capita 110 8,389 8720 523 50061

GDP/capita (PPP

value) ($) Letters / capita

Expenses / Letter

($) Trust Population Served

(%) Asia Pacific 6,106 28.0 0.474 4.38 97 Europe and CIS/ECA 8,030 38.8 0.672 4.70 99.7 LAC 7,017 17.5 0.702 3.03 93 Arab/MNA 4,275 8.3 1.92 4.70 85.7 Sub-Saharan Africa 2,750 9.1 0.5371 3.47 76.8 Industrialized countries 25,191 224.0 0.57 6.02 99.5

21

letters posted in a country based on what wouldbe expected given the country’s income. Broadly,a positive number suggests that the country isposting more letters than we would expect basedon income, a negative number means people areusing the (official) postal system less than ex-pected (for example, the 2000 score for Algeriais 0.99, Morocco 0.87, Trinidad.24, Tanzania1.82 and for Senegal –0.11).4

We can now study a correlation matrix based onthis data. The numbers in the matrix (Table 4)suggest the degree to which one indicator variesin step with another. If there is no relationshipbetween changes in one indicator and the other,the score in the matrix would be zero. If an in-crease or decrease across observations in one in-

dicator is perfectly matched with an increase ordecrease in the other indicator, the score in thematrix would be one. If an increase in one indi-cator is perfectly matched by a decrease in theother (and vice-versa), the score in the matrixwould be –1. Strong positive or negative corre-lation scores (greater than 0.5 or less than –0.5)are highlighted. The next 10 paragraphs discussthe results presented in Table 4.

We can use the postal index score (letters percapita performance allowing for income) for 2000to see what correlates with “better than expected”postal performance. Looking across the first rowof the correlation matrix, it is possible to see howclosely our index of postal performance is re-lated to the other indicators in our database. In

Table 4. Correlation matrix


GIT

Postal Trust

Illiteracy

Urban Population

Savings and Giro

Money orders

Letters/staff

Receipts/E

xpenditure

KM

/Office

% O

ffices Staffed Private

Expenses/L

etter

Receipts/Staff

Offices/Population

Staff/Office

Population Unserved

GD

P/C

apita

Postal index 0.55 0.13 -0.19 0.00 0.17 0.35 -0.01 -0.10 0.18 -0.30 -0.10 0.10 0.13 0.07 0.08GIT Postal trust -0.22 0.36 0.51 0.32 0.72 0.34 -0.33 -0.01 -0.24 0.65 0.48 0.46 -0.43 0.81

Illiteracy -0.60 -0.39 -0.21 -0.28 0.06 0.42 0.02 0.15 -0.33 -0.43 -0.12 0.43 -0.56Urban population 0.36 0.14 0.43 -0.07 -0.26 0.02 -0.09 0.40 0.21 0.37 -0.41 0.62Savings and giro 0.10 0.50 0.05 -0.16 -0.22 -0.03 0.83 0.29 0.19 -0.22 0.53Money orders 0.09 0.05 -0.10 -0.16 -0.04 0.04 0.17 0.20 -0.11 0.24Letters/Staff -0.02 -0.12 0.15 -0.10 0.56 0.03 0.46 -0.15 0.66Receipts/Expenditure -0.02 -0.01 0.05 0.04 0.11 -0.03 -0.04 -0.02KM/Office -0.16 0.60 0.13 -0.25 0.09 0.08 -0.22% of Offices staffed private -0.07 -0.01 0.21 -0.31 0.04 -0.02Expenses/Letter 0.50 -0.10 0.14 -0.02 -0.08Receipts/Staff 0.03 0.39 -0.19 0.60Offices/Population -0.11 -0.27 0.37Staff/Office -0.12 0.50Population unserved -0.30

22


short, the postal index appears to be strongly andpositively correlated with our measure of trust inthe postal system and the number of letters peremployee, and strongly negatively correlated withtotal postal expenditure divided by the numberof letters delivered.5 This suggests that trust inservices and efficiency in their delivery are bothimportant to increased postal volumes.

What do these simple correlations suggest aboutpossible explanatory factors driving postal per-formance? First, it suggests that demand doesnot appear to be simply correlated with literacyrates. Countries with literacy rates below 80percent have an average postal performance of0.24 –higher than those countries with a literacyrate of above 80 percent—at -0.20. Income ap-pears a better determinant of postal demand thanliteracy rates, then—perhaps in part because inmany developing countries many literates do notuse postal services, suggested by letters per capitavolumes of under one piece per capita per year.

On the supply side, and related to the above, itdoes not appear that the extent of the network (asmeasured by the percentage of the populationcovered or post offices per square kilometer), thelevel of urbanization or the presence of ancillaryservices is correlated with increased letter vol-ume per capita at a given level of income. Theseresults may suggest limited demand for a univer-sal letter service in developing countries, giventhat countries where more people can (theoreti-cally) post and receive letters do not see largerletter volumes as a result. This is an idea ex-plored further in Kenny (2004).6

The lack of a relationship between postal receiptsdivided by postal expenditure and performancemay further suggest that government subsidiesjustified on the grounds of extending use of thepostal network are not working. If subsidies were

extending access, countries where revenues wereconsiderably smaller than expenditure (and socountries where subsidies were large) would seea larger number of letters per capita, a lower ra-tio of square kilometers to offices, a higher num-ber of offices per capita, or a lower unservedpopulation. None of these results hold.7 Subsi-dies are apparently not extending access to oruse of the postal network (this could also suggestthat demand for letters sent in developing coun-tries is fairly independent of price, perhaps fur-ther emphasizing the importance of efficiency andreliability in increasing volumes).

There does not appear to be a particularly stronglink between the level of postal financial servicesused by the population and basic postal perfor-mance. Given that the postal financial networkmight be considered a good in its own right, thissuggests that a separate analysis of what leads toa strong network may be worth undertaking.Some suggestive correlations do appear, however.Savings and giro accounts per capita is stronglycorrelated with receipts per employee (unlike thepostal index itself), suggesting that the postal fi-nancial network may play an important role increating income for the postal network and re-flecting, perhaps, the greater importance to manypost office customers of financial services pro-vided through the postal network than letter ser-vices.

At the same time, the results strongly suggest thatefficiency improvements may be an importantelement of the story. Both postal expenses perletter and letters delivered per postal staff mem-ber are strongly correlated with postal perfor-mance—lower costs and fewer staff per letter arecorrelated with more letters sent at a given in-come (see Figure 2). As noted earlier, it is prob-able that causation runs in both directions. Highfixed-costs mean that factors that increase vol-

23

umes will improve efficiency, while improvedefficiency and quality will also increase the de-mand for mail services. Trust in the postal net-work is also a significant correlate with postalperformance—countries with low trust in thepostal system (a score of below 4 out of 7) aver-age postal performance of -0.59, compared tocountries with high trust (above 4) that average0.51.

Efficiency related indicators include letters de-livered per staff, ratio of revenues to expendi-tures and postal revenues per staff. Since theseindicators are correlated with each other, we canconstruct an index for efficiency as the principalcomponent of these four indicators. The regionalaverages of the index are illustrated in Figure 3.Industrialized countries lead at an index of 2.0.Asia Pacific region’s index of 0.448 is a step aheadother developing countries. East European andCentral Asia scores the last among all regions.9

This suggests the unsurprising result that effi-ciency is related in part to GDP per capita, butalso suggests other factors are at work—sub-Saharan Africa does better on this index thanwould be expected given its income, Eastern Eu-rope and Central Asia, as well as the Middle Eastand North Africa do worse than would be ex-pected.

Like efficiency, postal trust is correlated withincome, suggesting that it may be difficult fordeveloping countries to greatly increase trust inthe postal network, perhaps through institutionalreform. For those countries that do manage toimprove trust, however, the payoff may be high.One possible avenue suggested by the strong posi-tive correlation between letters/staff and trust, isto reduce over-staffing in the postal network. Itis worth noting that it appears efficiency and highquality of service are positively related. Coun-tries with low postal efficiency in the postal sys-

Figure 2. Postal performance by expenses per letter

y = -1.2492x + 0.8064R2 = 0.4544

-4

-3

-2

-1

0

1

2

3

4

0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50

Expenses/Letter(tw o outliers removed)

Po

stal

Per

form

ance


24


tem (an index of below 0) average a trust scoreof 3.62, compared to countries with high effi-ciency index (above 0) which average 4.97. Inother words, at least at first examination, theredoes not appear to be a necessary efficiency-qual-ity trade-off—instead, it is possible that greaterefficiency can drive improvements in both vari-ables.10

Turning to factors that might lie behind the effi-ciency result, expenses per letter seem to beuncorrelated with income per capita, despite thetheoretical argument that would suggest a strongrelationship based on scale economies in deliv-ery (driven by factors such as the fixed costs ofdelivery). The lack of such a relationship mayreflect the fact that developing country post of-fices provide fewer ancillary services, while hav-ing lower labor costs and a far lower level ofquality of service in developing countries. None-theless, it also reflects the possibility of increas-ing efficiency in poor countries.

Regarding the impact of reform on efficiency andletters per capita, countries where no secondary

Figure 3. Postal efficiency index by region

-1 -0.5 0 0.5 1 1.5 2 2.5

Asia Pacific

Europe and CIS/ECA

Latin America and Caribbean

Arab/MNA

Sub-Saharan Africa

Industrialized countries

Postal Efficiency Index

post office is staffed by people outside the postaladministration score an average of -0.43 onthe performance index, while those with at leastone such office score an average of 0.17. Hav-ing said that, the link appears statistically weak.

There are at least two possible reasons for theseresults:

n There are significant data weaknesses. Themeasure of “percentage of offices that areprivately staffed” is also only a crude andpartial indicator of sector reform. Further-more, countries that have a liberalized mar-ket are likely to be the ones where statisticson mail volumes provided to the UPU areweakest, with a likely bias towards under-reporting.11

n It may be that the indicator does not cap-ture what is important as part of a success-ful reform agenda (“Postal Policy andRegulatory Reform in Developing Coun-tries”). The case studies in Walsh (2001)covering Argentina, Tanzania, and Trinidad

25

and Tobago suggest that privatization with-out successful regulation (Argentina) canproduce worse results than corporatizationand various types of management contract(Tanzania and Trinidad and Tobago). Animportant factor behind increasing the let-ter flow of the incumbent captured in UPUstatistics may be better management at thetop of the incumbent itself—ill captured byour measure of reform.

The Annex contains results of a series of regres-sion analyses that confirm the above results—efficiency and quality appear to be important de-terminants of performance, but the link betweenfranchised offices and efficiency appears weak.

Conclusion

The results in this paper, it should be repeated,are very preliminary. They measure postal per-formance in a narrow way, excluding, for ex-ample, the important role that post offices playin the delivery of government and financial ser-vices. Further, they relate to single-period corre-lations, and such correlations can say nothingabout causation. The most the results can sug-gest is that the data does not disprove a giventheory. Nonetheless, they are suggestive. It ap-pears that it is possible for developing countriesto improve the performance of their postal sec-tors through efficiency reforms, especially thosethat increase trust that letters will be deliveredintact and on time. Such reforms may be able tostem postal sector losses while at the same timedelivering positive improvements in the qualityof service. The cross-country data on reform isnot fine-grained enough, to allow for a seriousstatistical analysis of what reforms work and howto improve efficiency. The tentative conclusionsthat can be drawn from the evidence we have (to

date) cannot move far beyond the level that itmay not be as simple as franchising offices.

However, we know that successful reform pro-grams can improve the indicators that make upour efficiency index. Both Trinidad and Tobagoand Tanzania’s reform programs saw letters de-livered per postal employee approximately double(estimated from Walsh, 2001). Careful case studyanalysis of reform, of the type summarized inWalsh (2001) should drive policy recommenda-tions in this area.

Given the preliminary nature of the results pre-sented here, there is clearly scope for consider-ably more research in this area. A first step forsuch research would involve the collection of moredata (where the UPU and its members should playa lead role). In particular, it would be useful tohave data covering mail volumes in the wholesector (not just volumes delivered by the incum-bent), prices for mail delivery services, and de-tails on the status of sector reform in a wide cross-section of countries. A second step would be amore advanced analysis of the data based on re-gression analysis of models of postal demand.Better data and more advanced analysis mightallow for the evaluation of the impact of variousreform models on volumes, quality of service andprice.

Annex: Regression Analysis ofEfficiency, Trust, and ImprovingPerformance

In order to test the determinants of the postal ef-ficiency and its impact on the reliability and per-formance of postal services, we estimate a simplelinear regression model on the efficiency indexand a two-stage least square model for the latter,as summarized in the following equations:


26


As for factors that might lie behind the efficiencyindex, regression analysis of Equation (1) sug-gests that income levels play a dominant role.Literacy (perhaps because of its link to the po-tential postal market and the capacity of the postalstaff), also has a statistically significant impactalthough the magnitude is rather small. Echoingthe correlation results above, however, the resultssuggest no link between policy reform as mea-sured by the percentage of offices that are pri-vately staffed12 and postal efficiency.

As would be expected, the results from Equation(2) suggest that trust in the postal network isstrongly correlated with postal efficiency. A 10percent efficiency improvement is associated withan eight percent increase in the trust index.

Variable Coefficient Student’s T Log GDP/capita 0.851 3.79 Urban population (%) 0.003 0.35 Private offices -0.003 -0.66 Literacy 0.016 1.84 Constant -7.575 -4.43

Variable Coefficient Student’s T Efficiency Index 0.485 5.11 Percent of population served 0.045 2.46 Constant -0.042 -0.02

Variable Coefficient Student’s T Log GDP/Capita 1.956 6.62 Efficiency Index 0.340 1.90 Predicted trust 0.911 0.48 Literacy -0.001 -0.09 Urban population (% of total) -0.027 -2.64 Constant -13.418 -5.51

Regressions with Efficiency as Dependent Variable

Regression with Trust as the Dependent Variable

Regression with Log Letters/Capita as the Dependent Variable

The analysis of the correlates with letters percapita as the dependent variable in Equation (6)suggests that efficiency is strongly related to thismeasure of postal performance. The results againsuggest that there is insignificant correlation be-tween possible environmental determinants ofperformance (urbanization and literacy) and thepostal performance measure. Performance on thepostal index does not appear to be determined bythese factors that are difficult for developingcountries to change over the short-term, then.This suggests the potential for reform efforts tohave an impact on performance.

27

Box 1. Formulas


)()()()( 44321 iiiii LITERACYPRIVATURBPOPGDPEFFINDEX ααααα ++++= (1)

)()( 321 iii POPSERVEEFFINDEXTRUST βββ ++= (2)

)(6)(5

)(4)(3)(21URBANPOPiLITERACY

predictedTRUSTEFFINDEXiGDPiLETCAP

γγ

γγγγ

++

+++=

(3) )(3

ˆ)(2ˆ

1ˆ

iPOPSERVEiEFFINDEXpredictedTRUST βββ ++=

(4)

(1)

(2)

(3)

(4)

countries with weak operators, much mail is deliv-ered by unofficial operators—which is not capturedin the statistics. The trust index could reflect trustin express or highly guaranteed mail services ratherthan the country’s core mail service.

4 As a first, very preliminary, step in overcoming thelack of evidence regarding postal performance andeconomic growth, using 1980 data to construct theindex, and data on economic growth rates over the1980-2000 period, we can see if countries that havea positive score on the postal index see faster subse-quent growth. Looking at the results suggests thatthere might be such a relationship. Overall, 72 per-cent of the 79 countries suggest the correct relation-ship (positive to positive or negative to negative)between the postal index score and growth outcome.Clearly, it is likely that a positive score on the postalindex correlates with a number of other factors (per-haps general institutional quality, for example) whichare also factors behind growth. However, the re-sults are suggestive of the role that the postal systemmay have to play in the economy and development.

5 It is important to remember that this is a measureof postal performance given income—if the measureused was letter per capita, it would be strongly cor-

Notes1 Senior Economist, the World Bank. Particularthanks go to Christine Qiang for both comments andfor providing the regression analysis presented inthe annex.

2 Letters/capita is clearly not a directly equivalentmeasure of telephones per capita—it is more of ameasure of use (equivalent, perhaps, to minutes ofcall time/capita) than of infrastructure rollout. An-other measure might be post offices/capita. However,as post offices are public, whereas the great majorityof telephones are private, this is not a good com-parator, either. For example, Hong Kong could prob-ably get away with very few really large post officesto serve the entire city, obviously it couldn’t do thesame with telephones. The most direct equivalentmay be home letterboxes per capita, but that doesnot appear to be available in the UPU database.

3 It should be pointed out that much of the data issubject to significant error. For example, the UPUcollects its data from the incumbent postal operatorand it has limited capacity to check the reliability ofthis information. Further, while such operators usu-ally have an official monopoly over at least parts ofthe postal delivery system, especially in developing

28


related with any indicator that was also related toincome per capita (including, for example, urban-ization). It should also be noted that letter/capitaand expenditures/letter are all arithmetically linkedby the inclusion of letter volume.

6 It is worth noting that countries with urbanizationrates below 50 percent have an average postal per-formance of 0.29—higher than those countries withan urbanization rate of above 50 percent—at -0.20.This may be an artifact of a nonlinear relationshipbetween income and letters per capita. Alternatively,it may reflect a problem with the data, related to thecomparative ease of competitive service delivery inurban settings. Developing countries with large ur-ban populations are seeing a greater level of legal orillegal competition to the incumbent, and this com-petition is delivering letters that are not reported inthe UPU statistics.

7 The weak correlation between urbanization andreceipts/expenditure suggests that it is not the casethat subsidies are going to deliver the same level ofservice in countries where delivery is more complex.

8 This is probably because Hong Kong is grouped inEAP, not in the industrialized countries.

9 This efficiency index has a sample mean normal-ized to 0, sample standard deviation of 1.429. Itranges from a minimum of –1.031 (Tajikistan) to amaximum of 8.029 (Hong Kong, China).

10 Alex Dieke (in correspondence with the author)notes a similar phenomenon occurred in Europe inthe 1990s, where TPG and Deutsche Post both sawimproved routing times and reduced costs and em-ployment.

11 Interestingly, it is positively correlated with thenumber of staff per office. Use of another measureof “reform”—countries with a World Bank projectlisted in the Bank’s postal projects database whichbegan before 2000, that had a policy/regulatory re-form component—suggests that project countries see

slightly above average performance on the perfor-mance index. However, the result is not statisticallysignificant and is based on only eight countries thatfit the criteria (Gabon, Guatemala, Guinea Bissau,Morocco, Tanzania, Togo, and Trinidad and Tobago).

12 The World Bank liberalization index was not usedin regression analysis because it was created basedon a small and unrepresentative sample.

29

Chapter 4

Questioning the Monopoly-SupportedPostal USO in Developing Countries

The monopoly-supported universal serviceobligation (USO) is usually defended onthe grounds that the monopoly allows for

cross-subsidy in letter services that in turn al-lows universal access to a service of great im-portance to all. This paper argues that letter de-livery (as opposed to other services that may beprovided by post offices) is not in universal de-mand in poor countries, that the size of the mar-ket in developing countries is such that USOscould not be met under the monopoly model, andthat the monopoly carries heavy costs for sectordevelopment and consumer welfare. It proposesin the place of the postal USO a competitive ap-proach involving universal access to a range ofservices that poor people have a need to access.

The Theory of the Monopoly-supported USO

The European Union (EU) defines universal ser-vice to post in the following terms:

Charles Kenny

“Member States shall ensure.. the perma-nent provision of a postal service of a speci-fied quality at all points in their territory ataffordable prices for all users… the univer-sal service provider(s) guarantee(s) everyworking day… one delivery to the home orpremises of every legal or naturalperson…[for] postal items up to two kilo-grams [and] postal packages up to 10 kilo-grams…”

Traditionally this USO has been linked with mo-nopoly provision of services. The theory of themonopoly-provided USO can be summed up asfollows. Mail delivery to the home is an impor-tant public service to which all should have ac-cess. For many low-volume or difficult to ser-vice users, however, the market price of mail ser-vices would be prohibitively high. In order toserve those customers while allowing for a fi-nancially sustainable postal network, a monopolyoperator charges a single stamp rate for nation-wide delivery, with “profits” made from the de-livery of mail to high-volume areas and users

QUESTIONING THE MONOPOLY-SUPPORTED POSTAL USO IN DEVELOPING COUNTRIES

30


(where the price of the stamp is higher than thecost of delivery) used to support “loss-making”delivery operations for low-volume areas andusers.

Many developing countries have adopted themonopoly USO model for their own postal sys-tems. They do not tend to have USOs as strin-gent as the EU’s, while in many cases, the obli-gation is rather imprecise or undefined, it rarelyinvolves home delivery, for example.1 Nonethe-less the general model holds that a provider willensure one-price letter delivery across its nationalterritory, funding loss-making routes with therevenues that it is guaranteed from a monopoly,over delivery on profitable routes.

The model has proven itself practicable (if notnecessarily efficient) in many wealthy countries.However, its application in developing countriesis likely to be inefficient, impractical, and ineq-uitable. This paper argues that a narrow, mail-based USO should be abandoned and the postalsector opened to competition. In place of a mail-only USO, governments in developing countriesmight decide to extend access to a bundled rangeof services widely in demand by urban, rural,poor, and rich alike. These services might bedelivered through post office franchises.

Do We Want a USO for LetterDelivery in DevelopingCountries?

Does everyone in the developing world need let-ter delivery in the same way as one might arguethe need for primary education or health care? Itis perhaps worth asking this question in the con-text of severe resource constraints that mean thegoals of universal primary education or univer-sal access to basic health care are far from being

reached. More narrowly, similar language ofuniversality is used in other infrastructure sec-tors including electricity. There (too) the lan-guage conflicts with the reality—in part becauseof inefficiencies and inequities of provision, butalso because there are only so many resources,and a great many needs for those resources tomeet.

Is there evidence that everyone in developingcountries needs (or would use) letter delivery toa nearby post office every business day or everyweek? To begin to answer that question it is worthlooking at demand for letter services in poor coun-tries. Overall demand for postal services is veryclosely correlated with the size of an economy.Smaller (poorer) economies see lower mail vol-umes (letters per capita). Countries with a GDPper capita (PPP) of below $1,000 see mail vol-umes of below 1 per person per year—comparedwith closer to 100 per person in countries with aGDP per capita of above $5,000 (Table 1).

Constraints on demand due to inadequate or ab-sent service delivery will be one factor behindlow levels of use, but by far the dominant causeis demand related to income—as suggested bythe results of both postal reform and extendingaccess. Under reform, service quality improves,and this tends to increase use of the postal sector.But the usage increase, while impressive, is inthe region of perhaps 100 percent, rather thanthe 10,000 percent differences between low in-come and high income country usage rates. Forexample, successful postal reform in Tanzaniasaw mail volumes increase from 0.87 to 1.26 let-ters per capita per year during 1994-98. Thispost-reform figure remains only approximately0.17 percent of US per capita mail volumes.There are considerable limits on the demand formail services in poor countries even where themail service provided is of a reasonable quality.

31

Regarding extended access, there does not seemto be a relationship between the percentage ofthe population covered by postal service and let-ter volumes per capita in developing countries—extending network access, like improving networkquality, does not apparently uncover significantunmet demand for postal services (Kenny andQiang, 2004).

The average consumer in poorer countries sendsperhaps one letter per year—and we know thatthe great majority will neither send nor receive asingle letter.2 In particular, there is little demandfrom businesses (which account for as much as80 percent of letters sent [Lee, 2004]) to delivermail to the poorest in developing countries. Thisis because most do not have services that re-quire billing (telephony or electricity, as it mightbe), and they are hardly an attractive target forcompanies in terms of marketing purposes.

If the majority of the population never receives aletter in the poorest countries, and might not evenafter significant postal reform, the concept ofuniversal service is somewhat nonsensical.3 Andfor the great majority of the over one billionpeople worldwide who live on less than a dollara day, it is not clear how they would write a letteror what they would do with one even if they gotit—they are illiterate (Kenny, 2002).

Overall, there does not appear to be significantdemand for universal letter service even if it were

to be a financially viable option. If universalitywas achieved using some sort of subsidy mecha-nism, given the direction of most mail, this wouldconstitute a regressive intervention. Given thatthe beneficiaries would be primarily large busi-nesses and wealthier individuals, general govern-ment revenues would be used to support thesewealthy few with little benefit to the poor andilliterate.

On the other hand, post offices can deliver farmore than mail. In many developing countries,postal payments represent over 50 percent of to-tal postal revenues, and post offices remain theprincipal point of access to financial services forthe majority of the working population (Walsh,2001). Many post offices also act as the inter-face for government-to-citizen interaction suchas license payments. Adding together this bundleof services that post offices frequently deliver,governments may well decide that it is vital forthe great majority or all of citizens to be able toreach the service delivery point.

It should be noted that we are some way from thegoal of universal (or even widespread) use of apost office. First, many people in poor countriesdo not even live within usable distance of a postoffice, as is suggested by the fact that, for coun-tries with a GDP per capita of under $1,000, thereis only one post office for each 4,700 square ki-lometers of territory as compared to one officeper 458 square kilometers for countries with a

Table 1. Postal structure and development


Indicator Average for GDP Per Capita (PPP) Band <$1,000 $1,000-$5,000 >$5,000 Letters/Capita 0.9 4.6 98.4 Total letter volume 10m 634m 2,468m KM2 per post office 4,702 1,738 458 Source: Calculated from UPU and World Bank data, 2001

32


GDP per capita of above $5,000. Second, it ap-pears that only a very small percentage of thatpopulation already in walking distance of a postoffice currently use it even in reformed develop-ing country markets. In Tanzania, for example,an average of about 0.4 percent of the popula-tion within the catchment area of a post officevisit that office each week (PWC, 2004).4

Regardless, if the goal is universal access to awhole range of services out of which delivery ofletter mail is an unimportant component for mostcitizens, again, a strictly postal USO makes littlesense. And if we are to look at the USO as abroad commitment to support a whole range ofgovernment-to-citizen interactions, perhaps alongwith financial and other services, the traditionalindustrialized country model of financing theUSO seems inappropriate even if it were work-able in developing countries. Can we justify afunding mechanism that forces the urban letter-stamp purchaser to subsidize the transactionscosts of a rural applicant for a car license as partof a system that frequently provides substandardletter delivery?

Could a Letter USO be Supportedin Developing Countries underthe Monopoly Model?

Even if developing countries desire a monopoly-supported USO model, it appears unlikely to besustainable in many countries. This section ar-gues that developing countries, with small postalmarkets, cannot benefit from significant scaleeconomies (which in turn drives up the per-unitcost of letter delivery) and face far more compli-cated environments in which to deliver mail.These factors combined mean that even a limitedUSO model supported by monopoly rights wouldbe unsustainable in many low-income countries.

In order for the USO to be sustainable throughcross-subsidy from profitable routes to unprofit-able ones, there have to be enough profitableroutes (and few enough unprofitable ones) to al-low for universal service at a reasonable price.The evidence suggests, however, that average perpiece costs in developing countries are very high,both due to scale effects but also a large propor-tion of expensive-to-service rural customers. Thissuggests that there is not the amount of profit-able service at an acceptable price to support theletter USO.

By and large (and with obvious exceptions suchas China and India), absolute mail volumes inpoorer countries are very small. Total mail vol-umes in countries under $1,000 GDP per capitaaverage 10 million pieces, compared to 2.5 bil-lion pieces in countries with a GDP per capita ofover $5,000 (Table 1).

This is a problem because it is clear that thereare fixed-costs associated with mail systems thatmake low-volume environments high-cost onesas well. Scale economies, due to factors such asautomated processing and (more significantly)scale effects in delivery, will be one factor be-hind a relationship between total mail volumesand letters delivered per employee. Countries thatsee delivery of fewer than one million letters ayear deliver on average below 4,000 letters peremployee per year (such countries include CapeVerde, with just over 939,000 letters deliveredby a staff of 3,947). This compares to over60,000 per employee in countries where over 100million letters are delivered a year (South Africadelivers nearly 2 billion letters using a staff of74,000, for example). This is a 15-fold produc-tivity difference (Table 2).

We can see the impact of scale economies by look-ing at estimates based on the United States Postal

33

Service (USPS) on the costs of a US-style ser-vice in lower volume environments with lowerinput costs. The USPS will differ dramaticallyin terms of automation, route topography, mix ofmail, efficiency, and the nature of its customerbase—and so the model is only an approxima-tion (made less accurate by a weak proxy forinput costs—that of GDP per capita). WhileCohen et. al. show that it functions reasonablywell in estimating features of rich country postaloperations, it is very likely that the adjusted modelwill become increasingly inaccurate as a guidewhen looking at low-income postal markets(Cohen et. al. 2003a).

It is also worth noting that UPU data suggests afairly weak relationship between GDP per capitaand total costs per letter (Kenny and Qiang,2003). This is somewhat surprising given evi-dence that poor countries in general see low vol-umes and so will benefit less from economies ofscale. The result will be due at least in part tothe facts that labor costs are lower in developingcountries and that most developing countries of-fer a lower quality, extent, and scope of services(not delivery to the door, not six days a week, notfast and reliable, and absent ancillary servicessuch as pension payments).

With those caveats, Figure 1 shows what theCohen study and World Bank data suggests thecost of delivering a letter under United States’USO standards would be in countries with lowermail volumes. As the number of pieces per capita

drop below 100 per year, costs rise dramaticallyabove 50 cents per piece.5 This may help to ex-plain why the United States only mandated uni-versal home delivery in 1958, when letter vol-umes were above 300 per person per year(Campbell, 2004).

Table 3 suggests the breakdown of costs of US-quality delivery per piece of mail in a system ofuniversal service in an economy where postalvolumes per capita were low, based on data fromthe Cohen et. al. study and the World Bank. Asseen, delivery would be expensive—over $0.76per letter in a country with mail volumes of 10per capita. Even without delivery to the door,the costs are high—over $0.60 per letter.6 Again,this data only suggests orders of magnitude, as itis based on United States’ postal data adjustingfor input costs—and the USPS is an operation ina very different (broadly more positive) environ-

Table 1. Cross-country evidence of scale economies in posts Average for Total Letters Delivered Band

<1,000,000 1,000,000 – 100,000,000

>100,000,000

Letters/Employee 3,587 20,693 59,739 Source: Calculated from UPU and World Bank data, 2001

0

0.5

1

1.5

2

2.5

3

0 50 100 150 200 250 300 350 400 450 500

pieces per capita

cost

per

pie

ce

Figure 1. Estimated relationshipbetween costs per letter and volumes


34


ment from most developing countries. Nonethe-less, the point is clear that letter delivery becomesconsiderably more expensive in low-volume en-vironments, not merely because of rising costs ofhome delivery, but also because the other fixedcosts of the network are divided up amongst farfewer stamp purchases.

The reality for a great majority of poor countriesis that total mail volumes will be small enough tosuggest severe scale diseconomies—mail deliv-ery will be comparatively very expensive to pro-vide to all, let alone the difficult-to-serve.

Making this problem more acute is the fact thata far greater percentage of people in developingcountries are difficult-to-serve rural customers—69 percent in low income countries (World Bank,2002). Income density—an important measureof “demand for postal services per square kilo-meter”—is far lower in developing countries thanrich ones—$39,000 per square kilometer in sub-Saharan Africa as compared to $658,000 in highincome countries, for example (Kenny, 2002).Both of these factors suggest that even less strin-gent USO may cost more to meet in poor coun-tries than they would in rich ones.

There is significant variation in all of these num-bers—some postal operations are more efficient

than others, and deliver more mail at the samelevel of GDP than others. It seems clear thatpostal performance (in terms of letters per capita)is correlated with greater operational efficiency(measured by letters per staff, for example) andtrust in the postal network (itself reflecting qual-ity of service) (Kenny and Qiang, 2003). At thesame time, some developing countries have verylarge, very dense postal markets where deliveryis far more straightforward. Nonetheless, thepoint remains that, in most cases, even less strin-gent USOs would be more expensive per letterthan similar obligations in the developed world.

At the same time, the profitable market thatmight provide cross-subsidy is far smaller, be-cause of the lower level of urban developmentand large corporate customers. Even in someof the richer, most population-dense, urbanizeddeveloping countries about 90 percent of routesare loss-making.7 This situation would be worsein countries where only 31 percent of the popu-lation were urban (the case in low income coun-tries). The cost and complexity of service de-livery, especially in small markets, will helpaccount for the fact that post offices frequentlymake significant losses in the developing world.Forty-five percent of countries with GDP percapita under $5,000 see revenues below operat-ing expenses (let alone total costs).8

Putting revenues and indicators of costs togethersuggests the scale of the problem. Low incomecountries will average total postal volumes lowerthan 10 million pieces a year. Imagine a nationalstamp price of $0.12.9 National stamp revenueswill bring in perhaps $1 million per year. This tosupport delivery to a population that is 69 per-cent rural and on average receives less than onepiece of mail per year.

Table 3. Estimated costbreakdown of USO servicesin low-volume environments

Cost of Mail/Piece ($) Pieces/Capita 10 5 1

Mail processing 0.06 0.07 0.16 Transportation 0.02 0.03 0.07 Window service 0.08 0.13 0.34 Delivery 0.76 1.19 3.06 Other 0.44 0.69 1.77 Total 1.36 2.11 5.40

35

The price of delivery and the related costs ofUSOs can be significantly reduced from the hy-pothetical levels of Table 3, if the USO is real-istically defined. In the United States, mail thatis nondelivered (i.e., is picked up at post offices)accounts for 21 percent of volume and 67 per-cent of profits. In Italy, similar numbers are 14and 42 percent. In other words, avoiding home-delivery dramatically reduces costs even in high-density environments such as in the United States(Cohen et. al. 2003a). Abandoning home deliv-ery for a post office box system (the de jurecase in many developing countries, the de factocase in the majority) will greatly reduce costs.This has been done in Senegal, for example,where home delivery carries an extra charge.

Asymmetric charging can also help raise rev-enues for rural access. Historically, differen-tial charges applied in nondeveloped markets—with the postal service in 17th century Englandcharging four times as much to deliver a letterfrom London to Scotland than within London(Ogilvie, 1893).10 Furthermore, “universal ser-vice” does not have to mean uniform service.In Tanzania, the quality of service is lower inrural areas, with service only provided to postoffices, and at D+4 rather than D+1. Again, inCosta Rica, the target is D+2 for urban and D+4for rural.

Nonetheless, even differential service targets—as part of the USO—are unlikely to reduce coststo allow for a sustainable universal service, to beprovided at a reasonable stamp price, on the backof the few profitable urban routes in low incomecountries. And so monopoly provision, even ifefficiently run, would not garner the profits largeenough to come close to true universal serviceprovision.

Against the Legal Monopoly

The two arguments for monopoly provision inthe postal sector are first, that the sector is a natu-ral monopoly and second, that the monopoly al-lows for cross-subsidy of services and so meet-ing USO targets.

The first argument is not a good argument, evenin theory, for enforcing a de jure monopoly. Thereare some reasons to believe that mail deliveryservices may be “natural monopolies”. Mail de-livery involves a network, and network externali-ties suggest that one big network serving a givenarea will, other things being equal, be more effi-cient than many overlapping networks providingthe same service. In other words, if you leave acompetitive postal market alone, it is likely thatone competitor will eventually win out to controlall, or nearly all, of the market.11 But this is nojustification for legally enforcing that market fromthe start. Quite the opposite. It is a reason toregulate the market so that a monopolist cannottake advantage of its position to squash competi-tion and over-charge consumers for services.

Furthermore, the “natural monopoly” character-istic of posts, especially in developing countries,may be oversold. We don’t need to go too farfrom basic posts to see that competition can workeven in “networked postal” industries in devel-oping countries—DHL, FedEx, and UPS dem-onstrate that competition can flourish where ob-ligations are few and prices can reflect costs. Andin a developing country environment where totalpostal volumes are frequently too small to gar-ner significant scale economies, the disadvantagesof market fragmentation are reduced. Some evi-dence for this is that in many developing coun-tries, a number of large postal users legally orillegally bypass postal incumbents to provide theirown services (in Jordan, legal competitive op-


36


erators provide services to deliver bills despiteserious handicapping, for example).12

The widespread bypassing of incumbent postalproviders in developing countries suggests that,whatever theoretical advantages to scale, the in-efficiency of monopoly government provision fre-quently outweighs them. Monopolies are likelyto be less efficient and less innovative than com-petitive environments because the usual incen-tive to innovation and efficiency (the fear of los-ing customers to the competition) is not there.When the monopoly in question is operated bythe government, the “theoretical” problem be-comes even more acute. There is not even theincentive to maximize the quality and efficiencyof services at a given cost in order to maximize(revenues and) profits. In an environment wherethe government monopoly service provider pro-vides an inefficient, poor quality service, any“natural monopoly” advantages it should theo-retically enjoy are frequently outweighed by thefact that post offices have not been run on soundbusiness practices.

Not only are staffing levels frequently very high,but pricing structures suggest that the post officemay even be losing money delivering to whatshould be their more profitable urban and corpo-rate customers. A monopoly will only producerevenues to fund the obligation if prices, some-where, are higher than costs. Given under-pric-ing and inefficiency of many government-con-trolled postal monopolies, this may be a ques-tionable assumption. Furthermore, service stan-dards remain low—as suggested by the low levelof trust in many postal operators. In response tothe question “do you trust your country’s postalsystem sufficiently to have a friend mail a smallpackage worth $100 to you?” the average surveyrespondent in Nigeria scored the post office 1.7on a scale of one (no trust at all) to seven (com-plete trust) (Kirkman et. al. 2002).

In these cases, not only is an enforced monopolyfailing to deliver on the promises of the USO,but by stifling competition, it is forcing peoplewho use the post to use an inefficient and unre-liable service provider. For little or no benefitto the rural poor, the monopoly penalizes thecorporate and urban user.

The practical example of the benefits of com-petition over monopoly in a poor developingcountry is Tanzania, where during a process ofliberalization, total mail volume increased from0.87 letters/capita/year to 1.26 between 1994and 1998 as we have seen, while the postal com-pany moved from loss to profit, and the numberof post offices increased (UPU and WB 2001).Prices did rise closer to costs, but consumerswere clearly willing to pay higher prices for ahigher quality of services.

The second argument for monopoly is that if itwere removed in a system of one-price nationalmail delivery, competitors unburdened by a USOwould serve those routes that cost less to ser-vice than the price of a stamp. Once competi-tors had “skimmed the cream” from these routes,the USO provider would be left serving only (ormainly) routes that were unprofitable at the cur-rent stamp price, leading to significant losses.

The cross-subsidy under monopoly model is aterribly inefficient way to support access tar-gets, however. The small number of large ur-ban customers posting mail in urban areas willhave to be heavily taxed (by paying stamp pricesfar above the cost to deliver their mail) to meetthe service requirements of the majority of ru-ral users. And theory suggests it is the worsttype of tax, in that its very specific—operatingon mail services only (theory suggest broadtaxes—like the income tax—are more efficientbecause they distort the market less). Heavily

37

taxing the large urban customer also encouragessubstitution, which depresses the volume of thevery traffic used to fund the USO. Furthermore,the sector looses the benefit of competition(greater efficiency, more innovation) in a mar-ket that accounts for some major part of thetraffic.

The monopoly is also a suboptimal way of “com-pensating” the operator because there is no nec-essary connection with the cost of providingthe USO. In rich countries, the value of themonopoly will be too much—in (rural, lowpopulation-density) New Zealand for example,the incumbent is providing USO services with-out monopoly and at no additional cost to thegovernment. As we have seen, in poor coun-tries, any potential profits from a monopoly onurban routes is likely to be far too little to meetany meaningful definition of universal service.A link between the value of the monopoly andthe cost of universal service can be made incountries where the monopoly is “worth toomuch” through regulation of pricing for mo-nopoly services. However, in countries with fewprofitable routes, it is doubtful that the monopolyprofit-maximizing price structure would gener-ate sufficient revenues to fund the USO, ren-dering pricing regulation irrelevant to meet thisgoal.13 The justification for the monopoly isnot practicable, the results of the monopoly arepoor-quality services and low use.14

It should be noted that the monopoly USO modelcan work. Trinidad and Tobago has increasedhousehold delivery to approximately 95 percentof population under the model, for example, aspart of a broader reform effort that saw postalvolumes and revenues approximately double,and quality of service and consumer satisfac-tion increase.15 But even in a wealthier devel-oping country, where the postal monopoly is

delivering a comparatively efficient cost-relatedservice and where the monopoly-funding-USOmodel may plausibly raise sufficient funds tosupport rural access, it is still far from the bestway of extending access.

A New Model

At the moment in many countries the “postalsector” is seen as coterminous with the post of-fice. It may be more constructive to view letterand parcel delivery as the “postal sector”. Someproportion of letters and parcels in developingcountries are delivered via the post office, whichis one service provider in the postal sector. Thepost office frequently also provides services inother sectors, including the financial and gov-ernment services sectors.

It is important to make these distinctions be-cause postal policy and regulation should beprimarily involved with improving the perfor-mance of the postal sector, not the post office.Conversely, the post office frequently has rolesin other sectors—it may be a vehicle for thedelivery of a range of government services, forexample. Ensuring the effective delivery of thiswider range of services is the concern of manyparts of government—the financial regulator forbanking services, for example, or the transportdepartment if the post office is involved in ve-hicle license issuance and delivery. Suggestingthat postal policy and regulation should be aboutimproved postal sector performance, and under-standing that the post office’s current and po-tential role spans many sectors, may help toclarify the roles and objectives for different partsof government and for the post office itself, al-lowing for a new model of both postal and gov-ernment services delivery.


38


Whatever is decided about the services providedto every citizen, in the poorest countries the mo-nopoly-supported USO for letter delivery is likelyto be unworkable and damaging to consumers ofpostal services. The first two steps on a path topostal sector reform should be to abandon themonopoly and seriously reconsider the postal USO.

What would be a sustainable model to replacethe monopoly, postal specific USO? There issome evidence in developing countries that in-creased private sector involvement in the incum-bent as part of a broader process of reform cov-ering issues such as staffing, tariffs, competition,and regulation may be of help. There should beno monopoly for any provision of postal services.However, either general competition laws orpostal-specific regulatory instruments shouldensure fair competition through (for example) theenforcement of termination agreements. Build-ing such capacity should be an early element ofthe reform process. Regulation of financial andother services provided would, naturally, fall un-der the purview of the suitable sector regulator.

Looking at universal service requirements, usingtaxes on operators may not be plausible in thepoorest countries—there aren’t likely to beenough postal and logistics profits to tax to meetthe costs of universal service schemes16—but itmay be a second-best solution for upper middle-income countries like Trinidad and Tobago, as-suming it does not discriminate against certaintypes of operators. But the issue of “universalservice” is one that should be tackled—in abroader context than the postal sector alone—and should almost certainly involve subsidiesbased on a broader tax-base.

In many countries, the most important social func-tion of the post office is not as a letter deliverymechanism, but in delivery of pensions, finan-

cial services, or other activities. Under those cir-cumstances, what users most want is “universalaccess” to those services, not necessarily to let-ter delivery. At the same time, there are cleareconomies of scope in siting services together inone location.

Governments provide or support a number ofservices that do, or it is felt should, reach thegreat majority of citizens—from primary educa-tion and basic health care to access to pollingplaces, registrations, and (perhaps) postal ser-vices. Some of these services can be provided“stand-alone,” others would benefit significantlyfrom economies of scope. A first step towards acoherent “universal service” strategy would beto decide on a list of services that governmentsbelieve should, and feasibly can, be provided tothe great majority of citizenry and then look foropportunities to “bundle” those services.

“Bundling” is the idea behind postal financialservices, the telecenter, and a number of otherinstitutions with a mixed track record. This isperhaps not the fault of the concept, but of itsimplementation—perhaps the wrong provider orthe wrong services provided. These failures sug-gest the importance of a community-driven exer-cise in choosing services and an entrepreneurial(sustainable) model for providing them.

How would we get from here to there? A five-step program might be:

(a) Abandon the monopoly. It doesn’t help, andit isn’t enforced.

(b) Carry out basic management and financialreform of the post office—restaffing, a man-agement contract, returning to financial sta-bility and setting up transparent accountingsystems. At the same time, reform and sepa-rate the postal financial system.

39

(c) Ensure the possibility of fair competition inmail delivery as part of a postal policy andlegal reform process This may be possibleby applying existing competition law andenforcement to postal operators or may re-quire specific regulatory institutional struc-tures to oversee the postal sector.17

(d) Develop a method to greatly increase pri-vate sector involvement in both the retailand delivery portions of current postal op-erations—perhaps including retail franchis-ing, and sale of the public operator’s net-work delivery functions.

(e) As part of a participatory process, agree thescope of bundled “universal” services to beoffered by postal retail outlets (starting at alevel that is minimal, implementable andsustainable), the number and approximategeographic location of the outlets needed tomeet USO goals, and a method to select thedelivery mechanism(s) for these services.

One possible model that would require consider-able development to be an implementable, prac-tical solution might involve a system where thereis a private franchise operated by a local entre-preneur that provides communications and ser-vices in every town or village with more than aminimum sustainable number of people. Servicesmay include basic finance, post office boxes (forthose that want them) and postal collection, li-censes, registrations and pensions. The govern-ment, as part of a participatory process, wouldset certain minimum services and standards thatmust be provided by the franchise. In addition,the franchisee would provide any other servicesit chose. The franchises (one per town or vil-lage) are auctioned to the highest bidder or, if noone will pay, a subsidy auction is used.

Most services apart from government-to-citizenbusiness are offered at prices that reflect costs.These prices would be regulated if the serviceprovided is an effective monopoly. A private com-pany provides (comparatively infrequent, regu-lar, sustainable) physical collection and deliveryto the franchises, picking up mail from a set num-ber of interconnection points. This company isalso selected competitively, and it receives a regu-lated (cost-based) amount per item delivered toor taken from the franchise plus (if required) anup-front subsidy. The company and franchiseswould have to provide letter delivery between postfranchises at one price nationwide. The com-pany is free to provide other postal services, buthas no monopoly on service delivery.

Again, it is important to make the point that thispaper has been based largely on generalities andaverages. There will be developing countrieswhere a monopoly-enforced cross-subsidy modelis providing services of value to the rural poorand where they are doing so at some reasonablelevel of efficiency. Even if they may benefit fromreform, the transition involved for such cases willhave to be carefully designed to ensure the con-tinuation of services. Furthermore, the record todate of reform is patchy, suggesting we have moreto learn about how to operationalize the competi-tive model.

Nonetheless, it is likely that moving from an un-workable monopoly model designed to deliver aservice in low demand, to a functioning competi-tive model that delivers a range of services thatpoor people find important and that might have adramatic impact on quality of life, at little cost inmany developing countries. Even taking just thefirst steps of abandoning the monopoly postalUSO would be significant progress.


40


As a final note, it should be pointed out that aban-doning a letter-post USO would pose a challengefor countries in their relations with the UPU. TheUPU Letter Manual states that “…in order tosupport the concept of the single postal terri-tory of the Union, member countries shall en-sure that all users/customers enjoy the right toa universal postal service involving the perma-nent provision of quality basic postal servicesat all points in their territory, at affordableprices.” (UPU, 2001). Perhaps it is fair to saythat this obligation was only introduced in the1990s and has since been met more in the spiritthan in the letter in many developing countries.

Notes

1 Even within the EU there remain questions as tothe nature of the USO—does it demand a uniformtariff, for example?

2 At least not through the network captured by Uni-versal Postal Union (UPU) statistics.

3 Regarding home delivery, many (most) poor andrural people do not live in households with a recog-nized address, so they could not receive mail if itwere being delivered.

4 In Malawi, the figure is closer to 0.2 percent.

5 Cohen et. al. Provide data on prices for US qualityUSO at United States’ input costs for various levelsof letter per capita. To adjust (at a high level ofapproximation) for differing input costs (primarilylabor), the author used a regression of GDP PPP percapita on letters/capita to obtain estimates for GDPper capita at various levels of letters/capita. I thenmultiplied the “unadjusted” Cohen estimates by (pre-dicted GDP)/(United States GDP) in order to getoutput cost estimates adjusted for input costs.

7 The procedure to adjust the Cohen et., al. estimateswas similar to the above except that GDP per capitaat a given level of letters per capita was calculatedfrom a regression of GDP on letters per capita forcountries with letters/capita values within 50 per-cent of those listed in the table (0.5 to 1.5 letters, 2.5to 7.5 letters and 5 to 15 letters per capita respec-tively).

7 Traffic is measured by revenues, in this case. Brit-ish Postal Consultancy Service (2002)

8 Raising prices to the levels suggested by the Cohenet. al. (2003a) study would be an unworkable solu-tion—this would doubtless reduce demand in favorof alternate providers, thereby further raising per-unit costs.

9 Trinidad and Tobago’s price after rate increases.

10 The United States only introduced one-price de-livery in 1885 (Campbell, 2004). It is worth notingthat in both the United Kingdom and the UnitedStates, the postal monopoly far pre-dates one-pricedelivery and universal service obligations, suggest-ing that the justificatory link between the three de-veloped ex-post.

11 This appears to be the case in a number of devel-oped markets where competition has been intro-duced—for example, New Zealand and Sweden.

12 Chapter 5 notes that many developing countrieswith weak or non-existent regulatory structures none-theless see more competition than developed coun-tries with aggressively pro-competitive regulatorssuch as the United Kingdom (where Royal Mail stillhas a 99.75 percent market share). It should be notedthat the presence of de facto competition, almost cer-tainly based on quality rather than price, suggeststhat an incumbent that improved the quality of ser-vice delivery in response to de jure competition mightactually see a growing market share even if it wereto raise its prices.

41

13 Furthermore, such efforts appear fraught with com-putational difficulties in practice and require far moremarket intelligence than most developing countryregulators (where there are such regulators) possess(see Chapter 5)

14 Again, the history of the British postal networkmay provide a suggestive example here—enforce-ment of the postal monopoly by Charles I was one ofmany grievances of Parliament that led to the En-glish Civil War (Ogilvie, 1893). In 16th century Brit-ain, the monopoly was used to “farm” profits,whereas today in many developing countries it hassome role as a source of government employment.In both cases, however, it is likely that the disadvan-tage to the consumer is greater than the advantageto the monopoly beneficiary.

15 Walsh, 2001 and comments by Juan Ianni. From1999 to 2003 delivery times fell from one week tonearly all mail by D+2, volumes rose 133 percent,revenues 75 percent, and a consumer satisfactionindex went from 50 percent to approximately 85percent.

16 Taking China as an example, a 2 percent tax onthe air express market would raise perhaps $30 mil-lion while costs of the USO can be conservativelyestimated at $340 million (Chapter 5).

17 Given that there appears to be de facto competi-tion for postal services in many developing coun-tries already, despite the hurdles presented by a dejure monopoly, the need for considerable regulatorycapacity may be limited.


42


Chapter 5

Postal Policy and RegulatoryReform in Developing Countries

Developing and transition countries cannot afford to ignore the competitive pressures that are transforming the postal

sector. Rather than taking an ad hoc approach topostal sector reform, developing countries needto implement a clear policy agenda that embracesliberalization and responds to customers’ needsfor improved service. Any policy reform strat-egy will have to address the complex question ofuniversal service. This paper demonstrates thatthe regulatory approach to universal service usedin developed countries is ill-suited to the low-vol-ume, low-demand environment of developingcountries.

A more realistic approach, based on universalaccess rather than universal service, echoes simi-lar developments in telecommunications policyand is warranted in a developing country con-text. Once the decision has been made to estab-lish a pro-competitive, liberalized postal regime,it is essential for developing countries to put inplace market-oriented regulatory rules, processes,and institutions by which the economic behavior

Boutheina GuermaziIsabelle Segni

of postal players (including the universal serviceprovider), is to be regulated. While there is nosingle recipe for regulatory reform, this paperpresents the argument that valuable lessons couldbe derived from the regulatory experience of de-veloped countries in the postal sector as well asfrom the experience of other network industries,notably telecommunications. These lessons needto be further tailored to better suit the peculiari-ties of developing countries’ low-volume, limiteddemand, and constrained finances of the postalenvironment.

Towards a Clear Policy Agendafor Postal Sector Reform

In most developing countries postal reforms havenot been implemented in the context of a clearpostal sector policy. Often decisions to reorga-nize the postal market and the incumbent are hast-ily taken to accommodate the timetable of thetelecommunication reform or to solve, momen-tarily, a financial crisis experienced by the postal

43

incumbent. Governments rarely proactively de-velop a long-term postal sector strategy, or thinkthrough the design of an effective regulatoryframework to achieve the postal sector’s reformobjectives. Formulating a clear policy agendafor postal sector reform is a critical task forpolicymakers in developing countries. This pa-per identifies two elements for sustainable policyreform in developing countries. First, the needto align the legal and regulatory framework withmarket realities; developing countries can nolonger ignore the phenomenon of de facto com-petition. Second, developing countries need torevisit their approach to universal service to makeit better suited to their postal market conditions.

Liberalizing the Market: Fromde facto to de jure Competition

The past decade has witnessed a fundamental shiftin public policies for network utilities from a long-standing natural monopoly approach towards anew paradigm embracing liberalization, open mar-kets, and the dismantling of public monopolies.Eperience has demonstrated that consumers havebenefited from open competition through improvedservice performance, lower prices, and expandedend-user choices—benefits not otherwise securedunder a monopoly environment.1 However, thissweeping paradigm shift has not effected all utili-ties equally. Postal services have lagged behindother utilities in responding to this evolution ineconomic thinking. In most countries the postalsector continues to rely on government ownershipand reserved areas designed to promote sociallyequitable access. The vestiges of the monopolysurvived even in countries that opted for liberal-ization as a method of sector reform. Unlike lib-eralization in other utility sectors, postal sectorpolicymakers in developed countries are continu-ously attempting to compromise between the ob-

jectives of introducing competition and disman-tling barriers to entry, and ensuring the financialstability of traditional monopolies entrusted withuniversal service mandate. In most cases, thiscomplex compromise has resulted in a marketstructure that is prone to distortions.

In developing countries, liberalization of thepostal delivery service has not followed a pre-defined, well-conceived policy reform agenda.Rather, it is developed as an ad hoc reaction torespond to the high cost and poor performanceof the public postal operator. Unsatisfied demandhas driven an unofficial liberalization process andgrowing de facto competition defying the mo-nopoly. In Latin America, incumbent postal op-erators have as little as 20 percent of the tradi-tional mail market due to their inefficiency andongoing failure to satisfy customer demands. Inother markets there is evidence of other forms ofsubstitution to ensure that mail services are pro-vided. Many utility companies have their ownmail distribution systems and smaller companieshave dedicated mail messengers to ensure maildelivery. Such alternative delivery mechanismsshow that if customers do not believe in the gov-ernment operated postal system, they will findways to ensure their mail is delivered. (See alsoChapter 3: Questioning the Monopoly-SupportedPostal USO in Developing Countries)

Given the failure of the public operator to pro-vide a better and more reliable postal service—and the increased customer pressure for such ser-vice—developing countries can ill-afford to ig-nore competitive pressures that are already in-formally transforming the market. A clear policyagenda responding to customer needs should bearticulated and implemented. New postal lawsreflecting market liberalization strategies are thenext major milestone in developing countries’sector policy reform.

POSTAL POLICY AND REGULATORY REFORM IN DEVELOPING COUNTRIES

44


From Universal Service toUniversal Access

Defined as “an obligation to provide a ubiqui-tous service at a uniform price at a reasonablyuniform standard of quality,” and long consid-ered a cornerstone for the postal sector, the con-cept of universal service has recently come un-der intense pressure. Its relevance as a sustain-able public policy goal is also being questioned.

In a low-volume, low-demand environment thetraditional monopoly-supported universal ser-vice—as experienced in high mail-volume indus-trialized countries—is of limited relevance (for adetailed economic analysis, see “Questioning theMonopoly”). The USP can be compensated forthe burden of universal service obligations (USO)through different schemes. The most frequentlyused scheme is that of the reserved area fromwhich revenues are generated to cover the cost ofUSO. Another scheme is a universal fund inwhich all sector operators contribute – usuallythrough license fees – to support the universalservice.2 Lastly, direct and indirect subsidies3

allow the USP to receive compensation for theextra costs related to USO. However, all theseschemes have shortcomings:

(a) In an environment with an average of lessthan one letter item per capita, per year, rev-enues coming from a letter monopoly areunlikely to cover the costs associated withUSOs (particularly the fixed-costs of postoffices throughout the national territory, andover-staffing with civil servants).4 Equally,it is unlikely that taxing private postal op-erators at an acceptable level will raiseenough resources to cover the USO costs.5

In addition, regulators do not have accessto sufficient information to define a fair li-

censing fee structure that is based on cleareconomic analysis, or to assess the real costassociated with the USO.

(b) Both the monopoly and the license feeschemes create potential impediments tosector development: monopolies lack com-petitive pressures to improve theincumbent’s performance, while high taxescan deter efficient private operators fromentering the market, hence, potentially lead-ing to unmet demand.

(c) Furthermore, in developing countries, themonopoly approach is a major burden forthe government. The public postal opera-tor often represents a significant fiscal bur-den on the state budget. Forty-five percentof countries with GDP per capita under$5,000 have postal revenues below operat-ing expenses (let alone total costs). Almostthree-quarters of African public postal op-erators reported a negative net result, andthis figure could well be under-estimated dueto unreliable accounting data reported byoperators.6 In some countries, the magni-tude of cumulated losses (up to 2 percent ofGDP in one case) is such that governmentsare unable to find politically and fiscallyacceptable schemes to bail out the publicpostal operator.

(d) The obligation to provide universal serviceamounts to providing often excessive cross-subsides between high-cost and low-costroutes and subsidies from the state budget.Subsidy schemes might not be sustainable.In a competitive environment subsidies(whether direct or indirect) risk a lack oftransparency (in the absence of cost ac-counting system) and can jeopardize thelevel-playing-field between competitors.

45

(e) Finally, universal service in developingcountries is no more than a “legal fiction”.Although the network of post offices oftenoffers the largest outreach to rural areas,there is no real “universal access” level.7

Delivery is infrequent, mail service is notsecured nor is it reliable, and there is onlylimited home delivery due to the absence ofaddressing systems.

This discussion shows that the traditional ap-proach is not suitable for developing countries.Instead of focusing on an ill-defined, all-encom-passing universal service that is not justified eco-nomically and is beyond the reach of developingcountries, governments are better off rethinkingthe notion of universal service concept. An evo-lution is needed from a static idealistic universalservice to dynamic and realistic universal accessschemes. This proposal is far from revolution-ary in the context of utility networks. In the tele-communications sector for instance, many devel-oping countries have reached the conclusion thatin their environment the Western concept of uni-versal service is unrealistic. Instead, developingcountries have their own version of universal ser-vice, wherein public access to telecommunica-tions services is at an acceptable distance frompeople’s homes. The mechanisms chosen toachieve universal access cover a range of policyoptions. Examples include: pay phones, publiccall offices, and community telecenters.

Given the peculiarities of the postal sector andthe legacy of universal service, policy reform isneeded to rethink the extent and form of a postaluniversal service policy, and define innovative andsustainable funding mechanisms to reduce thefiscal drain on governments. Formulating inno-vative access schemes and their implementationis by far the most important policy challenge in aliberalized framework. Smart subsidies combined

with innovative interconnection schemes couldenhance the provision of universal service in re-mote areas, at lower cost. In addition,policymakers should take into consideration thecurrent trend to bundle a variety of services dis-tributed through the postal network (government,financial, communication, see “Reinventing thePost Office”).

Towards an Effective PostalRegulatory Framework

Once the decision is made to establish a liberal-ized postal regime, it is essential that developingcountries put in place a regulatory framework toensure a smooth transition from a monopoly-re-gime to a market-based regime for postal servicedelivery. Market-oriented regulatory rules, pro-cesses, and institutions by which the economicbehavior of postal players (including the USP) isto be monitored are key to securing thesustainability of the reform.

Similar to other networked industries, regulationin the postal sector may be needed to ensure vi-able competition (starting with aligning the legaland regulatory framework to the actual level ofsector competition), correct market failures, andprotect both consumers’ and investors’ interests.In a perfect competitive scenario these goals couldbe reached by unleashing market forces. How-ever, in the context of developing countries apurely laissez-faire approach—subject only togeneral competition law—may not be adequateto protect public interest, at least immediatelygiven the lack of strong antitrust and competi-tion law tradition. Industry specific regulationmay be required.

While there is no single recipe for regulatory re-form, this paper presents the argument that les-


46


sons could be derived from the regulatory expe-rience of developed countries, in particular theEuropean Union’s (EU) approach, in the postalsector as well as from the accumulated experi-ence of other network industries, notably telecom-munications.

International Postal RegulationExperiences

Worldwide, postal experience suggests that com-petitive pressure has resulted (in most cases) inimproved service performance, increased end-user choices, lowered prices, and increased com-petitiveness of the business sector which relieson postal services as an input for transportingand delivering letters (bills, invoices, statementsissued by the public or the private sectors) andphysical goods (parcels, packages, usuallyshipped by the private sector).

Countries have adopted different regulatory ap-proaches and policies to ensure viable competi-tion in their postal sector. In rare cases (notablyin New Zealand) full unrestricted competition hasbeen introduced in all segments of the market withthe aim to maximize the benefits of competitionin the market.8 New Zealand embraced a light-handed approach to regulation and let marketforces set the final rules-of-the-game, subject toantitrust laws and enforcement of competition toprevent anticompetitive practices of dominantfirms. In the majority of cases including Austra-lia, Canada, the United Kingdom, and the UnitedStates policymakers preserved monopoly statusfor their incumbent postal operators for specifi-cally defined services in order to fulfill their uni-versal service mandate. In these cases a differ-ent regulatory approach has prevailed based onthe need for oversight of rates and services andfor safeguards against cross subsidies.

Paradoxically, even in developed countries thathave introduced full liberalization, the historicalpostal operator still maintains market sharesabove 90 percent; either because of regulatorybarriers to entry (in the form of excessive licenseconditions, for example, Sweden)9 or what canbe seen as the natural monopoly dimension ofpostal services (New Zealand).10 Some countrieshave established a strong regulator in charge ofimplementing an aggressive liberalization strat-egy, yet the historical operator maintains a strongdominant position as well (United Kingdom).

The European Commission’sApproach to PhasedLiberalization

The EU has followed a systematic and consen-sus-building approach to reform the postal inter-nal market. In the early 1990s it began by defin-ing a regulatory framework that according to Mr.Frits Bolkestein, the European Commissioner incharge of the Internal Market, would “…giveusers the benefit of competition while maintain-ing the availability of affordable services to all.”A key issue tackled by the 1997 and 2002 postaldirectives was to introduce a distinction betweenuniversal and nonuniversal services.Nonuniversal services cannot be reserved and arethus open to competition. For universal servicesthe directives set out a set of maximum reservedareas (Table 1). The EU plans for full liberaliza-tion of the postal market in 2009 although this isunlikely to be confirmed before 2006.

The Directives also established minimum require-ments for universal service which include allpostal items to 2Kg, domestic parcels to 10Kgand incoming international parcels to 20Kg. Itgoes on to specify collections and deliveries mustbe carried out five days per week (Monday to

47

Friday) and has targets for routing times of crossborder mail (85 percent J+3, 97 percent J+5).Member states have the freedom to work withinthese requirements (some member countries havecollections every day of the week and six deliver-ies per week) and can establish their own regula-tions concerning network access, complaints andredress, for example.

The phased approach to liberalization chosenby the EU, together with a sector wide consul-tative process, and constant monitoring andevaluation of liberalization impact is relevantto developing countries. Given the heightenedpolitical and social resistance to an outright andimmediate liberalization of the sector—eventhough it makes sense economically—the phasedapproached could be perceived as a pragmaticsector reform scenario for developing countries.Efforts to analyze the current market situation(supply, demand, use of technologies, contribu-tion to the economy, and spillover effects, etc.)will provide a baseline against which govern-ments can assess progress in reform implemen-tation and regulatory effectiveness, and definethe next stages of liberalization.

Lessons from theTelecommunications Sector:Is a Reference Paper onRegulatory Principles Required?

Market-oriented regulatory reform in the telecom-munications sector offers many references for

postal regulation. Over the years, regulatoryexperience has grown in areas like licensing ofcompetitive service providers, mandatory inter-connection rules, price cap regulation, targeteduniversal service funds and mechanisms, andprivatization. This experience was distilled intoan international trade instrument, the ReferencePaper (RP) on regulatory principles.

The RP consists of a negotiated set of commonguidelines for a pro-competitive regulatory frame-work that Members party to the General Agree-ment on Trade in Services (GATS) have attachedto their schedules as additional binding commit-ments in the context of the BTA concluded in1997.11 The RP is celebrated as a major accom-plishment of the Basic TelecommunicationsAgreement. It is the first international instru-ment that compiles guidelines for a global regu-latory framework to guarantee that trade liberal-ization is not impaired by domestic regulatorymeasures. This harmonized set of regulationsdeals with six regulatory principles includingcompetitive safeguards, interconnection, univer-sal service, licensing, allocation and use of scarceresources, and the creation of an independentregulator.

Many proponents of the RP’s approach have ad-vocated the importance of adopting a similar ap-proach in other “network industries,” especiallythose that are prone to regulatory barriers. Inthis context, specific proposals to adopt a simi-lar approach to postal and courier services havebeen tabled in the current round of GATS nego-tiations (Box 1).12

The argument is that, given the gradual openingof the sector to competition and the ongoing in-cremental reduction of “reserved” services inmany countries, regulatory authorities are likelyto face issues that are similar to those encoun-

Table 1. EU universal servicemaximum reserved areas


Weight Limit Price Limit 1998-2002 350 g 5 times tariff 2003-2005 150 g 3 times tariff 2006-2008 50 g 2.5 times tariff

48


tered in the liberalization of the telecommunica-tions sector.

The issue of postal universal service dominatesdiscussions on encouraging competition, creat-ing a level-playing-field, and acknowledging cus-tomers’ interests. In this regard, the most rel-evant aspects of the telecommunications RP arethose related to universal service and generalcompetitive safeguards. Under the RP, regula-tions pertaining to universal service are not perse prohibited. However, to be compatible withan open competitive environment, USOs have tosatisfy four conditions: they must be transpar-ent, nondiscriminatory, competitively neutral, andnot more burdensome than necessary for the kindof universal service defined by the member.Application of these principles will have far-reaching implications in revisiting the scope ofpostal universal service, and reserved andnonreserved areas in many countries. Such anapproach, if applied to the postal sector, can gofar in reconciling universal service objectives withthe workings of competitive markets.

Another interesting aspect of the RP is its com-mitment to adopt general competitive safe-guards—specifically those curtailing cross-sub-sidization to ensure an undistorted competitiveenvironment. Disciplines on public availabilityof licensing criteria are also relevant in the con-text of liberalization of the postal sector as li-censing can become a regulatory tool to erectbarriers to market entry. The RP’s disciplineson independent regulators are particularly rel-evant in the postal sector as many countries stillpreserve the ties between the operator and theregulator (as will be seen later in greater detail).Finally, it is important to assess the implicationsof cost-based principles (as those related to in-terconnection in the RP) on terminal dues andtariff setting practices. The move toward a postalRP is potentially controversial and calls for adeeper analysis of the tensions between the rulesand practice of the Universal Postal Union andthose of the GATS.13

Box 1. Chile

Negotiated during the Uruguay Round of multilateral trade negotiations, the GATS is the first multilateralframework for the global liberalization of trade in services. The GATS rules and disciplines extend to allservices except those provided in the exercise of government authority. Despite the prevalence of governmentownership of postal services in the overwhelming majority of countries, GATS ambit extends to postal ser-vices which, in most cases, are provided on a commercial basis and/or in competition with private suppliers.Postal services liberalization at the international level has been mainly driven by large private sector competi-tors in the postal sector. At the end of the Uruguay round, only a few WTO members undertook commitmentsin the postal/courier sector. When looking at these commitments, it seems that members were concerned aboutensuring their prevalent monopoly situation and preserving postal universal service. Since then, progress hasbeen made towards the continuing liberalization of the postal/courier sector. With the ongoing round of tradenegotiations, deeper liberalization is expected in the postal sector, supported by specific offers and commit-ments, and possibly even a postal RP.

49

From Policy to Implementation

In addition to facing the traditional issues of regu-latory reform (those pertaining to institutionaldesign, effective regulatory processes, and trans-parent regulatory procedures), developing coun-tries need to tackle sector specific regulatory is-sues in order to find innovative ways to encour-age liberalization and competition in the sector(under extreme market limitations and fiscal con-straints) while preserving the goal of a sustain-able, market friendly universal service policy.Several recommendations are given below to helpdeveloping countries in their design of a soundpostal regulatory framework.

Institutional Requirement:Regulatory Governance andIndependence

Establishing a regulatory framework for thepostal sector requires both substantive rules per-taining to universal service, licensing, pricing, anddesign of regulatory institutions. The role of aregulator as a referee in enforcing the rules be-tween market players is crucial for a successfulcompetitive market. For many developing coun-tries poorly planned regulatory reform will leadto negative outcomes. Failure to adopt effectiveframeworks that foster competition may “scare-away” potential players.

The institutional design of the regulator impactsits relationship with the policymaker, its account-ability, and its autonomy. To avoid potential con-flicts of interest, the postal regulator should beindependent of the operator and (preferably) ofthe policymakers. Yet in a fiscally constrainedenvironment, a stand-alone postal regulatory bodymight not be warranted: postal regulation couldbe located within the telecommunication regula-

tory institution, or in a multisectoral regulatorybody, as several support functions can then beshared to reduce overhead costs. Another argu-ment in favor of placing postal regulation withinan existing regulatory institution is that it mightaccelerate the learning curve of the postal regu-lation team through knowledge transfer from col-leagues working on other sectors. France, Mali,and Senegal have recently opted for this choice.

Improving Regulatory Capacity

Human resource constraints are a major handi-cap for developing countries to establish andimplement effective regulatory frameworks.Regulating the economic behavior of market play-ers requires adequate economic, financial, tech-nical, and legal capacity. Without such capaci-ties regulators do not have the tools to under-stand the market, analyze cost data, or measurethe operators’ performance.

Adjusting the USO

Expertise in the market is particularly importantwhen adjusting USOs to reflect market capacityand citizen demands. There is room for adjust-ment on the extent and form of the postal pres-ence (franchising the postal network), as well ason the quality of service (frequency of deliveryin rural and remote areas, delivering time adjustedfrom D+1 to D+5 for instance; place of delivery,including post office boxes vs. home deliveries)and tariffs (cost-based tariffs reflecting the addi-tional cost to deliver in remote areas).14 The im-pact of those adjustments to the USO should beassessed financially by the regulator so as tomonitor the subsidy scheme aimed at supportingthe USP. In this regard, the regulator should alsoencourage the USP to set-up a cost accounting


50


system compliant with international standards.Cost information provided by a reliable cost ac-counting system would help to reduce the debatearound the reliability of USO-related informa-tion provided by the USP.

Establishing Transparent andAccountable RegulatoryProcesses

Efficient and effective regulatory processes areneeded to reach the benefits anticipated fromthe reform exercise. Accountability of the regu-lator is an important guarantee for existing op-erators and potential entrants. Investors needinstitutional reassurance that the regulator willnot be captured by dominant operators. A pro-cess of appeal of regulatory decisions to juris-dictional or administrative bodies (depending oneach country’s institutional setting) may there-fore be warranted.

Transparency is also needed to increase the le-gitimacy of the regulatory exercise. It increasesthe pressure on the regulator to ensure open-ness, fairness, and objectivity of the regulatorydecisions. Transparency of the decision-mak-ing process can be guaranteed with the manda-tory publication of regulatory decisions, as wellas increasing public participation in importantpolicy decisions.

A postal regulatory system that is not optimizedcan hinder sector development. In China, someobservers estimate that the express mail segmentcould triple in value in view of the tremendousbusiness opportunities in the country. Yet theregulatory framework is currently a disincentivefor private sector investment as there is no vis-ibility in terms of government intentions to regu-late the sector (regulation is still implemented by

the public operator), liberalize the sector (or not),nor is there transparency on the licensing schemethat the authorities may develop.

Conclusion

An effective policy and regulatory framework iskey for postal sector growth. Few developing andtransition countries have taken the trouble of as-sessing their postal sector, its performance andpotential, as well as the policies and regulationsthat may foster its development. Too often, adhoc policy or regulatory measures are taken todeal with specific issues, such as express andcourier competition, office closures, postal defi-cits or diversification attempts. These measures,taken outside an overall postal sector strategy,rarely yield sustainable progress. The body ofsector reform experience is slowly growing, andthe European Union, as well as several othercountries, offer interesting approaches to sectorreform which should help policymakers in devel-oping and transition countries. No universal blue-prints are available, however, and reform needsto be tailored to the specific circumstances of eachcountry. Further data collection, analysis, andresearch would be needed to take stock of reformexperience to date and articulate best practicesrelevant to developing and transition countries.

Notes

1 “Reforming Infrastructure: Privatization, Regulationand Competition, A World Bank Policy Research Re-port, 2004.

2 This scheme is common in the telecommunicationsector, and is also stipulated by the 1997 Europeanpostal directive; it has not yet been successfully at-tempted in the postal sector.

51

3 Direct subsidies from the state budget, approved an-nually by the government, or some derogatory tax re-gimes.

4 The cost of USOs is seldom accurately assessed dueto the absence of cost accounting in most public postaloperators. Cost accounting systems are required, espe-cially in the cases where the post office provides bothmail-parcel services, and financial services through-out the network. However disagreement remains onthe best methodology to use (the European Commis-sion has favored the net avoided cost method, whileother stakeholders prefer the entry price method).

5 For instance in China, taxing the private postal op-erators present in the air express market (valued at $1.5billion) would raise less than one-tenth of the resourcesneeded to support the cost of the universal service. TheEuropean Commission estimated the average cost ofthe USO at about 5 percent of revenues in industrial-ized countries (likely to be relatively superior in devel-oping countries). Considering that China Post posted$6.8 billion of revenues in 2002, we would have antheoretical estimated cost of USO at $340 million (mostlikely in the low end of the actual cost). Taxing at, say,2 percent the express mail private operators would raiseabout $30 million, one-tenth of the cost of USO.

6 Source: Universal Postal Union - http://www.upu.int/statistics/en/index.shtml

7 There is one post office for about 3,000 inhabitants inFrance, compared to one for 102,000 in Mali and onefor 187,000 in the Democratic Republic of Congo.Countries with GDP per capita under $1,000 have onaverage a post office for each 4,700 square kilometersof territory compared to one office per 458 square kilo-meters for countries with GDP per capita above $5,000.

8 Volner, Brickman, Field, Postal reform: A Compara-tive Analysis and Critique of Regulatory Models –(Rutgers Conference on Postal and Delivery Econom-ics, Cork, Ireland) 2004.

9 Posten AB’s license features: every week-day deliv-ery (and at least five times a week), at least one clear-ance and one delivery per week day, high delivery stan-dards (85 percent overnight, 97 percent within threeweek-days), uniform prices for single piece mail witha price cap limiting price increases to the rate of infla-tion. In “Regulatory Challenges in the LiberalizedSwedish Postal Market,” Jonsson (Conference on Postaland Delivery Economics, Cork, Ireland, June 2004).

10 In New Zealand, an individual or company is able toregister as a postal operator with the Ministry of Eco-nomic Development; a one-off application fee of NZD95 ($62) is payable. In “The Liberalization of PostalServices in New Zealand,” Harford (Conference onPostal and Delivery Economics, Cork, Ireland, June2004).

11 The BTA was negotiated under the World Trade Or-ganization (WTO) and is appended to the GATS as itsfourth Protocol. It contains specific market access andnational treatment commitments in basic telecommu-nications services as well as a set of regulatory prin-ciples. This agreement covers basic telecommunica-tions services whether local, long-distance or interna-tional. It includes services for public and nonpublicuse provided on an infrastructure basis as well asthrough resale. It covers telecommunication servicesprovided by any means of technology including cable,wireless, and satellites.

12 See European Union’s proposal: http://europa.eu.int/comm/internal_market/post/doc/activities/job_en.pdf.

13 Study of TMC Asser Institute, The Study of the Rela-tionship Between the Constitution, Rules and Practiceof the Universal Postal Union, The WTO Rules andthe European Community Law, The Hague , 30 June2004 ( Unpublished Manuscript).

14 Regarding the nonuniform tariffs, numerous hurdles/technicalities of accounting systems, inconvenience forthe clients, explain that such cases are rare.


52


Chapter 6

Case Studies on Postal Sector Reform

Isabelle SegniJuam Ianni

The World Bank has been involved in postalreform projects throughout the world.This paper looks at three of those projects

in three very different countries; Morocco, Tan-zania, and Trinidad and Tobago. The case stud-ies review the reform process undertaken in eachof the three and analyses the impact of the re-forms with respect to performance, financial vi-ability, and sustainability.

In each of the three case studies, four reform el-ements were key: (a) commercialization of theoperator; (b) improved operational performanceof the operator; (c) improved financial perfor-mance of the operator; and (d) legal and regula-tory reform of the sector, including progressiveliberalization.

All three countries have become more commer-cially focused and all have performance contractswith government in the areas of quality of ser-vice, customer satisfaction and financial perfor-mance. This has led to improvements in each ofthose areas although there appears to be a trade-off between quality improvement and financialperformance. Trinidad and Tobago has made thegreatest quality improvements extending networkcoverage from 50 percent to 96 percent of thepopulation. However, they have failed to becomefinancially viable during the five years of reformdespite growing volume by over 150 percent.

Tanzania has managed to break-even in an envi-ronment where mail volumes are less than oneper capita per annum, by introducing prudentperformance and service objectives. Morocco hasconsistently made profits since its reform, in partdue to its more conservative operational targetsthan in Trinidad and Tobago. However, each canbe termed a success because they have all trans-formed their business from one with low customerperception to one providing reliable and secureproducts and services.

Legal and regulatory reform has occurred to somedegree in all three countries but market liberal-ization has not progressed as expected. OnlyTanzania has reduced its reserved area to 500gwhile Morocco maintains a 1,000g limit, andTrinidad and Tobago a 2,000g limit. This hasresulted in limited domestic competition in eachcountry. Tanzania also remains the only countryto fully separate the regulatory function fromgovernment, although the separation is plannedin the next phase of reform in Trinidad and To-bago as well as in Morocco, albeit with no changeto the reserved area.

The three case studies provide evidence of thebenefits of undertaking postal reform, but alsoidentify some of the trade-offs that must be madebetween financial viability and universal servicecoverage in a developing country context.

53

MoroccoSuccessful Public PostalOperator Restructuring butDelayed Sector Reform

Pre-reform Postal Situation

Market and RegulatoryEnvironment

Mail and domestic express services were the le-gal reserve of the monopoly postal operator “Of-fice nationale des postes et telecommunications”(ONPT) which also had some regulatory respon-sibilities. The ministry responsible for posts wasalso in charge of the policymaking and some regu-latory functions. Only the international expresssegment was open to competition with DHL be-ing a major player in the market. In absence ofan appropriate regulatory framework, interna-tional express carriers paid a fee to the publicpostal operator, as a contribution to the univer-sal service cost.

Although the postal sector represents a small partof the economy (0.2 percent to 0.3 percent ofGDP), its social importance is significant if wetake into account the universal postal servicesand the postal financial services through whichONPT provided basic financial services to manylow- to middle-income households.

Public Operator Status andPerformance

As an office nationale, the incumbent was insti-tutionally distinct from the administration, withan autonomous budget, but managed by processessimilar to those of the central administration.Approximately 70 percent of ONPT’s revenues

came from mail and parcels and the remainderfrom postal financial services.1 The postal rev-enues of ONPT grew by 5 percent from 1995 to1998, although the average number of mail itemsper capita remained under 8 and has stagnatedover the last years. The number of permanentpost offices also grew by 5 percent over the sameperiod to 1,469 offices in 1998, and while all of-fices accept financial transactions, fewer than halfoffer a full-range of postal services.

Reform Impetus and Strategy

Reform Environment

In 1988 the Government of Morocco split ONPTinto two government-owned companies, ItissalatAl Maghrib (IAM, the public telecom operator)and Barid Al Maghrib (BAM, the public postaloperator). This institutional restructuring wasmainly driven by the telecommunications sectorreform agenda, which aimed at a progressive lib-eralization and privatization of the incumbent.For the postal sector, a more cautious approachwas taken: BAM was created under public lawas a state-owned “établissement public àcaractère industriel et commerciall” (EPIC),2

and the government maintained a reserved areaup to 1kg for letters.

In 2001, the government adopted a postal sectorstrategy, which was included in the sector policyletter for the information infrastructure develop-ment. This inclusion signaled the government’sview of potential synergies between the postalsector and other elements of the information in-frastructure (for example, telecommunicationsand the Internet).

CASE STUDIES

54


Postal Reform Strategy

The sector policy letter noted three general ob-jectives for the postal sector: (a) implementingprogressive sector liberalization, (b) increasingBAM’s competitiveness through commercializa-tion and status change (if necessary),3 and (c)developing postal financial services to improveaccess to finance and channel postal savings tothe private sector.

The progressive liberalization, a key componentof the reform agenda, has not yet been imple-mented, but BAM’s competitiveness has madesatisfactory progress, as detailed below.

Regulatory Reform

While the telecommunications regulatory body,Agence nationale de régulation des telecommu-nications (ANRT) was created, the ministry incharge of the postal sector continues to share theregulatory function with the incumbent. BAMstill collects the license fee levied on internationalexpress operators. However a postal law is be-ing drafted that provides for ANRT to expand itsmission to include postal regulation.

Three years after the adoption of the sector policy,further liberalization of the postal sector has notbeen realized.

Commercialization of the PublicOperator and PerformanceAgreement

The Prime Minister presides over BAM’s boardof directors, which is comprised of representa-tives of eight ministries. Management and fidu-ciary autonomy and responsibility are relativelylimited, for instance, procurement of contractsabove a certain threshold are subject to a prioricontrol from the Ministry of Finance.

BAM has defined and implemented its corporateobjectives through a thorough reorganization,including territorial decentralization, and man-agers’ accountability on performance (based onagreed objectives and means). BAM hired (withinternal financing) a consulting firm to work, to-gether with BAM, on the planning and imple-mentation of the reorganization.

BAM has exceeded expectations in its strategicturnaround. Since its separation from telecom-munications it has been consistently profitable,avoiding burdening the government with subsi-dies (a pre-reform prediction). Through trainingand capacity building programs most staff havegained exposure to the new market-oriented cul-ture and management tools. Several senior man-agers have been hired from the private sector.Business agreements with private partners havedeveloped, and greater outsourcing to the privatesector is underway to optimize operational andfinancial performance.

One of the major instruments of the renewed dia-logue between the operator and the governmenthas been the performance contract. Although thecontract does not carry a formal incentive sys-tem (such as rewards if BAM meets its objec-tives [and penalties if it does not]), it offers theopportunity for the stakeholders to analyze thegovernment’s development objectives and to de-fine the conditions under which the public opera-tor can meet them. There are four categories ofperformance indicators that have been monitoredfor 2002 to 2004:

• Operational indicators: creation of postoffices and distribution routes; number ofletter items handled; deposits amounts andnumber of accounts for CCP and CEN;number and amounts of money orders; de-livery time (D+2); number of claims for let-ters and money orders;

55

• Investment indicators: quantitative andqualitative objectives of investment arespecified;

• Financial indicators: 11 indicators reflect-ing economic, financial, treasury produc-tivity performance;

• Human resource indicators: structure ofstaff and training budget are specified andmonitored on the basis of quantitative tar-gets.

Progressive upgrade of information systems al-lows BAM to more efficiently collect and aggre-gate operational and financial data generatedthroughout the network.

Sequencing

The postal sector reform began as a subset of thetelecommunication reform. Although the postalsector policy was developed well into the telecom-munications reform, it already needs to be updated.In 2004, the government initiated two studies, oneon universal service,4 and the other on the devel-opment of postal financial services. The govern-ment will use the results to finalize a new postallaw that emphasizes further liberalization, a bet-ter regulatory framework, and the transformationof BAM into a joint-stock company.

World Bank Participation

The World Bank’s support to Morocco’s postalsector reform has been limited to technical assis-tance support.5 However regular policy dialoguewith the stakeholders has yielded ideas and prin-ciples drawn from international best practice.Currently the World Bank is encouraging thegovernment to update the sector policy and ac-celerate the reform of the legal and regulatoryframework.

Reform Results

Measuring Progress

While the overall reform framework was balancedthe objectives and components appeared to be tooambitious given the government’s limited capac-ity to implement the reforms.

The performance contract for 2002 to2004 pro-vides a good framework for evaluating BAM’sperformance since 1998. As of 2002, BAM hassuccessfully met most of its objectives, and ex-ceeded them in terms of financial performance.

Sector-wise, the reform objectives have not beenmet, whether in terms of further liberalization,or in terms of regulation. Lack of market intelli-gence makes it difficult to assess sector growthor customer satisfaction. Hence the analysis be-low concentrates mostly on BAM’s results.

Volumes and FinancialPerformance

While mail volume has only slightly increased(Chart 1),6 BAM’s financial turnover (driven bypostal financial services and other sources of rev-enues), has increased by more than 45 percentover the 1998 to 2003 period (to 1,137 milliondirham [MnsDH] in 2003), while net revenueshave tripled from 55 MnsDH in 1998, to 153MnsDH in 2001. Domestic express activitieshave grown the most significantly over this pe-riod, reaching 16.5 percent in 2002, while postalfinancial services deposits have increased by 11percent in 2003. Net revenue in 2002 and 2003decreased as BAM funded its retrenchment plan(about 1,000 staff, Chart 2) which represent ashort-term cost (on incentive departure packages)with an expected medium-term return (reducedstaff costs, improved efficiency).7 In 2003 this

CASE STUDIES

56


translated into a 24 percent rise in productivity(increase in value added per employee).8

BAM is implementing a cost accounting systemto allow better identification of cost and profit cen-ters, and to provide more reliable information rela-tive to the cost of universal service and pricing.

Coverage

With about 1,500 post offices, BAM has the mostextensive network in the country, although theratio of post office per inhabitant remains low byregional standards (1 for 18,000 in Morocco,versus 1 for 9,000 in Algeria, and 1 for 8,000 inTunisia in 2002). There has been a slight im-provement in urban and rural areas.9 Efforts toincrease access are being considered taking intoaccount BAM’s financial sustainability. Solu-tions such as franchising are being examined.

Recently BAM implemented a geographic infor-mation system (GIS) to provide a better pictureof the adequacy of coverage between populationrepartition and postal presence.

Quality and Customer Satisfaction

Several investments (marketing and capacitybuilding efforts) were undertaken to improvequality of service. Customers are enjoying im-proved services—counter automatization (430 in2002, representing 95 percent of counter turn-over), upgrading applications and informationsystems, and the creation of a call center. Cus-tomer lines in post offices are closely monitored,resulting in organizational changes to optimizeservice. At this stage, most of the improvementin quality has benefited business clients, as wellas customers of postal financial services. Thedelivery time of mail remains lower than expected,with less than 60 percent of mail delivered onD+2 (2002 statistic).

With better service the number of claims hasdropped dramatically. Improving customer rela-tions is now a priority for BAM, though a sys-tematic customer satisfaction survey has yet tobe initiated.

Chart 1 - BAM letter volume

0.0050.00

100.00150.00200.00250.00

300.00350.00

1998

1999

2000

2001

2002

Letterdomestic(Mns)

Letterinternational(Mns)

Total letter(Mns)

0

20000000

40000000

60000000

80000000

100000000

1998 1999 2000 2001 2002

Chart 2 - BAM financial performance

Operating revenue Operating expenses Operating income

57

Investment and Innovation

BAM has initiated a program to upgrade and in-novate commercially. It started a joint-venturewith Chronopost for international express mail,it launched a new domestic express service, andit introduced a pilot hybrid mail center in 2003.Electronic payments are also increasing throughthe deployment of ATMs. BAM has acquired aGIS to facilitate informed strategic decisions rela-tive to its postal network. It developed a nationaladdress database that is the starting point to de-velop direct marketing activity. BAM is currentlyinvesting in a large hub to support logistics of itse-commerce activities (storage, ordering, prepa-ration, stock management, shipping and delivery).

Overall Market Development

Today there is not a government entity that iscapable of regulating the market, or of providingcomprehensive data on the market—and the Min-istry has limited capacity and resources to per-form these tasks. There are five internationalexpress carriers identified in the market, exclud-ing the recently created Chronopost-EMS joint-venture, demonstrating that the Moroccan inter-national express segment is attractive. Licensefees are collected by BAM (which has the rightto audit operational and financial accounts of itscompetitors). Given the lack of detailed account-ing information it is not clear whether BAM usesrevenues from reserved areas to cross subsidizethose in the competitive domain. For the remain-der of the postal market, it is impossible to as-sess volumes, values, market shares, output, per-formance, or trends.

Regarding postal financial services, the TreasuryDepartment is currently analyzing possible de-velopment strategies, which could mean retain-ing the status quo, or creating a new entity (pos-

sibly a subsidiary) with more or less extendedcapacities in financial, banking or nonbankingservices.

What Went Right,Areas for Improvement

What Went Right

The most visible success is the commercial ori-entation and financial sustainability of BAM, il-lustrated by the self funded modernization andthe increase in volumes realized through thoseinvestments. Prior to BAM’s separation fromItissalat, few observers would have anticipatedsuch a success. This can be partially attributedto the vision of its managing director and the com-mitment of the upper management team to re-form the organization, and also to the existingmarket conditions. Morocco has a thriving busi-ness environment that provides room for marketexpansion through fulfilling unmet demand forpostal services, assuming that BAM can providean acceptable level of quality of service.

Areas for Improvement

Morocco is facing the traditional “ universal ser-vice dilemma”: how to increase access withouthaving a negative impact on BAM’s financialviability. The government, through the perfor-mance contract, is encouraging BAM to increaseaccess through physical presence (the contractmonitors the number of new post offices), insteadof through alternative access options (franchiseor other private-public partnership). This willlead to increased fixed-costs for BAM with mar-ginal additional revenues realized.

With a reserved area of 1,000g the governmenthas not achieved its objective of market liberal-

CASE STUDIES

58


ization although the monopoly is not enforcedand BAM faces de facto competition. BAMhas announced that it is ready to open the sectorto competition, though the policymaker is hesi-tant to formally liberalize the market. The WorldBank supports the preparation of the draft lawthat will propose lowering the scope of the re-served area, and provide for an adequate regu-latory framework and universal service fundingmechanism.

The government has been slow to address theconditions to develop postal financial services—an important objective in the postal sector policy.The outcome of the ongoing study might callfor a substantial institutional modification (cre-ation of a subsidiary, new conditions of remu-neration for CCP, and savings deposits by theTreasury and the Caisse des depots et degestion). This mirrors the delay experienced inpreparing a new draft law. The slow pace is ahandicap for BAM as it is ready to change itsstatus to “limited company” to have greater com-mercial autonomy.

Sustainability

Morocco’s postal reform has a promising future,provided the government can quickly update itspostal legal and regulatory framework, andmodify BAM’s status. The government and BAMshould now build for this future by preparing thenew performance contract (2005 to 2007).

59

TanzaniaSolid Performance for aLow-income Country, butUncertain Sustainability



Tanzania’s postal market is limited by low GDPper capita (610, PPP 2001), minimal economicand commercial activity, and weak informationand transportation infrastructures. Over 80 per-cent of mail flows involves business-to-businessor business-to-customer communication with thelargest demand from utilities (water, telecommu-nications, electricity) and banks. Since less than10 percent of the population has access to grid-based electricity and 1.7 percent to fixed or mo-bile phones,10 Tanzania’s low annual per capitamail volume of 0.73 items is not surprising.11

Prior to reform, several unlicensed private op-erators functioned in an unregulated market eventhough a reserved area was stipulated in the ex-isting postal and telecommunications law. Themajority of these private providers are classifiedas “indirect” communication services such asfreight transporters, bus companies, or individualtravelers.


The pre-reform public operator showed little orno commercial orientation and offered a limitedproduct line and suffered from a low public im-age mainly due to unreliable service. For thefour years preceding the commercialization of thepublic operator in 1997, mail volumes fell by an

average 22 percent per annum; and parcel trafficdecreased by almost 50 percent from 1992 to1993. Performance was not monitored in areassuch as quality of service, customer satisfaction,volumes, or revenues. Prior to reform, the pub-lic operator was organizationally and financiallydependent on the public telecommunications op-erator, and functioned as a dependent agencywithin the Tanzanian Post and Telecommunica-tions Corporation (TPTC). The Post Office Sav-ings Bank functioned as a separate entity withinthe Ministry of Finance but provided accessthrough the postal retail network.


Reform Environment

During the late 1980’s and early 1990’s, the gov-ernment, with the support of the World BankGroup, initiated a program to restructure its pub-lic services to increase efficiency and lower costs.For the communications sector a critical first stepin this reform process was to separate the postaland telecommunications functions (as is commonin other African countries). This implied thatsubsidies from telecommunications would end,meaning that the public postal operator wouldneed to be financially viable.


The Tanzanian postal reform strategy was basedon a highly structured, three-stage approach:

n To establish a strong regulatory environmentand an autonomous public operator;

n To develop a contractual (performance)agreement establishing stakeholder account-ability; and

CASE STUDIES

60


n To commercialize the public operator to suc-ceed in a competitive and expanded market.

Regulatory Reform

In 1993, the government initiated this process bypassing a series of new communications andpostal laws to establish a sound legal and regula-tory framework for the sector.12 These laws sepa-rated TPTC into autonomous entities for postal,Tanzania Posts Corporation (TPC), and telecom-munications, Tanzanian TelecommunicationsCompany Ltd. (TTCL), and provided TPC withcommercial flexibility including the right to in-corporate subsidiaries and enter into commercialpartnerships.

The new postal law defined TPC’s universal ser-vice obligation (all letters, packets, and parcelsup to 10 kilograms) and associated privileges (areserved area up to 500 grams) and establishedthe “rules of the game” for competitive portionsof the postal market by establishing a coherentlicensing scheme. Finally, a new regulatoryframework was established to clearly demarcatethe functions of each postal stakeholder:

n Ministry of Communications and Transportsets overall policy for the postal sector

n Tanzanian Communications Commission(TCC) administers licensing and tariff set-ting for reserved services;

n Presidential Parastatal Sector ReformCommission (PPSRC) approves TPC bor-rowing, represents consumers, and deter-mines future restructuring or divestiture ofpublic entities (such as TPC); and

n TPC, under a license granted by the TCC,provides universal services at a level estab-lished in a performance agreement estab-lished with the government.

Commercializing the PublicOperator

Performance Agreement

The regulatory framework defined the rules ofthe game and the responsibilities of each of themajor players. However the government deter-mined that it would be necessary to measure thesuccess of these policy decisions and to ensurethat the public operators’ level of accountabilitywas commensurate with its newly granted man-agement autonomy. Therefore, a performancecontract was signed by TPC, PSRC, and TCCspecifying five performance targets:

n Quality of service, a negotiated percentageof inter and intratown letters to be deliveredwithin a specified number of days;

n Security, mail theft be kept below a definedlevel;

n Profitability, TPC must reach an establishedlevel of profit before taxes;

n Growth, expressed as a percentage of in-crease in mail volume annually; and

n Customer satisfaction, expressed as keep-ing customer complaints below a percent-age of mail volume.

Quantitative performance indicators were devel-oped for each area and measured by an indepen-dent third-party. Furthermore, financial bonusesfor exceeding, and penalties for missing the tar-gets were specified for the board of directors andTPC management. The first three-year perfor-mance contract was signed in 1997 and has beenrenewed once.

61

Building TPC’s Capacity

The government determined that to fulfill the re-quirements of the performance contract, compete,and be financially sustainable, TPC needed suf-ficient commercial capacity. A strategic plan wasdeveloped to establish corporate objectives to:improve the quality of service, increase employeeproductivity, support innovation of products andservices, and incorporate new technologies.Operating costs were reduced by adjusting theexisting network (mainly through franchising) andby staff reductions (nearly 45 percent, mainlythrough attrition).13 The most important factorin commercializing TPC was its separation intothree distinct business units: mail, counters, andexpress mail service (EMS). Each unit receivedadequate operational and financial autonomy tofocus on its customers and core products, therebydecentralizing authority and increasing manage-ment control. Significant capacity building wasalso undertaken in each unit: mail operationswere upgraded in the main sorting center in Dares Salaam, counter operations were automatedwith the support of the East African Develop-ment Bank, and EMS incorporated new trace andtrack systems.

Reform Process

The government not only undertook a well se-quenced postal reform (regulatory reform, ac-countability, and commercialization) it also al-lowed time to resolve the more challenging is-sues in the process. A good example is the oneyear “transition period” to split posts and tele-communications functions, to allow the legalseparation of existing balance sheets, determinehow to fund pension commitments, and how toallocate TPTC’s assets and liabilities between thetwo entities.


The World Bank’s support for postal reform inTanzania was one of its first projects in the sec-tor and began as part of a larger telecommunica-tions project. This support was an importantfactor in obtaining the technical expertise neededto develop effective market regulation and estab-lish an organizationally and financially autono-mous TPC. It was valuable in resolving highlytechnical issues such as the separation of postsand telecommunications and the design of a com-mercial accounting system for TPC. The tar-geted, if limited, investment in strengtheningmanagement information systems also helpedTPC to monitor and control its operations andfinances.14 The Bank’s participation was effec-tive in supplementing the government’s basicstrategy and vision for postal reform. The mo-mentum of reform has been maintained by postalstakeholders in Tanzania for many years follow-ing completion of the Bank’s support.

Reform Results

Measuring Progress

Tanzanian reforms have focused on the publicincumbent operator: trying to ensure it is finan-cially viable and can meet its universal serviceobligations. Consequently, the development ofthe supply-side of the postal market has consistedmainly of a few entrants into profitable marketsegments (for example, international courier andsome domestic interurban routes). This increasedthe choice for particular categories of consumersbut has not resulted in more choice or quality ofservice for most consumers in rural areas. De-mand served by the incumbent has failed to in-crease substantially over the period since re-form—it has risen and fallen from year-to-year

CASE STUDIES

62


but remained at around the same level over theperiod. TPC’s loss of market share to competi-tors suggests that aggregate demand has in-creased, although no data on volumes is avail-able from competitors. Consequently, the in-creased demand that arises from the reforms(such as from permitting new entry) does notnecessarily accrue to the incumbent. As a re-sult, TPC is looking to expand its services be-yond its traditional letters market and into otherforms of mail (like advertising mail) and otherservices (such as financial services). If thissucceeds it would be partly attributable to theincreased commercial focus instilled in TPC asa result of the reforms.

The requirements of TPC’s license and perfor-mance contract provide a good framework forevaluating the results of postal reform startingin 1998.


Chart 3 shows growth of letter class volume for1998 to 2002, but it does not appear that reformhas generated large increases in volumes. How-ever, domestic volumes, which are more respon-

sive to operational improvements, grew by nearly20 percent from 1998 to 2002, and overall letter-volume by approximately 8 percent.15 Mostpostal administrations in the world, includingthose in industrialized nations, have experiencedflat or declining volumes for letter class mailduring the same period. Productivity spiked in1999 (32 percent increase in items per employee;17 percent increase revenue/employee) due toreductions in staff and increased mail volumes;however, productivity has remained fairly un-changed in ensuing years.

During and following the reform, the policy de-cisions make it clear that TPC must be finan-cially self-sustaining and deliver an acceptablelevel of universal service without subsidy. A ris-ing cost per item handled (nearly $0.50 per itemin 2001) made these goals challenging. TPC hassucceeded by controlling costs (Chart 4)16 but atthe price of reducing coverage and increasingtariffs (by nearly 43 percent in real terms).17 TPCattempted to increase profit by aggressively mar-keting its new products; but revenues, like vol-umes, were relatively flat from 1998 to 2002.

-2.00

0.00

2.00

4.00

6.00

8.00

10.00

12.00

SDRs (millions)

Chart 4- TPC economic performance

Operating revenue

Operating expenses

Net (before taxes)

Operating revenue 10.30 9.60 11.00 10.60 9.70

Operating expenses 10.50 10.00 10.20 9.50 8.10

Net (before taxes) -0.20 -0.40 0.80 1.10 1.60

1998 1999 2000 2001 2003

Chart 1 - TPC Letter volumes

0.0

5.0

10.0

15.0

20.0

25.0

30.0

1998 1999 2000 2001 2002

mill

ion

s o

f it

ems

Domestic

Internationa

Total

3

63

Coverage (Outlets and Delivery)

Providing universal postal services is both chal-lenging and costly in a country of 37 million people(70 percent of whom live in rural areas) dispersedover 947,000 sq. km., serviced by a poor trans-portation infrastructure, and with a grossly inad-equate addressing system. Tanzania averagesabout 60,000 inhabitants per postal office, one ofthe highest ratios in the developing world. To con-tain costs the total number of post offices (TPCand franchised) was reduced from 500 in 1993, to422 in 2002. Almost all mail is delivered at postoffices, through private post office boxes or overthe counter. However, the reduction in networkreflects a closer approximation of access with de-mand, and there appears to be relatively uniformcoverage throughout the country.

Quality of Service and CustomerSatisfaction

Gathering data on quality of service is difficultdue to the dispersed nature of Tanzania’s postalnetwork and the light volumes of mail carried.However, recent studies show that TPC is meet-ing the quality standards of the performance con-tract (Table 1).18 As with quality of service, themeasurement systems used for monitoring cus-tomer satisfaction do not produce detailed infor-mation. They are based primarily on monitoringthe number of complaints submitted to the op-erator and regulator. While recent studies indi-cate this target is being met, 19 growing competi-tion from private operators show there is still dis-satisfaction with TPC’s efficiency.

Innovation

Each of the three corporate areas have made sig-nificant efforts to research, develop, and marketnew products in order to generate extra revenue.

n Mail. Tariffs were based on the speed de-sired by the customer (economy or priorityservices) rather than on the content and sizeof the mail piece. TPC is also free to nego-tiate individual rates with large mailers andis developing an advertising mail business.

n EMS. In addition to the normal express mailservices, new products such as EMS faxservice and money-fax were introduced. Anagreement recently was signed with DHLfor delivery of international items in desig-nated countries while TPC delivers domes-tic items for DHL.

n Counters. As part of its revenue-generat-ing strategy, TPC’s retail network sells othercommunications products including station-ary, packaging material, photocopying, andfax services, and a few sites offer Internetaccess. Perhaps TPC’s most successful newproduct area is that of financial servicessuch as a “giro-banking” that allows insti-tutions to pay salaries, pensions, and divi-dends to its employees and shareholdersthrough the postal network. TPC also re-tains 30 percent of the shares in the PostOffice Bank, which provides another sourceof revenue.

Performance measurement Targets for 2001

Area

Mail intra-town

Mail inter-town Zone A

Mail inter-town Zone B

Mail inter-town --District

Target

J+0 (98%)

J+1 (85%)

J+2 (88%)

J+5 (80%)

Table 1. Performancemeasure targets

CASE STUDIES

64



In Tanzania an original goal of the reform was toincrease competition and grow the postal mar-ket. Private sector participation has slowly andsteadily increased since reform was initiated.There are now two licensed international postalcompetitors (DHL and TNT) and 14 other li-censed competitors who primarily use their ownnetworks (e.g., bus operators) to transport mail.Competition is particularly keen in express mailand parcels due to their profitability; so far, nooperator has applied for a license to deliver mailin the reserved area (under 500g). Above 500g,competitors enjoy 15 percent of the overall lettermarket and a significant portion of the expressmail market.20 Their higher prices, greater reli-ability, and same day service have created a qual-ity niche above TPC’s economy and priority ser-vices. At this time there is little threat fromInternet services since less than 0.5 percent ofthe population are Internet users.21 TPC increasedprivate sector participation through franchisingand commercial partnerships and by interconnect-ing with private transportation companies to ar-range for delivery of mail in remote areas.

In summary, while the reform has created a mod-est growth in competition, private sector partici-pation, and volumes it has not decreased the costof providing basic postal services.

What Went Right, Areas forImprovement

What Went Right

n Tanzania’s postal reform established a solidlegislative and regulatory foundation for thesector with a clear division of responsibili-

ties and transparent processes for postalregulation and performance. These reformswere properly sequenced and timed and thegovernment ensured that sufficient resourceswere available to complete the reformagenda.

n TPC has commercialized its operations andfinances and has developed a solid market-ing strategy based on determining and meet-ing unmet demands in the postal market.Quality of service, security, and product di-versity all significantly improved. TPC hascontinued to be self-sustaining and has sig-nificantly lowered the cost to the govern-ment of providing universal services. A self-sufficient postal operator is rare in coun-tries at this economic level.


n While regulatory and institutional reformswere implemented and the cost to the gov-ernment of providing universal service waslowered, the wider sector reforms (such asdecreasing the cost of basic postal servicesand significant market expansion) have notbeen achieved.

n Although the performance contract has pro-vided a worthy yardstick for measuringprogress, to date penalties have not been as-sessed in areas where targets have beenmissed. It is unlikely that such penaltieswill be levied except in severe cases ofunderperformance. The incentive mecha-nism embedded in the performance contractmay not come into play if penalties are notlevied.

n Despite successful reform, postal tariffs inTanzania are high; the lowest postage rate(economy service, 20 grams or less) is the

65

same as the price of a loaf of bread, or 30minutes Internet access. Recent studies es-timate that the tariff for basic letter servicescould represent a cost between 3 and 9 per-cent of typical monthly expenditures.22.

While pegging TPC’s prices closely to costhas allowed it to maintain financial equilib-rium, such frequent rate hikes may drive thecost of postage beyond the means of theaverage mailer, and may even be depress-ing mail volumes. Further analysis of priceelasticity may be necessary to better appre-ciate impact on demand.

Sustainability

Without significant market growth, Tanzania mayfind it difficult to sustain the benefits achievedby its postal reform. Competition is growing andthe relatively high prices needed to maintainTPC’s financial viability may depress volumeover time. Furthermore, the TPC’s low level ofprofitability (recently 2.5 to 5 percent) does notprovide sufficient excess capital for investmentin future development. The government is con-sidering four options to increase TPC’s access tooutside investment: enacting legislative changesthat would allow TPC to borrow funds indepen-dently of the Ministry of Finance; separating com-mercial (EMS) from social (basic letter mail)services with the former services entering intopartnerships with the private sector; convertingTPC from a public corporation into a limited li-ability company (as was done with TTCL) whichwould open TPC to partial privatization; or com-pletely privatizing TPC.23 These strategies fo-cus on private investment to further extend thebenefits that have been achieved throughTanzania’s postal reform.

CASE STUDIES

66


poorly defined universal service for basic mail,was not accountable for performance, had nomarketing capacity, and ran its finances accord-ing to the government budget cycle. Private sec-tor participation did not exist at any level withinthe public operator.


Reform Environment

Prior to 1999, several reform programs had beenimplemented in Trinidad and Tobago to boost theeconomy and strengthen the country’s financialsystem through a redefinition of the scope andsize of the public sector and a wide-range ofprivatizations. More recently, the governmentdeveloped a “2020 Vision” based on raising in-frastructure services to world-class standards by2020. For the postal sector this means bringingpostal financial and operational performance tothe levels of the most successful postal servicesin the industrialized world.


In 1999 the government established a two-phasedapproach to modernize the postal industry withstrong support from the wider postal stakeholdercommunity. The first phase would restore pub-lic confidence in the public operator, build postalvolumes and revenues, improve service, extenduniversal network coverage, and gradually reducethe levels of government support. Increased pri-vate sector participation in the public operator’smanagement and operations was seen as the bestmethod for achieving these goals.

In the second phase, a revised legislative and regu-latory framework would be established to stimu-late overall postal market development. During

Trinidad and TobagoImproved Postal Services ThroughManagement Contract, butUncertain Financial Sustainability



Prior to reform in 1999, the public postal opera-tor in Trinidad and Tobago was caught in a “vi-cious circle” of poor financial performance, lowinvestment, poor customer service, and decreas-ing mail volumes. The per capita mail volumestood at 12.6 pieces—low for an upper-middle-income developing country—and overall mailvolumes dropped by 20 percent between 1995and 1999.24 Quality of service was poor withmail frequently taking seven to 10 days to be de-livered, and universal service reached only 50percent of households. It was estimated that per-haps no more than half the population had a posi-tive response to Trinidad and Tobago Post Of-fice and Postal Savings Bank, the public opera-tor.


Financial performance was dismal; the post hadlost money for years and revenues typically cov-ered only two-thirds of costs. There was a greatdeal of unregulated competition for regular mailservices, much based on “self-delivery” by util-ity companies. The existing postal law estab-lished the Post Office and Savings Bank as agencydivisions of the Ministry of Public Utilities andEnvironment and the Ministry of Finance respec-tively. The public operator was responsible for a

67

this phase, the commercialized public operatorwould gradually be exposed to more competition,and strengthened through increased private sec-tor participation through a long-term arrangementsuch as a concession or partial privatization.

Regulatory Reform

A delegated management arrangement (DMA)was identified as the vehicle for the first phase ofthe reform, namely to rehabilitate the public op-erator by introducing commercial discipline andinnovation into its operations and management.In order to provide the legislative and regulatoryframework needed to support a management con-tract, the Postal Act of 1999 was established toprovide more management autonomy and com-mercial flexibility for the new public postal cor-poration, Trinidad and Tobago Post (TT Post); areserved area was maintained for letters with theunderstanding that over time this would be sub-ject to gradual liberalization.

Commercialization of the PublicOperator and Performance Targets

The DMA, a five-year management contract, wasawarded to New Zealand Post International Lim-ited (now Transend Worldwide Limited) after acompetitive bidding process. Transend was obli-gated to manage TT Post in line with internationalbest practices that would be measured by annualquality of service and revenue targets; with pay-ment of management fees tied to achieving tar-gets. Independent third-party organizations wouldmonitor, measure, and review the progress basedon the following five performance indicators:

n Improved level of customer satisfactionamong mailers;

n Expanded reach of universal delivery inTrinidad and Tobago;

n Met standards for transit time of mail fromcustomer deposit to delivery;

n Growth in revenue from postal services; and

n Met net income target specified in the DMAcontract.

Reform Process

The reform process began with the government’sdesign for the broad strategies to rehabilitate thepublic operator and establish the groundwork forfuture market liberalization and increased privatesector participation. The next task was to en-gage stakeholders (including major mailers, la-bor unions, consumer groups, and private postaloperators) in building consensus for reformthrough a series of information sessions and work-shops. There was a general consensus concern-ing the significant need for reform and the gen-eral strategies for achieving this transformation.The government’s “reform champions” did anexcellent job of explaining the goals of the re-form at Universal Postal Union meetings andother international fora. This process helped tobuild interest among potential bidders interestedin the management contract.


Support from the World Bank was sought to helpconduct detailed postal market analysis, policy-setting, and long-range strategic planning activi-ties. During the detailed DMA design and inter-national bidding process; the Bank’s extensiveexpertise in these areas helped to ensure a com-petitive search for a competent and committedpartner in TT Post’s transformation. As the DMAhas progressed, the Bank has provided technicaladvice in areas such as financial audits needed tostrengthen TT Post’s institutional structure. Aloan package of approximately $11 million was

CASE STUDIES

68


secured to provide the investment required toimprove the public operator’s management struc-ture, processing and delivery capacity, and retailnetwork. During the transition from DMA tolong-term arrangement, there will be a greaterneed for policy advice and assistance in develop-ing long-range strategies and a supporting legis-lative and regulatory framework for the sector.

Reform Results

Measuring Progress

A key project objective was to open the sector tocompetition midway though the DMA which didnot happen. In respect to the reform of TTPostthe data provided by the DMA’s monitoring sys-tem at the close of Year 4 provides a good van-tage point to determine progress:


Chart 5 shows the dramatic growth during between1998 and 2002: total volume 157 percent; percapita annual volume 145 percent; items handledper employee 175 percent; and revenue per em-ployee 184 percent, all while achieving a 39 per-

cent reduction in staffing. The current per capitafigure of nearly 30 pieces annually exceeds theGDP/GDI predicted figure for Trinidad and To-bago and demonstrates the positive contributionthe postal sector is making to economic growth.Revenue growth met all targets imposed by theDMA and (Chart 6) grew by almost 75 percentfrom 1998 to 2002.25 Despite this growth TTPost did not make an operating profit by Year 4of the contract (as had been assumed) leading toa continued need for government support. Whilethere are institutional (inadequate capitalization,incomplete vesting of property, and a lack of acollective bargaining agreement) and operationalreasons (rapid expansion of network coverage andquality) for these continued losses, it also indi-cates that there is a weakness in cost contain-ment in the DMA design and in the ongoing op-eration of TT Post.

Coverage (Universal Delivery)

The DMA requires the management contractorto maintain a delivery network that reaches allparts of the country. Chart 1 shows that fromapproximately 50 percent in 1999, TT post nowreaches 93.6 of all addresses in the country on adaily basis.

-4,000,000-2,000,000

02,000,0004,000,0006,000,0008,000,000

10,000,000

SDRs (millions)

19981999200020012002

Chart 2 - TT Post financial performance

Operating income

Operating expenses

Net income

5 6

1086420

-2-4

Chart 5 - TT Post operational performance

0

20

40

60

80

100

1998 1999 2000 2001 2002 2003

Volume (millions ofitems)Items per employee(thousands)Percentage homedeliveryCustomerSatisfaction

69

Quality of Service (Transit Time)and Customer Satisfaction

The DMA set aggressive quality-of-service stan-dards for next-day mail delivery for mail postedby an established cut-off time. Prior to the intro-duction of the DMA it was estimated that lessthan 50 percent of mail was delivered within twodays, whilst the latest independently monitoredfigures show that over 90 percent of mail wasdelivered by the second day after posting. Moreimportantly, business and individual customersatisfaction with the speed and quality of deliv-ery by TT Post (as measured during CustomerSatisfaction Index [CSI] surveys) is at an all-time high. (The independent CSI measures cus-tomer perception on postal service performanceover a broad range of factors.) Despite two in-creases in rates TT Post’s CSI scores consistentlyimproved during the first four years of the con-tract. The current CSI level of nearly 85 percentis a dramatic increase over the estimated 50 per-cent at the start of Year 1.

Innovation

The most significant factor to growth was newbusiness, which increased by nearly 50 percenteach year during Years 3 and 4. The successfuldevelopment, marketing, and delivery of newbusiness products demonstrate the improved com-mercial culture within TT Post and the ability touncover and respond to emerging market de-mands. New products included bill payment atpost offices, the country’s first domestic expressmail service, and bulk mailing offerings tailoredto major mailers. Early in the project it was de-cided to terminate postal banking services sincemost of the accounts were dormant, and sinceTrinidad and Tobago already enjoys a very denseretail banking network.


It is difficult to determine whether the TT Post’sincreased mail volumes represent new mail, orwhether the increase is due to business capturedfrom private operators—due to the publicoperator’s extremely efficient network and con-tinuing monopoly. However, evidence suggeststhat TT Post’s increased level of efficiency, ex-panded coverage, and successful marketing ofnew products has created a new interest in us-ing mail as an advertising and business com-munications.

What Went Right,Areas for Improvement

What Went Right

n Enhancing postal network capacity and in-creasing efficiency through the introductionof an independent management contract hasled to significant increase in volumes andrevenues; this suggests that other countriescan achieve similar results by adopting asimilar approach where latent demand forefficient postal services exists.

n The measurement of five key performanceindicators has led to a higher accountabilityand the ability to measure progress. This inturn has led to adjustment (for example intotal revenue targets) as well as a clear un-derstanding of issues to be addressed (costcontainment), before the next project phase.This would be impossible without quantita-tive performance indicators, objective mea-surement systems, relevant performance tar-gets, and structured accountability. For yearsto come the benefits of having a comprehen-sive addressing system and increased accessto all population segments will pay dividends.

CASE STUDIES

70



n After five years under the DMA, TT Postis not financially self-sufficient and is a con-tinuing (though decreasing) drain on gov-ernment resources. Expanded coverage andaggressive quality-of-service targets haveadded to the cost of universal service provi-sion and it may not be feasible to achieveuniversal next-day delivery and financialequilibrium throughout the network, particu-larly in remote areas.

n Because of TT Post’s continuing financiallosses there will be a two-year “transitionperiod” between the DMA and the long-termarrangement. This delay may pose some riskin terms of losing momentum for increasingprivate sector participation in TT Post.

n For the immediate future (in light of its cur-rent financial situation), the government iskeen to delay market liberalization. It is feltthis decision should be reconsidered since lib-eralization will focus the incumbent on be-coming more efficient and increased compe-tition will give greater choice to the consumer.

Sustainability

Based on the first four years of the DMA, it issafe to venture that TT Post will emerge from thecurrent management contract with a strong in-frastructure and a thriving postal business. Sig-nificant investments have been made to improveprocessing capabilities (including a new sortingcenter), delivery efficiency (a new delivery fleet),and to refurbish TT Post’s retail outlets. Newbusiness has been identified and captured, andthe image and reputation of the public operatorhas been strengthened. In order to sustain thesegains, as well as to realize the primary goals of

the second phase of the reform project (increasedprivate sector participation and competition), itwill be necessary to address three key actions:

n Financial viability by resolving outstandinginstitutional and operational issues that in-hibit financial performance, and by devel-oping a cost containment program.

n Establish a sector policy that emphasizesthe drive towards a state-of-the-art postalservice, as well as the strategies to increasecompetition and private sector participation.

n Design of a legislative framework to sup-port and implement the sector policy. ThePostal Act of 1999 represented a good firststep towards reform by providing TT Postwith some basic autonomy and flexibilityneeded to operate as a business. The DMA,conversely, provided a contractual form ofregulation through contract monitoring andfulfillment. However both TT Post’ssustainability as well as the long-range sec-tor policy goals (such as a more competi-tive postal market) require a comprehensiveand sector-wide approach to regulation.

The DMA provided an efficient and productivevehicle for building the postal business in Trinidadand Tobago. Now the challenge is to build on thissuccess to ensure future sustainability and growth.

Notes

1 Mainly current accounts “compte-chèque postal”and savings accounts “caisse d’épargne nationale”and money orders.

2 IAM was set-up as a “société anonyme” (joint stockcompany) under private law.

71

3 The change of status aims at transforming the EPICinto a société anonyme. This would, among otherthings, remove the a priori budget control from theMinistry of Finance.

4 In the last few years several studies related to theuniversal service were conducted. These studies il-lustrate the government’s cautious approach to sec-tor reform and to the introduction of competition.

5 A larger adjustment credit focused on telecommu-nications and information technologies.

6 Source: Universal Postal Union.

7 Source: BAM.

8 Source: BAM.

9 BAM reached 1.13 postal counter per 10,000 inhab-itants in urban areas, and 1.03 in rural areas, againstrespectively 1.00 and 0.92 set as initial objectives.

10 Source: ITU 2002

11 The average for sub-Saharan Africa is 0.86.

12Tanzania Posts Corporation Act, 1993: createdTPC as an autonomous government corporation witha managing board appointed by the Ministry of Com-munications.Tanzania Telecommunications Company Organiza-tion Act, 1993: established TTCL as a limited li-ability company and provided general strategy fordividing common assets and liabilities between thetwo new organizations.Tanzania Posts and Telecommunications (Vestingand Liabilities) Act, 1993, provided detailed guide-lines on division of assets and liabilities between theorganizations.Tanzania Communications Act, 1993: establishedthe TCC to regulate both TPC and TTCL. TTC over-sees licensing and tariff setting for reserved services/Public Corporations (Amendment) Act, 1993: cre-ated the PSRC with responsibility to restructure anddivest government organizations. The PSRC has

CASE STUDIES

control over some TPC functions such as the raisingof new capital. It will decide whether and how TPCis privatized.

13 Universal Postal Union and World Bank Group.2002. The Postal Industry in an Internet Age Wash-ington, DC: UP and World Bank.

14The approximately $2.9 million technical assis-tance loan was also used to build TPC’s capacitythrough acquisition of computer hardware and soft-ware, postal equipment and supplies, vehicles, con-sultant services in postal operations and cost account-ing, and a wide-range of management, marketing,and financial training.

15 During this same time period, GDP and GDP/capita, the best indicators of postal volumes, werebasically flat in Tanzania.

16 Source: Universal Postal Union and World Bank.

17 TPC pays taxes on its net profits and a royalty of 1percent to the TCC based on its volumes.

18Frontier Economics Study of Tanzania and Uru-guay (unpublished).

19 Not in the public domain.

20 Private operator volumes have grown slowly sincereform was initiated, along the same lines as TPC’s.volumes.

21 World Bank Development Indicators, 2003. Lat-est available data on usage levels is from 2000.

22 Frontier Economics.

23 Interview with Director General Minde in UnionPostale.

24Analysis of GDP/GDI compared with per capitavolume in countries with levels of economic devel-opment similar to Trinidad and Tobago’s would in-dicate that this number should have been closer to26 items.

25 Source : Universal Postal Union.

72


Chapter 7

Conclusion

Postal networks of the 21st century are rel-ics of a bygone era when mail was the onlyform of distance communication. Postal

networks were established to ensure that the wholepopulation had access to this once unique formof communication. Those networks are still main-tained to ensure the continued provision of uni-versal service, albeit in a state of poor repair withoutdated equipment and poorly trained staff. Thisis particularly the case in developing countrieswhere networks were often established to fuel theeconomic activity of a once colonial power. Con-versely nowadays these networks depend for theirsurvival on heavy subsidies acting as a drain ongovernment resources with their extensive net-works and low volumes.

Today telecommunications and electronic com-munications provide easier and quicker means ofkeeping in touch. Indeed, such new technologiesare serving as substitutes for postal communica-tion and in turn diminishing the importance ofthe social function of the postal service. Suchsubstitution brings into question the continuing

need to burden the taxpayer with the provision ofpostal services. Governments are beginning toquestion whether such investment would be bet-ter spent on increasing access to other public ser-vices which may have a more significant impacton social interaction, poverty reduction, and eco-nomic development than the continued investmentin the postal service.

That is not to say that mail does not have a partto play in the social and economic developmentof a country, it certainly does. But the role of thepostal sector needs to be viewed on present needand not historic precedent. The definition of uni-versal service therefore needs to be debated withineach country and its provision defined based onthe balance of social needs and the cost of pro-viding the service. Industrial countries are oftenused as sources from which to adopt best prac-tice but in the case of universal postal service itis unlikely to be a good source. Next-day andeveryday delivery is neither required, attainable,nor cost effective. Developing countries shouldbe wary of the tendencies that exist to promote

73

and apply such highly defined universal serviceconcepts. In many developing countries, postalservices offer low quality, insecure service, pro-vide only the most rudimentary products, and tendto have low-quality management often appointedfor political reasons. At the same time they areoffered the protection of a reserved area so thatthey can theoretically maintain a universal ser-vice that they practically cannot provide.

The role of posts in the developing world mustevolve in a way that assists in the economic de-velopment of the country. But economic devel-opment does not come from having a postal net-work that links every citizen in the country to aservice that they neither need nor use. It comesfrom identifying the needs of the market and es-tablishing products and services that fulfill thoseneeds. That does not mean next-day-delivery, nordoes it mean daily collection and delivery to ru-ral areas that send and receive little mail, nor doesit mean establishing dedicated postal outlets invillages that conduct fewer than 10 transactionsper day. But such situations currently exist andgovernments must take on the task of reformingtheir postal services so that they can reduce theburden of supporting them financially.

Best practice often involves access provided notby dedicated post offices but by private agentsperforming postal services from already estab-lished businesses for a fraction of the cost of dedi-cated facilities. If post offices throughout theworld are to survive and serve their communitiesthey must undergo radical change. The time for“bandage” solutions has gone and the time formajor surgery has arrived. Governments mustreview the needs of their postal market and moveaway from the romantic notion of universal-for-all to a notion of universal access based on needand cost.

Governments must look at the sector as a wholeand provide those who drive economic develop-ment with the tools that will help them do justthat. Competition should be introduced in themarket to provide the business community withquick, reliable, and secure services that are oftennot provided by the incumbent. Such competi-tion will force the incumbent operator to improvetheir service, become more efficient, and focuson the customer which will benefit the commu-nity as a whole.

Post offices can also act as a platform for theprovision of other services and ensure that theyuse their infrastructure to its full potential. Theycan diversify and offer services on behalf of gov-ernment and other businesses. If they have toomuch space, they should offer it for rent to busi-nesses that can attract customers through the door.Post office outlets that receive no customersshould be closed or replaced by agents that oper-ate on a transaction fee basis. Distribution net-works do not solely need to carry mail; and ifspare capacity exists, the postal operator shouldlook at what else they can distribute on behalf ofother people or use other networks for the car-riage of mail.

Such diversification into other businesses requirescapacity that rarely exists within a largely civilservice or politically managed organization. Afundamental step in transforming the fortunes ofa postal service is the introduction of businesspractices and commercially-minded managementwith experience in changing postal services fromloss-making, internally focused government de-partments into profit making, customer focusedbusinesses. Governments should follow examplesof reform around the world that have success-fully transformed postal operators. They shouldlearn the lessons, good and bad, of those examplesand put in place their own process of reform that

CONCLUSION

74


will lead to a streamlined postal operator withgrowing volumes, increased quality and customersatisfaction and reduced subsidies from govern-ment.

Many people will maintain that the post office pro-vides a vital social service and that by default sub-sidies need to exist in order to meet this key ser-vice. This is often used as a reason for maintain-ing the status quo and avoiding postal reform. Inpractice, however, customers in many countriesare being failed by their postal service, and de-spite the extensive networks there is no real ser-vice. Trinidad and Tobago has increased accessfrom 50 to over 90 percent and customer satisfac-tion has also increased by a similar margin sinceundergoing a postal reform process. The post of-fice has not yet reached the reform goal of finan-cial viability but surely it is better for the govern-ment to financially support a postal service that isincreasing access, raising volumes, expanding ser-vices, and increasing quality and customer satis-faction than support one that meets none of thosecriteria.

The Argentina model of reform has come undercriticism after the concession was revoked in No-vember 2003 but the seven years of private man-agement has resulted in significant investment inautomation, leading to new products and services,a new brand image and a customer focus that hastransformed the old state run post office into anefficient customer focused organization.

Finally Tanzania is another example of whatpostal reform can achieve in developing coun-tries. Once a heavily subsidized and inefficientpostal service, Tanzania Posts Corporation is nowhailed as an example throughout the region onhow postal services can meet social objectivesand the needs of the economic community with-out burdening the taxpayer.

Postal reform should not be seen as an enemy ofthe postal service, it should rather be seen as itsliberator. Postal reform provides the users of thepostal sector with the freedom, choice, and qual-ity they require to sustain their economic devel-opment and in turn, contribute to the social needsof each country in a timely and cost effectivemanner.

75

Bibliography

Chapter 2: Reinventing the PostOffice

CGAP (Consultive Group to Reach the Poorest).2004. Occasional Paper, July 2004. 8:5.

Correos Chile. 2004. <www.correos.cl>.

Guislain, P. 1997. The Privatisation Challenge - AStrategic, Legal, and Institutional Analysis of Inter-national Experience. Washington, DC: World Bank.

PWC (PricewaterhouseCoopers). 2004. “PostalTelecentres: Rural Development Through InternetAccess is South African Development Community.”Report produced for the World Bank. Processed.

Shirley, M., and J. Nellis. 1991. Public EnterpriseReform - The Lessons of Experience. Washington,DC: World Bank.

Universal Postal Union. 2004. <www.upu.int>.

World Bank. 1995. Bureaucrats in Business - TheEconomics and Politics of Government Ownership.Washington, DC: World Bank.

Walsh, T. 2001. Delivering Economic Develop-ment. London: Consignia.

Chapter 3: What Drives PostalPerformance? A Cross-CountryAnalysis

Ahmed, H. 1981. India Post Through the Ages.Postal History Society of India. Processed.

Forestier, E., J. Grace, and C. Kenny. 2002.“Can Information and Telecommunications Tech-nologies be Pro-Poor?” TelecommunicationsPolicy.

Frontier Economics. 2002. The Impact of Lib-eralization on Efficiency: A Survey. London:Frontier Economics. Processed.

Kirkman, G.., P. Cornelius, J. Sachs, and K.Schwab. 2002. The Global Information Tech-nology Report 2001-2002. New York: OxfordUniversity Press.

Nankervis, J., I. Carslake, and F. Rodriguez.1999. “How Important Have Price and Qualityof Service Been to Mail Volume Growth in Crew”in M. Crew and P. Kleindorfer (eds) EmergingCompetition in Postal and Delivery Services.Boston: Kluwer Academic Publishers.

BIBLIOGRAPHY

76


Ranganathan, K. 1998. Redirecting Mail:Postal Sector Reform, Washington DC: TheWorld Bank and UPU.

Walsh, T. 2001. Delivering Economic Develop-ment. London: Consignia.

Walsh, T. 2002. “Postal Infrastructure and Eco-nomic Development” in M. Crew andP. Kleindorfer (eds) Postal Delivery Services:Pricing, Productivity, Regulation and Strategy.Boston: Kluwer Academic Publishers.

World Bank. 1995. Bureaucrats in Business: TheEconomics and Politics of Government Owner-ship. Washington, DC: World Bank.

Chapter 4: Questioning theMonopoly-Supported PostalUSO in Developing Countries

British Postal Consultancy Service. 2002. “Cost-ing and Financing of Universal Service Obliga-tions: Universal Service Costs and EstimatesReport.” PHARE Project 9728-02-01.

Campbell, J. 2004. “History of Universal Ser-vice in the United States”, presented at the 12th

Conference on Postal and Delivery Economics,June 2-5, 2004, Cork, Ireland.

Cohen, R. et. al. 2003a. “A Comparison of theBurden of Universal Service in Italy and theUnited States.” US Postal Rate Commission.Processed.

Cohen R. et al. 2003b. “Towards a GeneralPostal Service Cost Function.” US Postal RateCommission. Processed.

Kenny, Charles. 2002. “Information and Com-munication Technologies for Poverty Alleviation:Costs and Benefits.” Development Policy Re-view. 20(2).

Ogilvie, A. 1893. “The Rise of the English PostOffice.” The Economic Journal. 3 (11).

PWC (PricewaterhouseCoopers). 2004. “PostalTelecentres: Rural Development ThroughInternet Access is South African DevelopmentCommunity.” Report produced for the WorldBank. Processed.

UPU (Universal Postal Union). 2001. LetterPost Manual. Berne: UPU.

———. 2002. Postal Statistics. Berne: UPU.

UPU and WB. 2001. “The Postal Industry in anInternet Age: Case Studies in Postal Reform.”Washington, DC: UPU and World Bank.

Walsh, T. 2001. “Delivering Economic Devel-opment: Postal Infrastructures and Sectoral Re-form in Developing Countries.” London:Consignia.

World Bank. 2002. “World Development Indi-cators.” Washington, DC: World Bank.

Chapter 5: Postal Policy andRegulatory Reform inDeveloping Countries

Alvero, A. and A. Levy. 2004. “The WTO DohaDevelopment Agenda: Defining the Scope ofPostal Service Liberalization” in Michael A.Crew and Paul Kleindorfer (ed) CompetitiveTransformation of the Postal and Delivery Sec-tor. Boston: Kluwer Academic Publishers.

Guermazi, B. 2000. “Exploring the WTO Refer-ence Paper on Regulatory Principles.” Center ofStudies for Regulated Industries, McGill Univer-sity. <http://www.law.mcgill.ca/institutes/csri/paper-guermazi-reference.php3>.

77

BIBLIOGRAPHY

Harford, G. 2004. “The Liberalization of PostalServices in New Zealand.” Rutgers Conferenceon Postal and Delivery Economics, Cork, Ireland.

TMC Asser Institute. 2004. “The Study of theRelationship Between the Constitution, Rules andPractice of the Universal Postal Union, The WTORules and the European Community Law.” TheHague. Processed.

Volner, Brickman, and Field. 2004. “Postal re-form: A Comparative Analysis and Critique ofRegulatory Models.” Rutgers Conference onPostal and Delivery Economics, Cork, Ireland.

World Bank. 2004. “Reforming Infrastructure:Privatization, Regulation and Competition, AWorld Bank Policy Research Report.” Washing-ton, DC: World Bank.

Chapter 6: Case Studies

ITU (International Telecommunications Union).2002. World Telecommunications DevelopmentReport. Geneva: ITU.

UPU (Universal Postal Union). 2002. Postal Sta-tistics. Berne: UPU.

UPU and World Bank. 2002. “The Postal Indus-try in an Internet Age: Case Studies in PostalReform.” Washington DC: UPU and World Bank.

Frontier Economics. 2003. “Evaluation of PostalSector Reforms: A Synthesis.” London: FrontierEconomics. Processed.

78


Author Biographies

Boutheina Guermazi is a regulatory specialistin the Global Information and CommunicationTechnologies department at the World Bank, in-volved in telecommunications policy and regula-tory reform in a number Africa. Before joiningthe Bank in April 2004, Boutheina was AssistantProfessor at the Faculté des Sciences JuridiquesPolitiques et Sociales of Tunis and a telecommu-nications advisor to the Sector Reform Unit atthe International Telecommunications Union. Shealso worked as strategy analyst for the StrategicUnit at ITU and a telecom regulatory consultantfor the UNDP. Boutheina holds a Doctor De-gree in telecommunications law and policy fromthe Faculty of Law at McGill University (2003),an LLM degree in international law from Indi-ana University and a Bachelor degree in publiclaw from the university of Tunis. She held theFulbright scholarship (95-97) and was a researchscholar at the Social Science Research Council(USA) and the Center of studies for regulatedIndustries (Canada).

Pierre Guislain is the manager of the World BankGroup’s ICT Policy Division. Pierre, a Belgiannational, joined the World Bank in 1983 andworked successively in the West Africa ProjectDepartment, the Legal Department, the AsiaTechnical Department, and the Private SectorAdvisory Services (formerly PSD) Department,before joining GICT, a joint Bank-IFC depart-ment. From 1997 to July 2001, he managed ajoint program between the World Bank and theEuropean Commission on Private Participationin Mediterranean Infrastructure. Pierre is theauthor of several publications on privatizationand infrastructure sector reform. Over the last15 years, he has advised many governments ontelecommunications and postal reform, with em-phasis on issues of market structure, competi-tion, regulation and privatization. Pierre holdsan MPA in Economics and Public Policy fromPrinceton University and a graduate law degreefrom the University of Louvain.

79

AUTHOR BIOGRAPHIES

Charles Kenny is a senior infrastructure econo-mist in the GICT Department of the World BankGroup. He works on a number of projects re-lated to information infrastructure, including co-ordination of the bank’s information infrastruc-ture activities in Afghanistan and Kenya. Pre-viously, Mr. Kenny was an author on the 1999/2000 World Development Report. Mr. Kennyhas MAs from Johns Hopkins School of Ad-vanced International Studies and LondonUniversity’s School of Oriental and AfricanStudies and a BA from Cambridge University.He has published on issues including ICT in de-velopment, economic growth, and the link be-tween growth and happiness.

Graeme Lee is a senior postal policy specialistin the World Bank. His current portfolio of workincludes advising on World Bank postal projectsin Afghanistan, Malawi, Nicaragua and Trinidadand Tobago. Prior to joining the bank in January2004 Graeme was based in Buenos Aires as thestrategic advisor to Correo Argentino, advisingon issues as diverse as the UPU quality of ser-vice fund to costing of universal service. His 18month assignment in Argentina came during a 14year career working for Royal Mail where heworked as a consultant on international assign-ments in central and eastern Europe, Africa andthe Caribbean in the fields of key account man-agement, transport planning, costing systems andpostal reform. Graeme has a BSc (Hons) degreein Transport and Distribution from the Univer-sity of Huddersfield.

Isabelle Segni is a French national who joinedthe Bank in 2001 on a secondment program fromthe Ministry of Economy, Finance and Industry(MEFI). She covers regulatory issues mainly re-lated to postal projects and focuses on the mod-ernization of the postal industry. She spent 5years with the MEFI in the department chargedwith postal and telecommunication regulation andwas then sent to Washington for a 2-year assign-ment with the French Treasury office at the FrenchEmbassy. She holds a B.A. in political sciencesand is graduated from the École NationaleSupérieure des Postes et Télécommunications.

the postal sector in developing and transition countries

Documents