implications of electronic mail and message systems for

Implications of Electronic Mail andMessage Systems for the U.S. Postal Service

August 1982

NTIS order #PB83-265017

—.

Library of Congress Catalog Card Number 82-600599

For sale by the Superintendent of Documents,U.S. Government Printing Office, Washington, D.C. 20402

.

Foreword

This study of the implications of electronic mail and message systems (EMS)for the U.S. Postal Service (USPS) is one of four components of the OTA assess-ment of Societal Impacts of National Information Systems. It should be read alongwith the September 1981 OTA report on Computer-Based National InformationSystems: Technology and Public Policy Issues, which provides the larger contextfor understanding EMS-related public policy questions. Also relevant is the March1982 OTA background paper on Selected Electronic Funds Transfer Issues: Pri-vacy, Security, and Equity.

The House Committee on Post Office and Civil Service, one of the originalrequesting committees, indicated an interest in the possible impact of EMS andelectronic funds transfer (EFT) on the USPS mailstream and labor force, and im-plications for the future role of USPS in electronic mail and message systems.The Senate Committee on Governmental Affairs, Subcommittee on Civil Service,Post Office, and General Services, expressed a similar interest. In response, OTAinitiated a review of relevant EMS and EFT developments since the completionof the work of the Commission on Postal Service in 1977. Several working papersand computer-based models were developed by OTA staff and contractors. Thesewere reviewed by an OTA advisory panel representing postal, labor, telecom-munication carrier, computer service, business, academic, and consumer interests.On the basis of that review, a revised and integrated report was prepared.

OTA appreciates the participation of the advisory panelists, external reviewers,and others who helped bring this study to fruition. It is, however, solely the respon-sibility of OTA, not of those who so ably advised and assisted in its preparation.

Director

. . .I l l

OTA Electronic Message Systems Advisory Panel

Roy AmaraPresidentInstitute for the Future

Paul BaranPresidentCabledata Associates

Harold P. BelcherGeneral Manager for Advanced

MailR&DU.S. Postal Service

Robert L. Chartrand, ex officioSenior Specialist in Information

Technology and PolicyCongressional Research Service

Kent W. ColtonProfessor of Public ManagementBrigham Young University

Donald DunnProfessor of Engineering-

Economic SystemsStanford University

Lawrence J. GervaisFormer Vice PresidentAmerican Postal Workers Union

Maurice W. GreggSenior Vice President for FinanceThe Gap Stores, Inc.

Jeremiah GutmanAttorney at LawLevy, Gutman, Goldberg &

Kaplan

J. Robert HarcharikFormer PresidentTymnet, Inc.

Susan Nycum, ChairwomanAttorney at Law, Gaston Snow & Ely Bartlett

Rob KlingProfessor of Information and

Computer ScienceUniversity of California at Irvine

John C. LeGates*Director, Information Resources

Policy ProgramHarvard University

Henry LucasChairman and Professor of

Computer Applications andInformation Systems

New York University

Daniel McCrackenPast PresidentAssociation for Computing

Machinery

Arthur S. MillerProfessor of Law EmeritusThe George Washington

University

Sharon NelsonAttorney at LawConsumers Union (formerly)

Phil NyborgFormer General CounselComputer and Communications

Industry Association

Russell PipePresidentTranslational Data Reporting

Service, Inc.

Ithiel De Sola PoolProfessor of Political ScienceMassachusetts Institute of

Technology

James RuleProfessor of SociologyState University of New York at

Stony Brook

Richard M. SchmidtGeneral CounselAmerican Society of Newspaper

Editors

Sherry TurkleProgram in Science, Technology,

and SocietyMassachusetts Institute of

Technology

Ron UhligManager, Business and Services

PlanningBell Northern Research

Laboratory

Jacques ValleePresidentInfomedia

Carl VorlanderExecutive DirectorNational Association for State

Information Systems

Fred W. WardenFormer Director,

Telecommunication RelationsIBM Corp.

Willis WareSenior Computer SpecialistThe Rand Corp.

Iram WeinsteinSystem Planning Corp.

Alan F. WestinProfessor of Political ScienceColumbia University

*[{esigned in late 1979.

iv

OTA Electronic Message Systems Project Staff

John Andelin, Assistant Director, OTAScience, Information, and Natural Resources Division

Sam Hale, Interim Program ManagerCommunication and Information Technologies*

Fred B. Wood, Project Director**

Jean E. Smith, Study Coordinator

Zalman A. Shaven, Senior Analyst

Fred W. Weingarten, Senior Analyst

Elizabeth A. Emanuel, Administrative Assistant

Shirley A. Gayheart, Secretary

Jeanette Contee, Wordprocessor

Contractors

Charles C. Joyce, Jr., and Jeffrey Held,Richard L. Deal & Associates, Inc.

Renee G. Ford, Tifford Producers, Ltd.Keith Geeslin, Walter Hinchman Associates, Inc.Marvin A. Sirbu, Jr., Massachusetts Institute of TechnologyMichael F. Cavanagh, Cavanagh AssociatesRobert Anthony, Mitre Corp.Robert Johansen, Institute for the FutureHarley W. Radin, Consultant

OTA Publishing Staff

John C. Holmes, Publishing Officer

John Bergling Kathie S. Boss Debra M. Datcher Joe Henson

*March 1981 through February 1982; Stephen E. Doyle served as Program Manager through February 1981; Fred W. Weingarten was appointedProgram Manager effective March 1982.

**since June 1979; Rumn E. Pengov served as Project Director through May 1979.

v

Acknowledgments

The following individuals also contributed as reviewers or provided assistance during the course ofthis study.

William F. Bolger, The Postmaster General, andstaff members of the U.S. Postal Service

Ronald G. Carr, U.S. Department of JusticeKathleen Casey, GTE-Telenet Communications

Corp.Henry B. Freedman, Policy Studies Corp.Dean Gillette, Bell Telephone Labs, Inc.Henry H. Hitchcock, J. F. Coates, Inc.Herbert N. Jasper, Ad Hoc Committee for

Competitive Telecommunications

John F. McLaughlin, Harvard UniversityArnim H. Meyburg, Cornell UniversityRoger K. Salaman, National Telecommunications

and Information AdministrationStephen L. Sharfman, Postal Rate CommissionJack D. Smith, Federal Communications

CommissionScott K. Socol, Satellite Business SystemsFrederic Withington, Arthur D. Little, Inc.

vi

Contents

Chapter Page

1.

2.

3.

4.

5.

6.

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix

Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

Background and Purpose of Study. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

Market Penetration Model and Technology Assumptions . . . . . . . . . . . . 23

Market Penetration Results . . . . .,’ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

Revenue/Cost Model and Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51

Implications for Postal Rates, Service Levels, andLabor Requirements.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61

7. Implications for the Telecommunication and Computer Industries,EMS Privacy and Security, and USPS Long-Term Viability . . . . . . . . . 73

8. Congressional Policy Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

Appendix PageA. Supplemental Detail on the Market Penetration Model . . . . . . . . . . . . . 101B. Logistic Substitution Process. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105C. Equation Used to Calculate Labor Requirements . . . . . . . . . . . . . . . . . . 107D. E-COM Interconnection Arrangements . . . . . . . . . . . . . . . . . . . . . . . . . . . 108

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

vii

OverviewAdvances in communication and computer technology provide new ways to

convey messages and carry out financial transactions. These are called electronicmail and message systems (EMS) and electronic funds transfer (EFT) systems.Commercially offered EMS and EFT will increasingly compete with portions ofthe traditional market of the U.S. Postal Service (USPS). While there is disagree-ment on how fast EMS and EFT may develop, it seems clear that two-thirds ormore of the current mainstream could be handled electronically, and that the volumeof USPS-delivered mail is likely to peak in the next 10 years. Any decline in thevolume of mail has significant implications for future postal rates, USPS servicelevels, and labor requirements.

A key policy issue requiring congressional attention is how USPS will par-ticipate in the provision of EMS services, both in the near term and in the longerterm. If USPS does not attract and keep a sizable share of the so-called Genera-tion II EMS market (electronic input and transmission with hardcopy output)and conventional (especially first-class) mail volume declines, USPS revenues willprobably go down, with the likelihood of an unfavorable impact on rates and/orservice levels. If USPS does develop a major role in the Generation II EMS market,and if Generation II EMS costs are low enough, the effect on USPS rates and/orservice could be favorable.

Two of the major factors influencing USPS labor demand are total mail volumeand worker productivity. Regardless of whether USPS participates in Genera-tion II EMS, improving worker productivity and eventually declining conventionalmail volumes could lead to considerably lower labor demands in the future. If USPSdoes participate successfully in Generation II EMS, this potential decline in laborrequirements could be deferred or partially offset.

USPS is already involved in EMS to a limited extent. For example, it deliverssome industry EMS hardcopy, provides a portion of Western Union’s Mailgramservice, and in January 1982 introduced a domestic service called “electronic com-puter-originated mail” or E-COM. (In E-COM, USPS accepts letters in electronicform, converts them to hardcopy, including printing and enveloping, and deliversthem.) The role of USPS in EMS activities is already controversial, and it is like-ly to become more so if its role expands.

As a result of technological advances, historical (and legal and regulatory)distinctions between conventional and electronic mail have blurred, along withthe application of the the congressional mandates embodied in the Postal and Com-munications acts. Absent congressional action to provide a clear direction for USPSand to clarify or redefine regulatory boundaries, the controversy over the USPSrole in EMS is likely to continue indefinitely and Generation II EMS opportunitiesfor USPS, private telecommunication carriers, and mailers may be lost.

OTA identified a number of areas related to USPS participation in EMS thatwarrant congressional consideration:

● Potential Contribution to USPS Mail Volume.—A USPS role in Genera-tion II EMS has the potential to provide a volume “cushion” to partiallyoffset reductions in conventional mail. Since private firms are neither will-ing nor able to duplicate the nationwide physical delivery infrastructureof USPS, any large-scale Generation II EMS service depends on the par-ticipation of USPS. There is, however, little consensus on what USPS role

ix

would be most conducive to growth of Generation II traffic (and hence USPSmail volume).

● Potential Contribution to USPS Finances.— While Mailgram apparentlyprovides both a substantial avoidance of conventional USPS mailstreamcosts and a significant contribution to covering USPS fixed costs, it is notclear whether E-COM would do likewise at current rates and in its presentconfiguration. All parties, including USPS, agree that the original E-COMcost estimates prepared in 1978 for the Postal Rate Commission are nowoutdated. A comprehensive cost review of E-COM is needed if its contribu-tion to USPS finances is to be understood.

. Impact on USPS Labor Force.—USPS labor force requirements are sensi-tive to mail volume. Any significant decline in mail volume for whateverreasons (e.g., diversion to EMS and EFT, competition from nonelectronicalternative delivery services, reduced demand due to general economic reces-sion) would translate into labor force reductions beyond those that maybe needed as a result of higher worker productivity.

USPS participation in Generation II EMS would have two effects onUSPS labor requirements. A small number of new jobs would be createdto carry out EMS-related activities, and a considerably larger number ofjobs would be required in traditional USPS activities to process and deliverthe resulting hardcopy. Thus, USPS participation could offset, or at leastdefer, some of the reductions in the existing labor force that might other-wise be necessary. The need for sizable reductions from the present levelof employees is not likely to be felt until the late 1980’s or early 1990’s.

● Size and Nature of the E-COM Market.--OTA concluded that priorestimates of the Generation II market have probably been high, and thatprudent planning should be based on a mature market (20 years hence) inthe range of 7 billion to 17 billion messages annually, rather than 25 billionas previously estimated by RCA (for comparison, conventional mail in 1981totaled 110 billion pieces). Even the lower estimate depends on mailer ac-ceptance and successful institutional marketing strategies. Several modifica-tions to E-COM have been proposed that deserve consideration, such as1-day guaranteed delivery rather than 2-day, flexible letter formats, andthe use of telecommunication carrier and mailer logos to personalize theE-COM output and provide incentives for aggressive marketing. A fullreview of the E-COM market should include governmental as well as privatesector mailing needs.

. Impact on Competition.—USPS believes its participation in E-COM is au-thorized by the Postal Reorganization Act mandate to use new facilitiesand equipment to improve the convenience, efficiency, and cost effectivenessof mail service. However, various telecommunication and computer firmsview E-COM as: 1) the entry of a Federal agency into competition with pri-vate industry; 2) possibly subject to the Communications Act as well asthe Postal Act; and 3) raising questions about the fairness and legality ofa USPS role in EMS in general. The applicability of the Private ExpressStatutes to delivery of Generation II EMS hardcopy output has also beenchallenged.

● Regulatory Jurisdiction.—The E-COM case has surfaced disagreements overthe division and extent of regulatory jurisdiction by the Postal Rate Com-mission and Federal Communications Commission. As a result, USPSbrought suit against both commissions and the Department of Justicebrought suit against USPS–all on jurisdictional grounds, not on the meritsor faults of E-COM. Some private firms believe this demonstrates the dif-

X

ficulty of regulating USPS ratesetting and preventing the potential cross-subsidization of E-COM from conventional mail revenues. One proposedresponse to that concern would be to require a separate USPS entity forany EMS offering.

● E-COM Privacy and Security. –The involvement of USPS in E-COM haspointed out the need for additional security measures to protect the privacyof EMS messages. The Postal and/or Communications Act may need tobe amended to provide additional statutory privacy protection.

● Role in Telecommunication and/or Generation III EMS.—The USPS rolein E-COM does not involve either telecommunication or electronic delivery.Whether there are conditions that would constitute demonstrated need forUSPS to contract with a telecommunication carrier to transmit messageselectronically on behalf of USPS needs to be clarified. Further study seemswarranted on the possible use of EMS Generation III (as well as Genera-tion II) to help USPS maintain adequate service levels to rural and lesspopulated areas and to low volume, nonprofit, and educational mailers(E-COM is currently designed for high-volume business mailers.)

● Increased Cooperation with the Private Sector.—At present, it is difficultfor USPS to conduct effective long-range planning and market testing ofEMS, since this requires good working relationships with private telecom-munication and computer firms. If some clearer consensus can be reachedon the direction and limits of USPS involvement in EMS, perhaps a moreconstructive relationship with the private sector can develop.

OTA’s analysis suggests that advances in technology and increased compe-tition in the communications marketplace will significantly affect USPS finances,service levels, and labor force requirements over the next two decades. It furthersuggests that modification or clarification of the USPS role in EMS can, in turn,help determine how effectively USPS accommodates to these changes. Given thedifficulty of modifying institutions as large and complex as USPS and the lawsand regulations that govern USPS actions, it would seem prudent for Congressand USPS to address these issues aggressively. Changes are taking place so fastin the so-called “communications revolution” that by the time USPS actually ex-perienced significant reductions in conventional mail volume, most opportunitiesfor participation in EMS would have passed and it would be much more difficultto adjust.

x i

Chapter 1

Summary

Contents

—

Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

USPS Mainstream . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

USPS Rates and Service Levels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Labor Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Policy Implications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Provide a Clear Direction for USPS Involvement it ems. . . . . . . . . . .Reduce or Eliminate Further Regulatory and Judicial Delay . . . . . . . . .Strengthen Privacy and security protection . . . . . . . . . . . . . . . . . . . . . . .Maintain Oversight and Initiate Planning on USPS Long-Term

Viability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Page3

4

5

7

89

1010

11

Chapter 1

Summary

IntroductionThis study addresses three major questions:

1.

2.

3.

To what extent are privately offered elec-tronic mail and message systems (EMS)and electronic funds transfer (EFT) sys-tems likely to affect the volume of mailhandled by the U.S. Postal Service(USPS)?Are changes in USPS mail volume likelyto lead to significant adjustments inUSPS rates, service levels, and/or laborforce requirements? andWhat are the implications of EMS for thefuture of USPS and how it might partici-pate in the provision of EMS services?

These questions are of concern because his-torically USPS has served a variety of socialpurposes mandated by Congress, such as “tobind the Nation together through the personal,educational, literary, and business corres-pondence of the people” and to provide mailservice “to (postal) patrons in all areas . . . andall communities, ” including rural areas, com-munities, and small towns (Postal Reorganiza-tion Act of 1970). In recent years, there hasbeen a continuing revolution in computer andcommunication technology, a gradual dereg-ulation of the telecommunication industry (thecomputer industry being essentially unregu-lated), and a proliferation of new and old firmsoffering or planning to offer EMS and EFTservices that compete with portions of the tra-ditional USPS market. Technology is, in ef-fect, blurring the historical (and legal andregulatory) distinctions between conventionaland electronic mail.

USPS is already involved in EMS to a lim-ited extent. It currently delivers some in-dustry EMS hardcopy output, provides a por-tion of Western Union’s Mailgram service, andin January 1982 introduced a domestic servicecalled “electronic computer-originated mail”or E-COM. However, the role of USPS in EMS

in general, and in E-COM in particular, con-tinues to be in dispute before various regula-tory agencies, the courts, and Congress. USPSbelieves its participation in EMS is authorizedby the Postal Reorganization Act mandate touse new facilities and equipment to improvethe convenience, efficiency, and cost effective-ness of mail service. Various mailer organiza-tions, consumer groups, and postal laborunions see a USPS role in EMS as essentialto USPS long-term viability and to maintain-ing, or at least minimizing any reductions in,mail services that are vital to a large part ofthe U.S. population. They point to the criticalrole of USPS in providing a universal, low-cost, nondiscriminatory nationwide commu-nication service. Various private telecommu-nication and computer firms view USPS in-volvement in EMS as the entry of a Federalagency into competition with private industry(possibly subject to the Communications Actas well as the Postal Act), raising difficultquestions of ratesetting and potential cross-subsidy.

There are no easy answers to the questionsaddressed in this study. Prior studies haveproven to be oversimplified. In order to bet-ter understand the complexities involved,OTA used computer-based models to projectindependently the levels of conventional andelectronic mail volumes under different setsof assumptions, and to project the possible ef-fects of changes in USPS mail volumes onUSPS rates, service levels, and labor re-quirements. Still, while computer modelingpermits consideration of a larger number ofvariables and interrelationships than wouldotherwise be possible, the precision of the pro-jections can be misleading. The models arehighly sensitive to initial assumptions andhave limited ability to anticipate unexpectedevents. The study as a whole, and the use ofcomputer modeling in particular, is intended

3

- -——

4 ● Implications of Electronic Mail and Message Systems for the U.S. Postal Service

to help Congress better understand the possi- intended to makeble implications of EMS for USPS, and is not course of events.

USPS Mainstream

a prediction of the future

To a substantial extent, the volume ofUSPS-delivered mail in the future is beyondthe direct control of USPS. It will be affectedby diversion to electronic modes of “written”communication (EMS and EFT), by overalleconomic factors, and by competition from pri-vate message and parcel delivery systems.Taken together, it seems clear that two-thirdsor more of the mainstream could be handledelectronically, and that the volume of mail islikely to peak in the next 10 years and fallbelow today’s level sometime in the 1990’s.

EMS penetration of the mainstream will bepaced by the introduction and widespread useof technologies such as high-quality electronicprinters, office automation, home computerterminals, viewdata/teletext, and inexpensivehome hardcopy terminals. EFT will be pacedby the increased use of automated teller ma-chines and point-of-sale terminals, and con-solidation of bills and payments through tel-ephone bill payer, debit cards, and direct de-posit. In the long run, both EMS and EFT arelikely to be possible over the same electronicterminals and communication networks. Butin the shorter term, they are separate technol-ogies. EMS itself has two distinguishablemodes-Generation II EMS (electronic inputand transmission with hard copy output) andGeneration III EMS (all-electronic). This di-version of mail from conventional paper-basedto electronic form is likely to stretch overmany years and probably decades, dependingon the rate of technological advance, on futurepostal rates, on regulatory constraints, and inpart on intangible factors such as consumeracceptance and institutional marketing strat-egies.

OTA made several estimates of the rate ofdiversion of conventional mail to EMS andEFT. Mail diverted to Generation III EMSand EFT was assumed lost to the USPS mail-

stream. Because of the need to sort and deliverthe hardcopy output from Generation II EMS,mail diverted to Generation II EMS was as-sumed to remain in the USPS mainstream.Mailgram and E-COM are Generation II serv-ices. In all of 1980, about 40 million Mailgramswere sent; in the first half-year of its existence(January through June 1982), about 660,000E-COM messages were sent. During July1982, E-COM averaged about 172,000 mes-sages weekly. USPS also delivers an unknown,but small, number of letters that represent thehardcopy output of private sector GenerationII EMS.

OTA did not independently estimate overallfuture economic growth or competition fromprivate delivery services; representative pastgrowth rates of the USPS mainstream wereprojected into the future. For example, theUSPS mainstream has grown by about 2 per-cent per year if averaged over the 1900-77period, and 3 percent over the 1947-77 period.In 3 of the last 4 years, the growth rate ex-ceeded 3 percent. However, between 1971 and1976, the growth rate was only about 1 per-cent.

If the recent 3 percent growth rate held forthe next 20 years, USPS conventional mailwould exceed the 1981 level (110 billion pieces)until the mid-1990’s, even in the face of com-petition from high but plausible EMS andEFT growth. Assuming that USPS deliversthe hardcopy output from Generation II EMSin that timeframe, USPS total deliveries wouldexceed the 1981 level until the turn of the cen-tury. Similarly, at 2-percent annual growth,and the same conditions, USPS conventionalmail volume would exceed the 1981 level untilabout 1990, and total mail until themid-1990’s.

OTA conducted several sensitivity analysesto determine the conditions under which the

.

Ch. 1—Summary ● 5

USPS mainstream might decline even faster.OTA concluded that the most likely conditionwould be continued economic recession or de-pression, which in the past (early 1970’s andearly 1930’s) has resulted in a flat or evennegative mailstream growth. This, coupledwith even faster than anticipated introductionof all-electronic Generation III EMS and EFT,or escalation of USPS costs and rates abovethe level of alternative conventional and elec-tronic delivery services, would accelerate thedecline of the USPS mainstream.

Regardless of the underlying mainstreamgrowth, the effect of Generation II EMS vol-ume is to “cushion” or offset some of thedecline in conventional mail, assuming USPSdelivers the Generation II hardcopy output.Put differently, if Generation II volumereached significant levels, USPS-deliveredmail volume (conventional plus Generation II)might be maintained at or above a given levelfor an additional 5 years or so.

For planning purposes, it is reasonable toassume that mail volume is likely to remainstrong for most of the 1980’s, and declinesignificantly in the 1990’s. Under any plausi-ble scenario, USPS is still likely to be handlinga large volume (70 billion to 110 billion pieces)of mail in 2000.

Independent of the total USPS mainstream,the size of the potential Generation II EMSmarket itself takes on considerable importancewith respect to decisions concerning USPS in-volvement. OTA has concluded that prior mar-ket estimates probably have been high. For ex-ample, RCA Corp. previously estimated a ma-ture Generation II market (20 years hence) of

25 billion pieces. However, this exceeds evenOTA’s high but plausible Generation II pro-jection by roughly 40 to 80 percent, dependingon the underlying growth in demand for mail.It appears that RCA was overly optimisticabout Generation II market development, ig-nored competition with Generation III EMSservices, or both. OTA also identified a slow-growth Generation II path that projects a vol-ume of about 40 million messages 5 yearshence, increasing to about 600 million mes-sages after 10 years and around 3 billion after15 years. Given the highly volatile and unpre-dictable nature of the EMS market, it appearsthat prudent planning should be based on amaximum of one-half of the RCA-projectedvolume on down to the OTA-projected volumefor Generation II slow growth. This wouldplace the projected Generation II market 20years hence in the range of 7 billion to 17 bil-lion pieces rather than 25 billion.

There are legitimate differences of opinionas to how Generation II would fare after 2000,which was beyond the timeframe of the OTAstudy. Some analysts believe that GenerationII would taper off very slowly and remain sig-nificant for many years. Others are convincedthat Generation II might decline rather pre-cipitously. However, it is likely that Genera-tion III would surpass Generation II in ab-solute volume well before 2000. Indeed, unlessGeneration II grows at a high or very highrate, it is possible that Generation II wouldnever exceed Generation III. Various privatetelecommunication carriers have indicatedthat most research and development (R&D)and marketing effort is going into GenerationIII, not Generation II.

USPS Rates and Service LevelsUSPS has some control over the way in

which changes in mail volume might be re-flected in rate and service level adjustments.While USPS is not allowed to make a profitoverall, it need not “markup” all classes andsubclasses of mail by the same amount over

the costs specifically allocatable to each (“var-iable” costs). Thus, individual classes and sub-classes make varying contributions to cover-ing common (“fixed”) costs. For example, first-class mail, with high volume and relativelyhigh markup, historically has made the largest

97-918 0 - 82 - 2


contribution to fixed costs of any class of mail.In fiscal year 1980, first-class mail made upabout 57 percent of the total mainstream. Itscontribution to USPS fixed costs was about$4.2 billion, based on an actual volume of 60billion pieces (and assuming 20¢/piece revenueand 13¢/piece variable cost). This was about55 percent of total fixed costs ($7.6 billion) inthat year.

The example of first-class mail just cited isparticularly relevant to this study because itis likely that most of the conventional mail di-verted to EMS or EFT over the next 20 yearswill be first-class mail. Not only does it repre-sent the largest volume of mail, but it is gen-erally more amenable to electronic transmis-sion than are other classes. Three-quarters offirst-class mail is made up of correspondence,negotiable instruments (e.g., checks), and billsand financial statements.

If the first-class mainstream declined 10 bil-lion pieces by the year 2000, and the 1980 costand markup is assumed, first-class mail wouldcontribute $700 million less to USPS fixedcosts than it does today. Making up that lossby raising first-class rates alone would requirea 7-percent increase. To the extent that USPSwas delivering Generation II EMS hardcopyand making a profit on it, some or all of thisrate increase could be avoided. This latter pos-sibility is highly dependent on the cost of Gen-eration II EMS (not well known) and the mar-ket price (also not well known).

Cost cutting would ease the necessity of in-creasing rates. This strategy would be par-ticularly prudent if first-class rate increaseswould lead to a further volume reduction,which could set off a spiral of rate increasesand volume reductions. Whether or not cut-ting costs would result in service cuts woulddepend in part on the USPS cost structure.Under the current cost structure (36 percentfixed costs, 64 percent variable), some servicecuts would appear to be necessary. For exam-ple, USPS officials have estimated that deliv-ery 5 days a week (instead of 6) would save

about $650 million (1980 dollars). A 1976USPS staff study projected a $1.1 billion (1977dollars) savings for delivery 3 days a week.

In the longer term, USPS fixed costs maybe reduced below the current 36 percent; otherways to improve productivity might be iden-tified such that service levels could be main-tained even at lower volumes and revenues.However, some significant portion of USPScosts clearly is required to pay for maintain-ing the basic nationwide delivery system andinfrastructure, largely irrespective of thevolume. For example, a substantial number ofcarriers are required to cover the approximately 69 million city USPS delivery points and15 million rural delivery points (as of 1980)each day, 6 days a week, and to maintain win-dow service at over 30,000 post offices, 9 hoursor more a day, 5 or 5½ days a week. Likewise,some minimum level of truck transportationbetween post offices must be maintained tomeet delivery performance standards, regard-less of whether the trucks are carrying severaldozen or several thousand letters. Labor cur-rently accounts for about 85 percent of USPScosts and transportation about 7 percent(largely for trucks and other postal vehicles,plane and rail transportation). Thus, from thisperspective, it is not clear that fixed costscould be reduced substantially without cuttingservice levels.

In sum, any projections of USPS revenues,costs, and rates are difficult at best, given:

1.

2.

3.

4.

5.

the complexity of USPS revenue and costrelationships;the fact that costs obviously vary by typeof mail route (e.g., urban, suburban, rural),although USPS does not collect such costdata;the problem of how to assign commoncosts properly to different mail services;the uncertainty in determining what costsare variable with volume changes overvarious time periods; andthe uncertainty over future costs, rates,and volumes of Generation II EMS.


Labor Requirements

USPS is a labor-intensive organization, withlabor representing about 85 percent of totalUSPS costs. By making assumptions aboutthe kind of service provided by USPS, the divi-sion of labor among its employees, and theirproductivity, it is possible to estimate thelabor requirements for projected future mailvolumes.

For the purposes of this labor force analysis,OTA assumed that the kind of service pro-vided by USPS (service levels) would remainconstant. That is, there would be no changein the number of delivery days (e.g., 6 days aweek), post offices (over 30,000), weekday win-dow service hours (typically 9 hours), or cityand rural delivery points (over 84 million).

OTA calculated the division of labor amongUSPS employees by assuming that each func-tional group of employees would allocate theirtime in the future the same as they do now.That is, each employee group spends a certainfraction of its time in activities that are in-dependent of the volume of mail (“fixedlabor”); most groups also spend time in ac-tivities that do vary directly with the volumeof mail (“variable labor”).

OTA estimated variable and fixed percent-ages for each major group of USPS employeesfrom the USPS fiscal year 1980 Revenue andCost Analysis. The variable labor percentageis based on the variable attributable cost fromthe USPS analysis; the fixed labor percentageon the sum of specific fixed attributable costsplus all other institutional costs for eachemployee group. The variable and fixed laborpercentages determined by OTA were re-viewed with USPS and found to be reasonable.

The overall cost split for the entire USPSlabor force was calculated to be 61 percentvariable and 39 percent fixed. Individualemployee groups varied from those independ-ent of USPS mail volume to those almost com-pletely dependent on it. As of 1980, 14,268employees, or about 2.1 percent of the totalUSPS labor force, had no activities that varied

with mail volume. Included would be head-quarters, regional, and inspection serviceemployees. On the other hand, almost all ofthe activities carried out by clerks and mailhandlers varied with mail volume (86 percentvariable/14 percent fixed). In fiscal year 1980,this group included 303,560 full- and part-timeemployees, or about 45 percent of the total1980 USPS work force of 667,000 employees.The clerks and mail handlers would have lim-ited participation in a Generation II EMSservice, which would bypass many of the tradi-tional mail sorting and processing functionsperformed by clerks and handlers. The hard-copy output of Generation II EMS would stillrequire physical delivery by city or ruralcarriers.

Most other employee groups fall in between,having some activities that vary with totalUSPS-delivered mail volume (conventionalplus Generation II EMS) and some that donot, but with a larger fixed component thanclerks and handlers. These would include su-pervisory and technical personnel (48 percentfixed), city delivery carriers (50 percent fixed),maintenance personnel (about 55 percentfixed), and rural delivery carriers (73 percentfixed).

With respect to productivity, OTA assumedan average labor productivity improvement of1.5 percent per year as a baseline. For com-parison, USPS labor productivity is creditedas increasing by roughly 3 percent annuallyduring the 1970’s, as measured by the numberof pieces of mail per workyear. In fiscal year1970, 741,000 postal employees delivered 85billion pieces of mail, while in fiscal year 1980,667,000 employees delivered about 106 billionpieces. The 3-percent figure may not reflecttrue labor productivity since service levels didchange. For example, mailbox collection fre-quencies (per day) were generally reduced andcluster boxes were substituted for home de-livery in many new suburban residential de-velopments. Still, between 1971 and 1980 thenumber of city delivery points increased by


about 21 percent and the number of rural de-livery points increased by about 50 percent.Even assuming that the 3-percent annual aver-age during the 1970’s is an accurate measure,this does not appear to be a realistic expecta-tion for the 1980’s in view of the fact that mostproductivity improvement from automationand mechanization has already been realized.Even the expanded ZIP code program, knownas ZIP + 4, would realize a total labor pro-ductivity improvement of only 2.3 percent, ac-cording to USPS estimates.

Taking these variables together, and assum-ing high but plausible EMS growth, OTA con-cluded that the USPS labor force requirementin 2000 is most likely to be roughly 20 to 25percent below the 1980 level. This result is pro-jected for the base case of 1.5-percent annuallabor productivity improvement and 2-percentannual underlying mainstream growth, andalso for the cases of 3-percent productivity im-provement/3-percent mailstream growth, andO-percent productivity improvement/1-percentmainstream growth. Under all three of thesescenarios, the need for significant labor forcereductions is not likely to be felt until the late1980’s and early 1990’s, but would increasequite rapidly thereafter.

Whatever the level of reductions, they arenot likely to be spread evenly among all em-

ployee groups. The post office clerks and mailhandlers group would be expected to be hit thehardest, losing perhaps two-fifths of their 1980complement by 2000. Post office supervisorsand city delivery carriers could, by 2000, bereduced by about one-fifth and rural deliverycarriers by about one-tenth of their 1980 com-plements.

Whether or not these labor force reductionscould be handled through attrition dependslargely on future USPS retirement, quitting,and new hire rates. But the uneven impact ofreductions on various employee groups makesthis unlikely. In addition, the uneven distribu-tion of minority employment among employeegroups raises the possibility that such reduc-tions may fall disproportionately on black andperhaps other minority employment. For ex-ample, as of late 1978, the mail handlers,whose employment would be reduced themost, had one of the highest percentages ofblack employment. Involuntary labor force re-ductions in this area, if needed, would likelyraise some difficult social and political issues.

Overall, the impact of labor force reductionson promotion opportunities, upward mobility,employee morale, and union contract negotia-tions could be significant. These areas warrantfurther study.

Policy Implications

The OTA analysis indicates that, regardlessof what role USPS plays in Generation II elec-tronic mail, reductions in USPS-delivered mailvolume due to diversion to Generation IIIEMS and EFT could reach significant levelsby 2000. The threat to USPS-delivered mailcould come even sooner if Generation III EMSservices (all-electronic) develop faster than cur-rently anticipated, if the underlying growthin the mainstream is less than the historicalaverage, or if diversion of second- and third-class mail to alternative (nonelectronic) deliv-ery services increases significantly beyond cur-rent levels.

Although a USPS role in Generation IIEMS has the potential to provide a volumeand revenue “cushion” to partially offset re-ductions in conventional mail volume and rev-enue, there is little consensus among USPSand major stakeholders on exactly what theUSPS role should be in the provision of Gen-eration II EMS.

The market penetration results indicate thatUSPS-delivered mail volume (conventionalmail plus Generation II EMS hardcopy out-put) is one key factor in considering a USPSrole in Generation II EMS. USPS-deliveredvolume is a function in part of the rate of

.


Generation II EMS growth. Assuming thatUSPS delivers Generation II EMS hardcopyoutput, the faster the rate of growth (and theearlier the takeoff), the larger the GenerationII EMS volume and USPS-delivered volume.There is currently little agreement on whichUSPS role would contribute the most to Gen-eration II EMS growth and volume.

The revenue/cost results indicate that Gen-eration II EMS cost displacement and con-tribution to covering USPS fixed costs are alsokey factors in considering a USPS role. Thegreater the cost displacement (avoidance ofconventional mainstream costs) and contribu-tion to covering USPS overhead, the less likelythe need for service (and/or labor) reductions.Mailgram apparently provides both a substan-tial cost displacement and contribution tofixed costs; it is not clear whether E-COMwould do likewise at current rates and in itspresent configuration. All parties, includingUSPS, agree that the RCA cost estimates pre-pared for the Electronic Message Service Sys-tem in 1977 and the original E-COM cost es-timates prepared for the Postal Rate Commis-sion in 1978 are now outdated. If E-COM isto be fully evaluated and its role in USPS’future understood, a comprehensive cost re-view of E-COM is needed.

In contrast, there is general agreement thatUSPS participation in Generation II EMSwould generate only a relatively small numberof new jobs. An estimated 200 persons (125operations, 50 maintenance, 25 marketing andadministrative) currently work on E-COM. Afully deployed service (at 150 serving post of-fices (SPOs) rather than the current 25) is es-timated to require perhaps 2,000 persons. Theadditional volume from USPS delivery of Gen-eration II EMS hardcopy output could helpto offset some of the reductions in the existinglabor force that will be necessary if the pro-jected decline in USPS-delivered mail materi-alizes.

Based on interviews with many of the stake-holders and a comprehensive review of the his-torical record, OTA has concluded that absentcongressional action, the controversy over the

USPS role in EMS is likely to continue indef-initely. The fairness and legality of a USPSrole in EMS, the impact on innovation andcompetition in the EMS industry, and implica-tions for EMS privacy and security continueto be in dispute. Although the U.S. DistrictCourt of Appeals has denied a Department ofJustice petition to block E-COM, further reg-ulatory proceedings are anticipated and addi-tional legal actions are possible.

With continuing uncertainty over the futureof E-COM, and in general over the USPS rolein EMS, the prospects for a successful USPSentry into domestic EMS services are uncer-tain. This affects both USPS and its potentialcompetitors in the private sector. Some firmshave indicated to OTA that they are reluctantto make any major commitments until theyare certain what role USPS is going to have.Meanwhile, most private sector R&D effortsare going into Generation III EMS, whichwould completely bypass USPS. In addition,given the continuing adversarial atmosphere,USPS is unable to establish effective workingrelationships with many private carriers andpotential Generation II EMS users.

Should Congress wish to take action, thereare several possibilities: 1) provide a cleardirection for USPS involvement in EMS; 2) re-duce or eliminate further regulatory and ju-dicial delay; 3) strengthen privacy and securityprotection; and 4) maintain oversight and in-itiate planning on USPS long-term viability.These possibilities are outlined below.

Provide a Clear Direction forUSPS Involvement in EMS

There is a range of alternatives for a USPSrole in EMS: .

1. no real involvement other than delivery

2

3

of Generation II EMS hardcopy outputwhen deposited into the mainstream;delivery of all hardcopy output when con-veyed over postal roads;hardcopy delivery plus location ofcarrier-provided EMS terminal equip-ment on USPS premises (as in Mailgram);

—


4. the current E-COM role or variations ofit (e.g., the use of logo envelopes to retaincarrier and mailer identities);

5. involvement in the telecommunicationportion of EMS as well-as printing, envel-oping, and delivery; and

6. involvement in Generation III EMS aswell (e.g., through lease or contract withprivate industry).

Each of these alternatives is technicallyfeasible. In evaluating each, Congress maywish to take into account some or all of thefollowing broad considerations:

●

●

●

●

●

●

●

the extent to which each alternativewould contribute to Generation II EMSgrowth and volume;the extent to which each alternativewould favorably affect USPS finances,i.e., EMS cost displacement and contribu-tion to covering USPS fixed costs;the extent to which each alternative,through creating new jobs and increasingUSPS mail volumes, would defer or par-tially offset labor reductions that mightotherwise be necessary;the extent to which each alternativewould provide incentives for marketingby USPS and/or private firms;whether changes in the Postal Reorgani-zation Act are needed to permit more flex-ibility in the USPS decisionmaking proc-ess (including regulatory review) with re-spect to USPS involvement in EMS;whether the term “demonstrated need”for USPS to contract with a telecommu-nication carrier to transmit messages elec-tronically on behalf of USPS needs to beclarified; andwhether or not there are any conditionsthat would constitute demonstrated needfor USPS involvement in electronic deliv-ery (presumably by contract with privateGeneration III EMS firms); for example,in geographic areas where conventionalmail service could no longer be main-tained at present levels.

With regard to E-COM itself, Congress maywish to review the following specific issues:

●

●

●

●

whether or not space should be providedin SPOs for carrier output equipment;whether or not the technology selected byRCA for E-COM is the best available;whether technical modifications to thecurrent E-COM interconnection arrange-ment could permit more total lines (andat what cost) for carrier and user access,and whether alternative access allocationschemes should be considered; andwhether E-COM performance standardsand design should be modified to guaran-tee 1-day delivery.

Reduce or Eliminate FurtherRegulatory and Judicial Delay

The most important action Congress cantake to reduce delay is to provide clear direc-tion for USPS involvement in EMS. A noteof caution is in order. If the direction set outis not well understood and does not reflect asubstantial consensus, further regulatory dis-putes and litigation could result.

Additionally, Congress could:●

●

●

clarify the applicability of the Private Ex-press Statutes to delivery of GenerationII EMS hardcopy output;delineate the division of regulatory juris-diction between the Postal Rate Commis-sion and the Federal CommunicationsCommission; anddecide on the desirability of a separateUSPS entity for any EMS offering.

Strengthen Privacy andSecurity Protection

Privacy protection in a USPS EMS serviceis a continuing issue. Preliminary discussionswith USPS indicate that while the E-COMequipment is apparently physically secure, thepotential for security breaches does exist. Useraccount numbers are visible on the outside ofE-COM envelopes. When combined with an ac-cess code and familiarization with the E-COMtechnical interconnection standards, this in-formation would permit unauthorized use of

Ch. 1—Summary ● 1 1

E-COM. Incoming messages are stored for 1week in computer memory or on magnetictape, providing another target for securityviolations. These archived messages could alsobe tapped via the management informationsystem, since the E-COM computers in the 25centers are connected electronically with thissystem.

Congress may wish to consider: 1) whetheran independent review of E-COM security iswarranted to ensure that adequate securitymeasures are in place to protect the privacyof EMS messages, and 2) whether the PostalAct and/or Communications Act should beamended to provide additional statutory pri-vacy protection (including the possible man-dating of data encryption to provide additionaltechnical privacy protection).

Maintain Oversight andInitiate Planning on USPS

Long-Term Viability

Although the immediate focus is on E-COM,and on providing a clear direction for USPSinvolvement in EMS and resolving currentregulatory problems and delays, EMS issuesare likely to be with Congress for many years,driven by the impact of EMS on USPS, therole of USPS in EMS, and the broader impactof EMS on American society and the publicat large. For a discussion of these broader im-pact areas, see the related OTA report on Com-puter-Based National Information Systems(1981). Within this context, Congress will needto maintain oversight and initiate planning onUSPS long-term viability.

The following areas warrant further study:

● USPS initiatives designed to develop im-proved working relationships with privatetelecommunication and computer firms;

●

●

●

●

●

joint technical and market tests with pri-vate firms to evaluate various EMS alter-natives;use of EMS to help USPS maintain ade-quate service levels to rural and less pop-ulated areas;use of EMS to help USPS offset the re-duction (or elimination) of the revenue for-gone subsidy (which is provided to offsetrevenue losses from mail service providedat reduced rates) and permit continuationof a lower rate to nonprofit and educa-tional organizations;use of EMS in the future in combinationwith the USPS infrastructure (perhapsscaled down) and delivery network to pro-vide other Federal Government services(e.g, printing and delivery of forms anddocuments); andUSPS long-range planning on the possi-ble need for labor force reductions, job re-training, adjustments in retirement andnew hire rates, and implications for unioncontract negotiations.

In view of aggressive private sector Genera-tion III EMS activity and the continuing eco-nomic trends that work in favor of electronicmail and against paper-based mail, it seemsclear that Congress and USPS should beginplanning now for the future viability of USPS.Changes are taking place so fast in the so-called “communications revolution” that bythe time USPS might actually experience sig-nificant impacts on mail volume, most oppor-tunities for participation in EMS will havepassed and it will be much more difficult toadjust.

Chapter 2

Background andPurpose of Study

Contents

PageIntroduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

Congressional Interest.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . + . . . . . . . . . . . . 17

Study Purpose and Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

Chapter 2

Background and Purpose of Study

IntroductionBy some measures, the U.S. Postal Service

(USPS) has done remarkably well in the 10years since postal reorganization.* Gross pro-ductivity, as measured by number of pieces ofmail per workyear, has increased by 34 per-cent since fiscal year 1970 when 741,000 postalemployees delivered 85 billion pieces of mail.In fiscal year 1980, 667,000 employees de-livered 106.3 billion pieces of mail. In 1980,mail volume increased by 6.5 percent over1979. First- and third-class mail volumes con-tinue to show strength with increases of ap-proximately 15 and 36 percent, respectively,over the last 5 years. During the same period,pieces of mail per capita have increased byabout 13 percent, from 418 pieces per personin fiscal year 1976 to 480 pieces in 1980.1

On the financial side, USPS more than brokeeven over the combined 1979-80 period. Thesurplus of $470 million in fiscal year 1979 off-set a deficit of $306 million in 1980. The costof postage continues to rise, but since 1975 ata rate that is lower than the consumer priceindex. If fully adjusted for inflation since 1975,a first-class stamp would cost 22¢ rather thanthe current 20¢. However, if fully adjusted forinflation since 1967, a first-class stamp wouldcost only about 14¢. Thus, about 6¢ out of thecurrent 20¢ represents a real increase in thefirst-class letter rate since 1967. The letter ratein the United States continues to be well belowrates in most other countries. For example, in1980 the letter rate was about 29¢ in Britainand Japan, 31¢ in France and Sweden, and 33¢in West Germany.2

*The po9~ Mrgani.zation Act of 1970 enacted by Congressabolished the Post Office Department as a cabinet level agen-cy of the executive branch. Postal functions were transferredto an independent Government agency known as the UnitedStates Postal Service, which commenced operations on July 1,1971.

IAnnual Report of the Postmaster General fiscal 1980.‘Ibid.

Why, then, is there concern about the futureof USPS? While recent years have been rela-tively good, the 1980’s and 1990’s will posea number of new and perhaps more difficultchallenges.

First, the potential for further improve-ments in postal productivity through presortdiscounts and through mechanization andautomation is limited. Even a fully im-plemented expansion of the ZIP code to ninedigits (known as ZIP + 4) would mean acumulative reduction of only about 15,600workyears by 1987 according to USPS esti-mates. Compared to the current USPS annualworkyears of 679,000, the productivity im-provement would amount to about 2.3 per-cents This means that continued automationwill provide only a small part of the total pro-ductivity improvement needed by USPS tooffset inflationary increases in employee com-pensation and transportation, which togetheraccount for over 92 percent of total USPS an-nual expenditures.4

Second, while USPS operating statistics in-dicate that ontime delivery has been main-tained for 95 percent of first-class mail des-tined for local or metropolitan area overnightdelivery, ontime delivery within 600 miles andfor cross-country fist-class mail generally hasdeclined since 1977. Two-day delivery of let-ters within 600 miles was achieved 86 percentof the time in 1980 compared to 90 percent in1977. Three-day delivery of cross-country let-ters was achieved 87 percent of the time in1980 compared to 91 percent in 1977.5 Part of

‘Comptroller General of the United States, Imph”cations ofEkctmnic Mad for the Postal S&vice Work Fome (Washington,D. C.: U.S. General Accounting Office, Feb. 6, 1981), p. 32. Theestimated changes in workyears per year from the nine-digitZIP are as follows: 1981 ( +3), 1982 (+87), 1983( – 1334), 1984(–2382), 1985 (–3301), 1986 (–4378), and 1987 (–4295) fora total reduction of 15,600 workyears. Per Nov. 4, 1981, discus-sion with Douglas Lynn of the USPS Operations Group.

‘Annual l?epor~ op. cit., p. 24.‘Ibid., pp. 8, 11.

15

16 ● Implications of Electronic Mail and Message Systems for the U.S. Postal Service—.

the problem can be attributed to cutbacks inair carrier service, but it is also possible thatthe rising total volume of mail has begun totax the capacity of the overall mail distribu-tion system. For example, postal officials havetestified that the elimination of Saturdaydelivery alone would create a mail backlog suf-ficient to overload the system for Mondaydelivery. On the other hand, postal officialspoint to the system’s ability to handle peakvolumes during holiday periods as evidencethat overall capacity has not been reached.

Third, a variety of technical regulatory, andmarket developments are contributing to arapid increase in commercially offered elec-tronic mail and message systems (EMS) thatincreasingly will compete with USPS As earlyas 1977, studies projected that EMS coulddivert substantial portions of mail from USPS,to the extent that total mail volume might ac-tually start to decline by the early 1980’s.6

None of these projections has yet been real-ized. However, the large number of commer-cial firms now in the electronic messagemarket,* coupled with recent developments inpersonal computers, viewdata/teletext, andother home information systems, suggeststhat this prospect is much more realistic thanit was just a few years ago. Furthermore, regu-latory decisions by the Federal Communica-tions Commission over the last few years,coupled with congressional actions to rewritethe Communications Act of 1934, are clearingthe way for increased competition in the elec-tronic message market. Thus, while EMS willundoubtedly stimulate new message “traffic”between individuals and organizations, theyalso have the potential for diverting existing

%%, for example, F. B. Wood, R. W. Anthony, et al., USPSand the (hnxnunications Revolution: Impacts, Options, andIssues, Final Report to the Commission on Postal Service,prepared by the Program of Policy Studies in Science and Tech-nology, The George Washington University, Washington, D. C.,Mar. 5, 1977. Also see Arthur D. Little, The Impact of Elec-tronic (%mmunication Syst%ms on First Class Mail Volume inJ9801990, Cambridge, Mass., April 1978.

*Ex~ple9 include Quik-COmm (General Electric), Telemail(GTE Telenet), On-Tyme (Tymnet), InfoPlex (Plexus), Faxgram(Graphnet), Mailgram (Western Union), and Datapost (southernPacific). Electronic mail is also one of several services offeredby Satellite Business Systems and other specialized or value--added common carriers.

traffic from other media, including a portionof the message traffic currently handled by theU.S. Mail. This diversion would come at a timewhen mail volume may also be reduced by theconsolidation of many bills and payments, orthe elimination of some billing and paymenttransactions altogether, as a consequence ofthe implementation of electronic funds trans-fer (EFT) systems.

Substantial erosion of U.S. Mail volumes,particularly first-class mail, could tend to raisethe unit cost of carrying the remaining mailvolume and/or force a reduction in the quali-ty and convenience of mail service, thus pro-viding further incentive for mail users toswitch to alternatives such as private deliveryservices, newspaper inserts, and the like. Sucha development could disadvantage users with-out a viable alternative to the U.S. Mail,jeopardize the ability of USPS to provideuniversal service, and adversely affect USPSemployees.

An important issue for USPS is whether andhow it will participate in the provision of elec-tronic mail and message services. USPS cur-rently provides a portion of Western Union’sMailgram service, and in January 1982 in-troduced a domestic service called “electroniccomputer-originated mail” (E-COM). An inter-national EMS service, known as “interna-tional electronic post,” has also been initiated.INTELPOST is outside the scope of thisstudy. USPS has been developing a more ad-vanced “electronic message service system”or EMSS which, for the purposes of this study,is considered to be an extension of E-COM tofull nationwide deployment at all or most serv-ing post offices. There are a variety of waysin which USPS could play one or more rolesin the provision of EMS services, ranging fromthe delivery of hardcopy output to the provi-sion of a complete end-to-end electronic mailservice.

The Annual Report of the Postmaster Gen-eral for fiscal year 1979 states flatly: “In thefuture, the only way the Postal Service will beable to keep its volume rising and financesdependable is through participating in elec-

Ch. 2—Background and Purpose of Study ● 1 7———-——

tronic mail services.”7 In a July 1979 policystatement, the White House agreed, statingthat “the national interest requires a PostalService which can serve all Americans and in-terface with the world’s postal services effi-ciently and economically. The service has pro-gressively achieved productivity improvementby mechanization and automation in process-ing of conventional mail . . . A postal EMS isthe next logical step to achieve further costreduction and mail processing improvement.”8

Over the last 3 years, the role of USPS inEMS has been in dispute before various reg-ulatory agencies, the courts, and Congress.USPS initiated E-COM service in January1982 after the USPS Board of Governors ap-proved the Postal Rate Commission’s (PRC)1980 recommended decision, with the excep-tion of PRC’s “experimental” designation (ofE-COM as an experimental rather than a per-manent service) which was successfully ap-pealed by the Governors to the courts. How-ever, several private firms and the current ad-ministration believe that E-COM as present-

7AnnuaJ Report of the Postmaster Genera4 fiscal 1979, p. 6.8Administration Policy Statement, The White House, July

19, 1979.

ly implemented differs significantly in otherways from the concept originally recom-mended by PRC. A 1981 inquiry opened byPRC to review what form of E-COM USPSshould be offering was suspended after itslegality was challenged by USPS.

Comments filed before PRC jointly by theDepartments of Commerce and Justice and acourt challenge to E-COM filed by Justice in-dicate that the current administration is notsupportive of E-COM as presently operating,or possibly of any USPS role in EMS that in-volves telecommunication, data processing, orprinting. This in part reflects continuing con-cern that E-COM places an independent Gov-ernment agency (USPS) in competition—per-haps unfairly and/or illegally-with privatefirms. Some of these firms believe that the de-mand for EMS can be met by private offer-ings, and that the USPS role should be re-stricted to the delivery of hardcopy outputfrom electronic message systems. Others areconcerned that in the future USPS may ex-pand its EMS role from printing, enveloping,and physical delivery-as in E-COM—to in-clude telecommunication and perhaps elec-tronic delivery as well.

Congressional InterestThe implications of electronic mail and

message systems for USPS, and especially therole of USPS in EMS, have been and continueto be a primary concern of congressional com-mittees with direct jurisdiction over USPSand the Postal Reorganization Act of 1970,i.e., the House Committee on Post Office andCivil Service (and particularly the Subcommit-tee on Postal Operations and Services and theSubcommittee on Postal Personnel and Mod-ernization) and the Senate Committee on Gov-ernmental Affairs (especially the Subcommit-tee on Civil Service, Post Office, and GeneralServices).

The role of USPS is also of interest to thecommittees with jurisdiction over telecom-munications to the extent that USPS becomes

involved with an EMS service that is subjectto the Communications Act of 1934, in wholeor in part. These committees include theHouse Committee on Energy and Commerce(and the Subcommittee on Telecommunica-tions) and the Senate Committee on Com-merce, Science, and Transportation (Subcom-mittee on Communications). In addition, othercommittees, such as the House Committee onGovernment Operations Subcommittee onGovernment Information and IndividualRights, have an interest in the privacy, com-petitive, and related implications of a USPSrole in EMS.

Despite a variety of legislative initiatives inrecent years, Congress has yet to agree on aclearly defined EMS role for USPS. H.R. 2813,introduced in the 97th Congress, would require


USPS to establish a separate organizationalunit to provide EMS service, would prohibitcross-subsidization of EMS from public funds,would prohibit USPS from owning telecommu-nication services (but would permit USPS tocontract for same), and would limit USPS toEMS services where the output is physicallydelivered through the U.S. Mail.9 H.R. 4758,also introduced in the 97th Congress, wouldprohibit all Federal agencies, including USPS,from providing data-processing or telecom-munication services to non-Federal persons orentities unless explicitly authorized bystatute. This bill would appear to prohibitUSPS from offering telecommunication trans-mission and data-processing services withoutspecific congressional approval.10 In the

‘H.R, 2813, 97th Cong., 1st sess., Mar. 25, 1981, to amendtitle 39 of the United States Code, referred to the Committeeon Post Office and Civil Service.

10H.R. 4758, 97th Cong., 1st sess., Oct. 15, 1981, to amendthe Federal Property and Administrative Services Act of 1949.See (2mgressiona] Record-House, Oct. 15, 1981, p. H7425.

Study PurposeThis study addresses three major questions:

1.

2.

3

To what extent are privately offered EMSand EFT systems likely to affect thevolume of mail handled by USPS?Are changes in USPS mail volume likelyto lead to significant adjustments inUSPS rates, service levels, and/or laborforce requirements? andWhat are the implications for the futureof USPS and how it might participate inthe provision of EMS services?

At the heart of the study are two computer-based quantitative models. The first is amarket penetration model used to project thelevel of conventional and electronic mailvolumes under different sets of assumptions,and the second is the USPS revenue and costmodel. There are four basic inputs to themarket penetration model: 1) the baselinedescription of the mail flows derived from asurvey based on 1977 data conducted forUSPS by the Survey Research Center at the

Senate, S. 898, “The TelecommunicationsDeregulation and Competition Act of 1981,”as enacted includes an amendment intendedto clarify provisions of the act relating to elec-tronic mail. The amendment makes clear thatS. 898 does not authorize or prohibit USPSfrom offering telecommunication services orthe electronic delivery of messages, whetherby resale or otherwise. If, at some future time,Congress should authorize USPS to offer suchservice or if current law is interpreted toauthorize it, the amendment stipulates theconditions under which such service would beoffered, including the establishment of aseparate organizational entity, among otherthings. Thus, in effect, S. 898 and the relatedSenate floor debate prior to enactment reaf-firm the absence of congressional consensuson the participation of USPS in EMS.11

11% ~wssjonal &ord—&na@ O c t . 7, 1981 ~ PP.S.11211-11216.

and Approach

University ofMichigan;122) a set of EMS andEFT technology assumptions; 3) a set ofassumptions about the underlying growth rateof the mainstream; and 4) the range of selectedalternatives (e.g., low, medium, high growth)for EMS development. The market penetra-tion model is explained in chapter 3 and ap-pendixes A and B, and the results are outlinedin chapter 4.

The second quantitative model is the USPSrevenue and cost model. It is designed to pro-ject the impacts of the growth or decline ofoverall USPS mail volume (conventional andEMS) on USPS rates, service levels, and laborrequirements. The USPS revenue and costmodel and the results for first-class mail arepresented in chapter 5.

IZM. K~& W. Rodgers, et al., Household MaJ%Jtream Stidy,Find Report prepared for Mail Classification Research Divi-sion, U.S. Postal Service, 1978.

—.

Ch. 2—Background and Purpose of Study ● 1 9

This study gives primary emphasis to im-pacts on the USPS mainstream (based on re-sults of the market penetration model) and onUSPS rates, service, and labor (based on com-bined results of the market penetration modeland the revenue and cost model). Implicationsfor rates, service, and labor are summarizedin chapter 6. Secondary emphasis was placedon the potential implications for the telecom-munication and computer industries, EMS pri-vacy and security, and the long-term viabil-ity of USPS, discussed in chapter 7. Congres-sional policy considerations are discussed inchapter 8.

A note on computer modeling is in order.Prior studies on mail diversion have provento be oversimplified. In order to better under-stand the complexities involved, OTA devel-oped and used computer-based models to per-mit consideration of a larger number of vari-ables and interrelationships than would other-wise be possible. While this approach is moresystematic and complete than those used inprior studies, computer modeling has itslimitations.

First, the precision of the projections can bemisleading. The reader should focus on generaltrends and relationships rather than the spe-cific numbers projected. Second, the models

are highly sensitive to initial assumptions. Forthis reason, sensitivity runs were conductedto see how much projections would changewith different assumptions. Third, the modelshave limited ability to anticipate unexpectedevents. For example, the possibilities of aWorld War III, second Great Depression,mandatory wage/price freeze, nationwidepostal labor strike, or repeal of the Private Ex-press Statutes were not included. In otherwords, the models are based on a relatively“surprise-free” future. Fourth, the models donot fully reflect the possible effect of rates onmail volumes. There is a feedback process, butits exact nature is unknown. That is, changesin rates may have a significant effect on mailvolume, which in turn affects mail rates 1, 2,or 3 years later (in the next ratesetting cycle).Despite these limitations, computer modelingcan be a useful analytical tool.

Again, the study as a whole, and particular-ly the use of computer modeling, is intendedto help Congress better understand the possi-ble implications of EMS for USPS. The studyis not intended to make a prediction of thefuture course of events. Many variations arepossible. Finally, the study specifically avoidsmaking judgments about the impacts identi-fied and makes no recommendations relativeto the role of USPS.

Chapter 3

Market Penetration Modeland Technology Assumptions

Contents

Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Mailstream . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

EFT Diversion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

EMS Diversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Generation II Growth and Timing Estimates . . . . . . . . . .Generation III Growth and Timing Estimates . . . . . . . . .Relationships Between Generation II and Generation IIIAlternative Generation II Growth and Timing Estimates

Mainstream Growth . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .Estimates. . .. . . . . . . . . . . .

. . . . . . . . . . . .

TABLES

Table No.l. Baseline Mainstream . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2. Mail Content . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3. Major Mailstream Segments Vulnerable to Penetration by Electronic

Funds Transfer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4. Major Mainstream Segments Vulnerable to Penetration by Electronic

Mail and Message Systems ....,... . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5. Assumptions About Rate of EFT Penetration . . . . . . . . . . . . . . . . . . . . .6. Assumptions About Rate of EMS Penetration . . . . . . . . . . . . . . . . . . . . .7. Assumptions for Generation II Growth Alternatives . . . . . . . . . . . . . . .8. Alternative Assumptions About Underlying Mainstream

Growth Rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Page23

24

26

2729303132

33

Page2424

25

25262932

33

FIGURES

Figure No. Pagel. Market Penetration Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232. Comparison of Conventional Mail Service With Generations I,II,

and III EMS Service . . . . . . . . . . . . . . . . . . . ........... 28

Chapter 3

Market Penetration Modeland Technology Assumptions

Introduction

The primary purpose of the market penetra- 1995, and 2000. The basic elements of the mod-tion model is to estimate the level of electronic el are shown in figure 1. (See app. A, fig. A-1,and conventional mail volumes in 1985, 1990, for further details.)

Figure 1.— Market Penetration Model

SOURCE: Office of Technology Assessment, see app. A, table Al for further details.

23

24 . Implications of Electronic Mail and Message Systems for the U.S. Postal Service

Baseline MainstreamAs one input to the market penetration mod-

el, the baseline mainstream was divided intoa number of different submarkets (subclassesof mail) in two ways—by mail content and bysender/receiver pairs. The mail content cate-gories included correspondence, merchandise,bills, financial statements, and advertising,among others. Senders and receivers weregrouped into households and nonhouseholds.Thus, the four possible sender/receiver pairsincluded household-to-household, household/nonhousehold, nonhousehold/household, andnonhousehold/nonhousehold.

The baseline volume for every class of con-ventional mail (first, second, third, fourth,other) was estimated for each category of mailcontent and sender/receiver pair. These esti-mates were based on data in two studies con-ducted for the U.S. Postal Service (USPS) bythe University of Michigan Survey ResearchCenter known as the Household MainstreamStudy and Nonhousehold Mailstream Study,which in turn were based respectively on 1977and 1979 mainstream data.1 For consistency,the 1979 nonhousehold data were used to es-tablish ratios among the types of mail andthen applied to 1977 mail volumes so that alldata would be for the 1977 calendar year.

The resulting baseline mainstream is high-lighted in tables 1 and 2 and detailed in ap-pendix A (table A-l). It is possible that someshifting among the mail segments has oc-curred since 1977, although a comparison ofdata for the 1977 and 1980 fiscal years indi-cates no major changes. The total mail volumehas grown from about 92 billion to 106 billionpieces between 1977 and 1980. First-class mailhas decreased from 58 percent of the total to56.6 percent; second-class mail has decreasedfrom 9.4 percent to 7.9 percent; and third-classmail has increased from 26 percent to 28.5 per-cent of the total mail volume.2

‘M. Kallick, W. Rodgers, et al., Household Maiktnwn Study,Final liepor~ prepared for the Mail Classification Division,USPS, 1978. Also, Nonhouwhold MaiMrearn Study, InterimReport for First Postal Quarter PFY 1979, July 1979.

‘Annual Report of the Postmaster General, fiscal 1980, pp.28-29.

Table 1.— Baseline Mainstream

To households To nonhouseholds

From 7.60/o 9.90/0households . . . . . . . . . (7.1 billion pieces) (9.2 billion pieces)

From 5 3 . 3 % 29.20/ononhouseholds. . . . . . (49.7 billion pieces) (27.3 billion pieces)

SOURCE: 1977 data, University of Michigan Millstream Study conducted forUSPS; percentages shown are based on total 1977 mail volume of93.3 billion pieces, See app. A, table A-2, for further details

Table 2.—Mail Content (illustrative)

To households To nonhouseholds

From Correspondence—7.1 Negot iab le Ins t rumentshouseho lds percent of total mail (e .g . , checks)–7.0

v o l u m e percent

C o r r e s p o n d e n c e — 1 . 5percent

From Third-class (mostly bulk Bills and financial state-non- rate) mail— 10.2 ments—9.1househo lds percent percent

Bi l ls—9.5 percent C o r r e s p o n d e n c e – 6 . 1

Adver t is ing—9.1 percent percent

F inanc ia l s ta tements— Adver t is ing—5.7 percent

2.8 percent

SOURCE: University of Michigan Millstream Study conducted for USPS. Per-centages shown are based on total 1977 mail volume. See app. A,table A-2, for further details

Mail originating from nonhouseholds consti-tutes over four-fifths of the total mainstream.Nonhousehold-originated bills and financialstatements alone account for over one-fifth ofthe total, advertising about one-seventh, andbulk rate mail over one-tenth. Of the one-fifthof the mail originating from households, mostis either correspondence (letters and cards) ornegotiable instruments (checks).

All mainstream segments were evaluated todetermine whether they potentially could behandled (in whole or in part) by electronicfunds transfer (EFT) and/or electronic mailand message systems (EMS). Those major seg-ments judged to be vulnerable to penetrationby EFT and/or EMS are listed in tables 3 and4. (See app. A, table A-2, for a complete list.)

As shown in tables 3 and 4, accounting onlyfor major mainstream segments, about two-thirds of the mainstream is vulnerable to pene-tration by EFT and/or EMS. This translatesinto about three-quarters of first-class mail

Ch. 3—Market Penetration Model and Technology Assumptions ● 2 5

Table 3.—Major Mainstream Segments Vulnerable to Penetration byElectronic Funds Transfer

1977 volume PercentageMail (billions of of first- Percentage of

Mainstream segment class pieces) class mail total mail

Nonhousehold to householdbills . . . . . . . . . . . . . . . . . . . . . . 1 8.9 15.4% 9.50/0

Non household to householdfinancial statements. . . . , . . . 1 2.6 4,5 2.8

Non household tononhousehold bills andfinancial statements. . . . . . . . 1 8.4 14.6 9.0

Household tononhousehold negotiableinstruments . . . . . . . . . . . . . 1 6.5 11.3 7.0Totals . . . . . . . . . . . . . . . . . . . . 26.4 45.8% 28.3%

SOURCE: University of Michigan Mainstream Study conducted for USPS. Percentages based on total 1977 mail volume of93.3 billion pieces and 1977 first-class mail volume of 57.7 billion pieces. First-class mail defined to includepenalty and franked mail.

Table 4.—Major Mainstream Segments Vulnerable to Penetration byElectronic Mail and Message Systems

Mail 1977 volume Percentage of mail class

Mainstream segment class (billions of pieces) First Third Total

Household to householdcorrespondence . . . . . . . . . . . . . . . . 1 6.6 11.4% — 7.1 %

Household to nonhouseholdcorrespondence . . . . . . . . . . . . . . . . 1 1.4 2,4 — 1.5

Non household to household third-class (mostly bulk rate) mail . . . . . . 3 9.5 — 38.80/o 10.2

Nonhousehold to household 1 1.1 1.9 — 1.2advertising . . . . . . . . . . . . . . . . . . . . . 3 7.2 — 29.4 7.7

Nonhousehold to nonhousehold 1 2.3 4.0 — 2.5advertising . . . . . . . . . . . . . . . . . . . . . 3 3.0 — 12.3 3.2

Non household to nonhouseholdcorrespondence . . . . . . . . . . . . . . . . 1 5.2 9.0 – 5.6Totals . . . . . . . . . . . . . . . . . . . . . . . . . 36.3 28.70/o 80.50/o 39.00/0

SOURCE: University of Michigan Mainstream Study conducted for USPS. Percentages based on total 1977 mail volume of93,3 billion pieces and 1977 first-class mail volume of 57.7 billion pieces, and 1977 third-class mail volume of24.5 billi~ pieces. First-class mail defined to include penalty and franked mail.

volume and about four-fifths of third-class natures, endorsements, and documentation formail. Actually, only the merchandise and mis- many such instruments. Even here, the possi-cellaneous segments and nonhousehold to non- bilities for electronic certified mail and elec-household legal/financial instruments were tronic contract signing are being researched.3

assumed to have no vulnerability. Legal/finan-cial instruments were assumed to have nopotential for electronic handling due to the fre- “’Cryptographers Gather to Discuss Research,” Science, vol.quent requirement for verified and original sig- 214, NOV. 6, 1981, p. 647.

26 ● Implications of Electronic Mail and Message Systems for the U.S. Postal Service—

EFT DiversionThe next step in the market penetration

model was to subtract from the baseline mail-stream the mail that could be diverted to EFT.While for some purposes EFT might be viewedas a special type of EMS, other EFT applica-tions, such as the point-of-sale use of debitcards, could eliminate certain paymentmessages altogether. Accordingly, in thisstudy EFT was considered to be separate fromEMS. Mail diverted to EFT was consideredunavailable for EMS. For mainstream seg-ments such as bills and statements where bothEFT and EMS could produce diversion, EFTdiversion was assumed to occur first. Theresidual mail volume in each mail segmentafter EFT diversion was then considered thepotential market for EMS diversion. Thediversion to EFT was modeled using the logis-tic substitution process described in appen-dix B.

Based on the results of a separate OTAstudy,4 current trends suggest that a signifi-cant consolidation of bills and financial state-ments is likely to take place via EFT, but thatit will take many years. OTA has assumedthat the use of EFT for bills and financialstatements in the long run would result in a90-percent reduction in total bills and state-ments received via conventional mail by theaverage household or nonhousehold. Thus, themaximum potential fraction (or penetrationpotential) of bills and statements that couldbe diverted to EFT is 0.9, as shown in table5. OTA assigned an initial growth rate of 20percent, as indicated in table 5. Given thenature of the logistic substitution process, a20-percent initial growth rate would decline toa 5-percent growth rate for the 20th year out.It would take 20 years to progress from 5 to75 percent of the maximum potential diver-sion. The year of 5-percent diversion (timewhen 5-percent diversion occurs) was esti-mated to be 1985. The year of 75-percent diver-

4See OTA report Sleeted Electronic finds Transfer Issues:Privacy, Security, and Equity, OTA-BP-CIT-12, March 1982.See also El?/l The Next Fifteen Y- Electronic Banking, Inc.,June 1980, a working paper prepared for the above report.

Table 5.—Assumptions About Rate of EFTPenetration

Year of 5 percent penetration—1985Year of 75 percent penetration—2005Initial exponential growth rate (1985)—20 percentGrowth rate at 50 percent penetration (year 2000)–5 percentPenetration potential —0.9 for bills and financial state-

ments1.0 for negotiable instruments

Key technologiesAutomated teller machines (ATMs)

25,000 in operation (1981)ATMs estimated by industry to at least double by 1990and could increase to 120,000 (an annual growth rate ofroughly 10 to 20 percent):

. deposit● cash withdrawal● bill or loan payment● cash advance

Point-of-sale terminals87,500 in service (1981):

• check validation● credit card authorization● debit of transaction balance

Telephone bill payment (TBP)302 financial institutions offer (1980)TBP transactions estimated by industry to be growing by27 percent a year:

● bill or loan payment● account status inquiry● interaccount transfer

SOURCE Off Ice of Technology Assessment, see app A, table A-3, for furtherdeta{ls.

sion was estimated to be 2005. As shown intable 5, this growth rate is generally consist-ent with rates of growth projected by industryfor key EFT technologies.

Likewise, the results of the OTA study sug-gest that EFT is likely to displace checks andother paper-based negotiable instruments, butthat this displacement will take many years.OTA has assumed that all such instrumentseventually could be displaced by EFT. Thus,the EFT penetration potential is 1.0 for nego-tiable instruments sent to households or non-households. As with bills and financial state-ments, OTA has assigned an initial growthrate of 20 percent and estimated the year of5-percent diversion to be 1985. (See app. A,table A-3, for details.)

The OTA assumptions for bills and finan-cial statements and for negotiable instrumentswere optimistic in the sense that the actual

—


penetration potential might be lower than 0.9 the mainstream. The assumptions about EFTand 1.0, respectively. Therefore, the actual may be affected by various intangible consid-EFT penetration is more likely to be lower erations important to EFT and EMS users,than assumed. EFT was defined, in effect, as especially those relating to consumer prefer-an all-electronic service completely outside of ences and institutional marketing strategies.

EMS DiversionAs noted earlier, the residual mail volume

in each mail segment after EFT diversion isthe potential market for EMS diversion. EMSdiversion is divided between Generation II andGeneration III and was calculated through useof the same logistic substitution process usedfor estimating EFT diversion (see app. B). Theterms Generation II and Generation III areexplained and compared in figure 2.

The EMS diversion model was based on aset of EMS technology assumptions discussedbelow, highlighted in table 6, and detailed inappendix A (table A-4). The assumptions re-late to the following six categories of technol-ogy as applied to the various combinations ofmail content and sender/receiver pairs:

1.

2.

3.

4.

5.

Generation II EMS systems with earlyelectronic printers (no color). This cate-gory includes such industry offerings asMailgram, Datapost and Tyme-Gram,and USPS offerings such as E-COM.Generation II EMS systems with ad-vanced electronic printers (including a col-or capability).Electronic data processing and officeautomation. This category includes Gen-eration III technologies such as computernetworks, communicating word process-ors, public and private message and pack-et-switching networks, and facsimile sys-tems oriented toward nonhousehold use.Home computer terminals. Included arehome computers and associated commu-nications concepts/services such as PCNet (Personal Computer Network).Viewdata/teletext. This category includesservices, primarily to the home, based on

6.

the use of the television set and the tele-phone.Inexpensive hardcopy receiver. Facsimilereceivers or character printers at a pricewhich could find acceptance in a major-ity of homes are included in this category.

The maximurn market penetration potentialwas estimated for each mainstream segment.As with EFT, the assumptions about EMSpenetration potential were optimistic in thesense that the actual penetration potentialmight be lower due to restrained consumer ac-ceptance and other intangible factors. In mostinstances, the entire segment was judged 100-percent susceptible to Generation II and Gen-eration III EMS. The exceptions are asfollows.

About 30 percent of the “other nonadvertis-ing” segments (nonhousehold to nonhouseholdand nonhousehold to household) is made up ofpamphlets, newsletters, official documents,coupons, and stockholder communications.Items of this type were judged not likely tobe susceptible to EMS technologies that areexpected to achieve widespread use over thenext 20 years. Hence a maximum potentialpenetration of 70 percent (P = 0.7) wasestimated.

The displacement of direct mail “advertis-ing” and greeting “cards” segments to thehome (nonhousehold to household and house-hold to household) by TV-based GenerationIII home terminals was judged to be limitedby the constraints of the video medium. Thus,a maximum Generation III penetration poten-tial of 30 percent (0.3) was estimated for thesesegments.


Figure 2.—Comparison of Conventional Mail Service With Generations I, II, and Ill EMS Service

SOURCE: Office of Technology Assessment; and National Academy of Sciences, .Hectromc ~essa9e Systems for the U.S. Postal Service, 1976.

.—— .


Table 6.—Assumptions About Rate of EMS Penetration (illustrative)

Early Generation II EMS (using current technology blackand white printers) for correspondence, bills, third-classbulk mail:

Year of 5 percent penetration—1983Year of 75 percent penetration—1996Initial exponential growth rate (1983)—30 percent

Advanced Generation III EMS (using inexpensive homehardcopy receiver) for nonhousehold to household bills andstatements:

Year of 5 percent penetration—1990Year of 75 percent penetration—2010Initial growth rate—20 percent

Advanced Generation II EMS (using high resolution colorprinters) for advertising, greeting cards:

Year of 5 percent penetration—1995Year of 75 percent penetration—2015Initial growth rate (1995)—20 percent

Generation Ill EMS (using public and private message andpacket-switching networks, communicating word proces-sors, computer networks) for intraoffice correspondence:


for interoffice correspondence:Year of 5 percent penetration—1984Year of 75 percent penetration—2004Initial growth rate (1984)—20 percent

Generation Ill EMS (using viewdata/teletext) for householdto household cards:

Year of 5 percent penetration–1985Year of 75 percent penetration—2005Initial growth rate—20 percent

Advanced Generation Ill EMS (using home computer ter-minals) for household to household correspondence:


for nonhousehold to household correspondence and bulkmail:


Key technologies:Home computers:

● 500,000 installed (1980)● Estimated by industry to grow to 4.5 million installed

by 1985 and 33 million by 1990 (roughly a 50 percent an-nual growth rate).

Video computer games:● Revenues increased from $308 million in 1978, to $968

million in 1979, to $2.8 billion in 1980 (roughly a 300 per-cent annual growth rate).

Mini and small business computers:● Revenues of about $9.4 billion worldwide (1980)• Estimated by industry to continue to grow at 25 to 35

percent a year.Computer software products:

● Revenues of about $1.5 billion (1980). Estimated by industry to grow at 30 percent annually

over the next 5 years.Data communications:

● Estimated revenues of about $4 billion (1979) and grow-ing at 30 to 35 percent a year.

SOURCE: Office of Technology Assessment, see app. A, table A-4, for further details,

Generation II Growth andTiming Estimates

The attractiveness of Generation II servicesis determined primarily by the capabilities andcost effectiveness of the devices for convertingthe electronic signals back to hardcopy. De-vices that have black and white capability onlyand limitations in page size and print style arefrequently not as attractive as conventional-ly printed material. Also, printing systemsmust be very cost effective or EMS prices willbe too high to compete successfully with con-ventional mail.

For the purposes of this study, OTA as-sumed that the electronic printers available inthe 1980’s will be limited in resolution andflexibility and will lack color capabilities. Ad-vanced electronic printers, which are expectedto become available at cost-effective prices inthe 1990’s, will add greater resolution, grey

scale, and color capabilities and probably willinclude greater flexibility in materials han-dling. Recent technology and product an-nouncements suggest that advanced printersmay be available earlier than assumed for thisstudy.

Generation II EMS services using early elec-tronic printing capabilities, if priced competi-tively with mail service, could begin to findsubstantial use by nonhousehold senders inthe next few years. For correspondence, bulkstatements, and other nonadvertising content,OTA estimated a 5-percent diversion of exist-ing mail to Generation II EMS by 1983, witha high initial rate of growth (30 percent) whichcould lead to a 75-percent market share about13 years thereafter, as summarized in table 6.The use of Generation II for advertising pur-poses, however, is expected to be largely de-layed until color capabilities become available,and even then growth will be slower to the ex-


tent that cost and relative inflexibility contin-ue to limit the advantages of electronic colorprinting over conventional printing. Thus, forthe advertising segment (nonhousehold to non-household and nonhousehold to household) a5-percent market share was forecast for 1995,with an initial growth rate of 20 percent.

Households will not be able to initiate a sig-nificant volume of Generation II EMS untilhome terminals capable of originating textcome into widespread use. About 500,000home computers had been sold by 1980,5

though many of these were not equipped forcommunications. OTA assumed that it willtake several more years before 5 percent ofhouseholds, or roughly 4 million homes, areequipped with communications-capable homecomputers, and that there will be additionaldelays before many of these home computersare used routinely for correspondence. Thus,OTA assumed that the EMS market share forcorrespondence originating in the home willnot reach 5 percent before 1987. A high initialgrowth rate (30 percent) was projected, whichis consistent with growth rates projected byindustry for home computers, as indicated intable 6.

The requirement for a color capability is ex-pected to put greeting cards in the same posi-tion as advertising, thus delaying a 5-percentmarket share for Generation II EMS until1995.

Generation III Growth andTiming Estimates

Generation III EMS services between non-household senders and receivers are expectedto be based largely on electronic data-process-ing and office automation technologies. Thereare strong incentives within this sector forkeeping information in electronic form and formachine processing by the receiver. As a resultof these incentives, there is a healthy competi-tion among several technologies for this mar-ket, including word processors, computer-

6U.S. Congress, Office of Technology Assessment, (%mputenBased National Information [email protected]: Technolo~ and PubficPolicy Issues OTA-CIT-146, September 1981.

based message systems, intelligent communi-cation networks, and store and forward mes-sage systems. Until recently, this competitionhas tended to impede the development ofstandards among different vendors supportingeach technology, and for information exchangebetween systems based on the different tech-nologies. The International Standard Organi-zation, the Consultative Committee for Inter-national Telephone and Telegraph, and theU.S. standards groups continue to work ondeveloping standards. The process of agree-ing on and then implementing standards hasbeen slow but appears to be accelerating. Thetime required to achieve and implement stand-ards at a variety of system levels will be a prin-cipal determinant of the rate of growth of Gen-eration III message systems within the non-household sector.

The problem will be easiest to resolve withinindividual companies. OTA estimated a 5-per-cent market share for Generation II EMS in1983, and initial growth at a fast rate (30 per-cent). As shown in table 6, OTA estimated a75-percent market share for Generation IIIEMS intra-office correspondence in 1996. Aslower initial growth rate (20 percent) was pro-jected for interoffice correspondence due to in-compatibility and the number of differentstandards issues involved. These rates ofgrowth are generally consistent with industryprojections (listed in table 6) for small businesscomputers, computer software, and data com-munications.

The standards problem will begin to be re-solved first for correspondence, which requiresa minimum of content standardization. OTAestimated a 5-percent market share in this seg-ment in 1984. Generation III will become at-tractive for bills and statements when the re-cipients can automatically process the infor-mation received. This requires considerablestandardization of data elements and formats.OTA anticipated a slow penetration of thesecomplex standards to other sectors, in partdue to the software development required toemploy them. Initially, exchanges frequentlyare likely to be via hand-carried or mailed com-puter tapes substituting for numerous paper


documents. Five-percent penetration in thismarket for Generation III is not expected tooccur until 1985.

Generation III systems will begin to replaceadvertising within the nonhousehold sectorwhen “online” catalogs and order entry sys-tems are implemented between corporate buy-ers and their suppliers. Such systems will re-quire extensive software development andstandardization. Thus, OTA projected an addi-tional 2 years (compared to bills and state-ments) to reach a 5-percent penetration ofadvertising (1987).

Generation III systems involving the house-hold as either sender or receiver will be pacedby the rate of acceptance of one of three prin-cipal home terminal technologies-the homecomputer terminal, the video-based viewdata/teletext terminal, and the inexpensive homehardcopy receiver.

A home computer terminal or its equivalentwill be required for households to originate cor-respondence. As discussed above, OTA pro-jected that a 5-percent market share for mes-sage services using home terminals would oc-cur about 1987. By that time, however, manystandards issues relating to home computerservices are likely to have been resolved.Hence, OTA assumed very rapid initialgrowth (40 percent) for Generation III corre-spondence and other nonadvertising messagesbetween households and nonhouseholds.

For correspondence between households,Generation III EMS growth is expected to beslower, since both sender and receiver mustbe equipped with a terminal device. For exam-ple, with 50 percent of households equipped,only 25 percent of household pairs on the aver-age would have a terminal available at bothends. For this reason, the projected initialgrowth rate for household-household Genera-tion III EMS correspondence is slower (20 per-cent).

Viewdata/teletext systems are most likelyto penetrate the advertising and greeting cardsegments involving household receivers,though the maximum penetration potential is

limited. These systems are projected toachieve a 5-percent market share somewhatahead of home computers–OTA estimated1985. Advertising by viewdata/teletext is ex-pected initially to grow very rapidly (40 per-cent), since it is paced only by availability ofthe home receiver. The greeting card segmentis likely to grow slowly, again because of therequirement that both sender and receiver beequipped.

The use of Generation 111 EMS services totransmit bills and financial statements tohouseholds requires home terminals as a pre-requisite. In addition, it seems likely thatmany consumers will desire a hardcopy oftheir statements or bills for tax records andother purposes, thus making consumers morereluctant to accept bills and statements overa viewdata-like terminal. OTA assumed thathome hardcopy equipment sales will bedelayed 3 years behind video terminal salesand will grow at a slower rate, thus affectingthe use of Generation III for bills and state-ments. OTA estimated that inexpensive hard-copy printers capable of reproducing the con-tents of a TV display will be produced in vol-ume quantities for under $200 (1980 dollars)per unit when market penetration reaches 5percent (1990).

Relationships BetweenGeneration II and

Generation III Estimates

Generation II growth rate and timing esti-mates are not assumed to have any significanteffect on the rate or timing of mail diversionto EFT systems or on the rate and timing ofGeneration III growth. The latter assumptionmay seem surprising at first since GenerationII and III are in some sense competing. How-ever, while the decision to send messages byGeneration II as opposed to conventional mailis almost entirely at the discretion of thesender, the decision to receive mail electronic-ally-and hence via Generation III–is large-ly at the discretion of the receiver. Thus, ifGeneration III is available, the recipients of


messages will elect Generation III instead ofhardcopy delivery when it suits their purposesor is convenient to do so, regardless of whetherthe hardcopy comes from conventional mailor is the output of a Generation II system.Thus, the rate of penetration of Generation IIIis not likely to be significantly affected by thestate of penetration of Generation II.

It may also be thought that Generation IIIis but a simple extension of Generation II—perhaps just running a telecommunication linefrom a Generation II terminus to the ultimateuser. In this view, earlier introduction ofGeneration II would speed the introduction ofGeneration III. However, many GenerationIII EMS systems are quite different in char-acter from Generation II EMS systems.Therefore, sending greeting cards by view-data, transmitting bills or statements betweencomputers, reading messages from a comput-er-based message system, or placing ordersagainst a supplier using an online catalog andorder entry system are not functional exten-sions of a store and forward message systemlike that which forms the basis for GenerationII. Both developer and user decisions for Gen-eration III message systems are expected tobe quite independent of the status of Genera-tion II.

Alternative Generation IIGrowth and Timing Estimates

The assumptions outlined above for thegrowth and timing of Generation II EMS areintentionally on the high side (i.e., optimisticin terms of rate and extent of development),but still are plausible in terms of technical,economic, and market realities. Henceforth,this set of assumptions will be referred to asthe baseline alternative for Generation IIEMS development.

In order to test the sensitivity of the marketpenetration model and the projected mail vol-umes to changes in the baseline Generation IIEMS assumptions, OTA has defined threeother alternatives, as presented in table 7: 1)very high Generation II EMS growth, 2) mod-erate Generation II EMS growth, and 3) slow

Table 7.—Assumptions for Generation II GrowthAlternatives

High but plausible Generation II EMS growth (baselinealternative):

● Peak volume (year 1995) about 60 to 75 percent of RCAprojected year 2000 volume

● Early Generation II—5 percent penetration in 1983initial growth rate 30 percent

● Advanced Generation II—5 percent penetration in 1995initial growth rate 20 percent

Very high Generation II EMS growth:● Peak volume (year 1995) about 110 to 130 percent of

RCA projected year 2000 volume● Early Generation II—5 percent penetration in 1983

initial growth rate 40 percent● Advanced Generation I I—5 percent penetration in 1992

(accelerated by 3 years)initial growth rate 30 percent

Moderate Generation II EMS growth:● Early Generation II—5 percent penetration in 1987

(delayed by 4 years)● Peak volume (year 2000) same as very high alternative

peak in 1995● Advanced Generation II—5 percent penetration in 1993● Initial growth rates same as very high growth alterna-

tiveSlow Generation II EMS growth

● Peak volume (year 2000) about 25 percent of RCA pro-jected year 2000 volume

● Generation II growth rates cut in half compared tomoderate growth alternative—20 percent for Early

Generation II—15 percent for Ad-

vanced Generation II

SOURCE: Office of Technology Assessment (see flg 8, ch. 4, for graphic com-parison of alternatives).

Generation II EMS growth. When comparedto the 1977 estimates made by RCA for theUSPS with respect to EMSS,6 the baselinealternative (high but plausible Generation IIgrowth) would project a peak volume of about60 to 75 percent of the RCA peak of 25 billionmessages. The very high growth alternativewould project a peak of 110 to 130 percent ofthe RCA peak, and the slow growth alterna-tive about 25 percent of the RCA peak. Themoderate growth alternative would project apeak of 55 to 65 percent of the RCA peak, butwould show a growth track substantially slow-er than the baseline alternative but faster thanthe slow growth alternative. (See ch. 4, fig. 8,for a graphical comparison.)

‘RCA Government Communications Systems Division, Elmtronic Message Service System: Growth and Economic Anal-yses, Camden, N. J., 1977, p. 6-13.

Ch. 3—Market Penetration Mode/ and Technology Assumptions ● 3 3

These alternatives are used in later chapters force requirements to assumptions about theto test the sensitivity of the projected mail vol- growth and timing of Generation II EMS.umes, first-class revenues and costs, and labor

Underlying Mainstream GrowthBetween 1970 and 1980, the overall volume

of mail handled by USPS grew at an annualcompounded rate of about 2.2 percent. Be-tween 1900 and 1977, the annual compoundedrate of growth was about 2.0 percent. There-fore, OTA used an estimate of 2 percent as thebaseline underlying rate of growth in the mail-stream. In other words, absent any diversion,it was assumed that the mainstream wouldgrow by about 2 percent per year over the next20 years (the timeframe of this study).

However, there are several indications thatthis assumption may be on the low side. Forexample, the University of Michigan study es-timated an annual mainstream growth rate of2.9 percent prior to technological diversion.7

Also, in 3 of the last 4 years the actual rateof growth in the mainstream exceeded eventhis 2.9 percent level.* The USPS estimatedrate of growth for fiscal year 1981 is 3.5 per-cent. 8 Finally, between 1947 and 1977, theannual compounded mainstream growth wasabout 3.2 percent.

Table 8 summarizes the various justifica-tions for assuming a 2- v. a 3-percent underly-ing mainstream growth rate. The balance ofevidence appears to suggest that a 2- or evenl-percent rate is typical of economically de-pressed periods, while the rate has been 3 per-cent or more during better economic times.

These possibilities were accommodated inthe market penetration model by making sen-sitivity runs of the model under different setsof assumptions, and by adjusting all resultsupward by 10 percent to be consistent with

7Kallick, Rodgers, et. al., op. cit.*5.1 ~rcent for figc~ year 1978; 2.9 percent for fiscal year

1979; 6.5 percent for fiscal year 1980; and 3.5 percent (estimated)for fiscal year 1981.

‘Per Nov. 22, 1981, telephone conversation with LouEberhardt, USPS Public Information Office.

Table 8.—Alternative Assumptions AboutUnderlying Mainstream Growth Rate

Justifications for 3-percent underlying growth rate1947-77 average mainstream growth = 3.2 percent per

year1977-81 average mainstream growth = 4.2 percent per

year1951-77 average first-class mail growth = 2.8 percent

per yearUniversity of Michigan mainstream study estimatea = 2.9percent per year

Justifications for 2-percent underlying growth rate1970-80 average mainstream growth = 2.2 percent per

year1900-77 average mainstream growth = 2.0 percent per

year (includes 1930’s depression and both WorldWars)

1951-77 ratio of first-class mail to disposable personalincome declined from 76 million letters per $1billion to 57 million letters per $1 billionb

Justifications for l-percent underlying growth rate1971-76 average mainstream growth = 1.1 percent per

year1971-76 average first-class mail growth = 0.7 percent

per year

afvI. Kall ick, W, Rodgers, et al., Househo/d Malktream Study, F ina l ~ePorf,prepared for USPS Mail Classlflcatlon Division, 1978.

bJohn F, M~L~ughlin, et al,, Te/ep/rone.Letfer Conrpetltlon A ~irsf ~oo~, Har”

vard Unwersity, 1979.

SOURCE: Off Ice of Technology Assessment

the estimated fiscal year 1981 USPS overallmail volume. Sensitivity runs were carried outfor both a higher (3 percent) and a lower (1 per-cent) underlying mainstream growth rate. Theresults of selected sensitivity runs are pre-sented in chapter 4.

The one-time 10-percent adjustment in re-sults for consistency with 1981 data wasneeded because the actual mainstream growthduring the 1977-81 period far exceeded the2-percent annual rate that was assumed initial-ly. As noted in chapter 4, the market diver-sion model was based on 1977 mainstreamdata, which was the only detailed data avail-able at the time of the study. The model as-sumed that, under the baseline growth of 2

—


percent per year, the total mail volume wouldgrow from about 93 billion pieces in 1977 toabout 100 billion pieces in 1981. The actualUSPS mail volume grew from about 92 billionpieces in 1977 to an estimated 110 billionpieces for fiscal year 1981.9 In order to com-pensate for this higher-than-expected growth

‘Ibid.

rate, all results of the computer modeling wereincreased by 10 percent to make the projectedand actual figures for 1981 consistent and toremove the effects of the 1977-81 growth dis-crepancy from future year projections. With-out the 10-percent adjustment, future yearprojections would have been penalized for thisdiscrepancy.

—

Chapter 4

Market Penetration Results

—. —

Contents


Base Case Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

Sensitivity Analyses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40Three-Percent Underlying Growth Rate . . . . . . . . . . . . . . . . . . . . . . . . . . . 40Other Sensitivity Runs. . . . . . . . . . . . . . . ...... . . . . . . . . . . . . . . . . . 42

Comparison of Alternative Generation II EMS Growth andTiming Estimates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

TABLE

Table No. Page9. Sensitivity Analyses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

FIGURES

Figure No. Page3. Market Penetration for High But Plausible Generation II EMS

Growth Alternative . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384. Breakdown of Mail by Class for High But Plausible Generation II

EMS Growth Alternative . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 395. Market Penetration for High But Plausible Generation HEMS

Growth Alternative . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 406. Breakdown of Mail by Class for High But Plausible Generation II

EMS Growth Alternative . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 417. Sensitivity Analyses of Market Penetration Projections Assuming

High But Plausible Generation II EMS Growth . . . . . . . . . . . . . . . . . . . 438. Generation II EMS Market Projections . . . . . . . . . . . . . . . . . . . . . . . . . . 47

Chapter 4

Market Penetration Results

IntroductionThis chapter presents the results of the mar-

ket penetration analysis using the model andtechnology assumptions described in chapter3. The primary use of the model here is toestimate future levels of conventional and elec-tronic mail volumes under a variety of condi-tions.

To recap, the starting point for the modelis the baseline mainstream, which is thenbroken down into a number of different sub-markets (classes or subclasses of mail). Themodel estimates the maximum potential frac-tion of each submarket which appears to besuitable for handling by (i.e., diverted to) elec-tronic funds transfer (EFT) or electronic mailand message systems (EMS). Then, based onspecific assumptions about the relevant tech-nology, the model estimates the rate and tim-ing of penetration of EFT and/or EMS intoeach submarket.

For each submarket, the model is thus ableto estimate the portion of the mainstream that

would be diverted to EFT and EMS and theportion that would remain as conventionalmail. OTA has assumed that, unless otherwiseindicated, the U.S. Postal Service (USPS)would deliver the hardcopy output of EMSservices but not the electronic output. The por-tion of the mainstream diverted to EMS is fur-ther divided into Generation II (defined asEMS with hardcopy output and delivery) andGeneration III (defined as EMS with elec-tronic delivery). The total remaining USPSmainstream for any given set of assumptionswill then be the sum of all submarkets of un-divertable conventional mail (mail not suitablefor electronic handling), plus residual conven-tional mail (mail that is suitable for electronichandling but has not yet been diverted), plusGeneration II EMS volume. As defined in thisstudy, Generation III EMS and EFT both in-volve electronic delivery and therefore areassumed to be completely outside of the USPSmainstream.

Base Case Results (High But PlausibleGeneration II EMS Growth, 2-Percent

Underlying Mainstream Growth)The results of the market penetration anal-

ysis for the high but plausible Generation IIEMS growth alternative under the baseline as-sumptions are shown in figure 3.

Under the base case, assuming a 2-percentgrowth rate in the underlying mainstream,USPS-delivered mail (conventional plus Gen-eration II EMS) would rise to about 118 billionpieces by 1990 and then decline to about 89billion pieces in 2000. By 1995, USPS-de-livered mail would be about equal to the 1980USPS mail volume of 106 billion pieces. Con-

ventional mail would have declined signifi-cantly as a percentage of total mail, fromabout 94 percent in 1985 to only 47 percentin 2000. Thus, conventional mail would con-stitute less than one-half of the total mail-stream, although still representing a substan-tial absolute volume of about 75 billion pieces.In contrast, the combined total of EFT andEMS would have risen from about 5.6 percentof the total mainstream in 1985 to about 53percent in 2000. Of the roughly 85 billionpieces of “electronic” mail in 2000, GenerationIII EMS would account for about 56 percent,

37

97-918 0 - 82 - 4

Figure 3.—Market Penetration for High But Plausible Generation II EMS Growth Alternative(assuming 2°/0 growth in underlying mainstream)

.;“1 mail (conventional -

and Generation II EMS)

.,1 . USPS - delivered., ‘ .; ~“,.

~, , (conventional and.i.’; .,!* - IT ‘*. f

Generation II EMS) . . . ...115.77 118.21 106.20 88.52 ‘: ““%,:1

SOURCE: Office of Technology Assessment.

EFT about 28 percent, and Generation IIEMS about 17 percent. By 2000, GenerationIII EMS and EFT would still be increasingat a fairly rapid rate, while Generation II EMSwould have peaked and started to decline.

Overall, the picture that emerges is onewhere conventional mail volume would declineby about 32 percent by 2000 compared to the1981 volume of 110 billion pieces. USPS-

delivered mail would decline somewhat less–by about 20 percent–due to the offsetting ef-fect of increases in Generation II EMSvolume. However, this Generation II EMS“cushion” would peak about 1995 and be onthe decline by 2000 in the face of competitionfrom Generation III EMS.

The breakdown of USPS-delivered mailamong the various classes of mail would also

Ch. 4—Market Penetration Results ● 39

change significantly. As shown in figure 4, thesplit in conventional mail between first classand all other classes would essentially reverse.In 1985, first-class conventional mail would ac-count for about 61 percent of total conven-tional mail, and all other classes would accountfor about 39 percent. This is essentially thesame as the split indicated by actual fiscal

Figure 4.— Breakdown of Mail by Class for High(assuming 2°/0 growth

70

60

50

40

30

20

10

0

year 1980 mail volume statistics. By 2000,however, first-class conventional mail wouldaccount for only about 43 percent of total con-ventional mail, while the share for all otherclasses of mail would increase to 57 percent.When Generation II EMS first-class mail istaken into account, all first-class mail (conven-tional plus Generation H EMS) declines some-

But Plausible Generation II EMS Growth Alternativein underlying mainstream)

Volume of mail(in billions of pieces)

Type of mail 1985 1990 1995 2000

Conventional first-class . . . . . . . . . . . . . 68.84 63.56 47.15 32.37. Convent ional o ther a . . . . . . . . . . . . ....44.32 45.60 43.57 42.72

Generation II EMS first-class . . . . . . . . . 1.90 7.26 13.64 12.51

Generation II EMS otherb . . . . . . . . . . . . 0.61 1.79 1.84 0.92

USPS - delivered totalsFirst-class. . . . . . . . . . . . . . . . . . . . . . . 70.74 70.82 60.79 4 4 . 8 8 . .

Other classes . . . . . . . . . . . . . . . . . . . . 44.93 47.39 45.41 4 3 . 6 4 .

asecond, third, fourth classes, and miscellaneous.w,,

.I t

1985 1990 1995 2000

Year

SOURCE Office of Technology Assessment.

40 Implications of Electronic Mail and Message Systems for the U.S. Postal Service

what less as a percentage of total USPS-delivered mail. However, even including Gen-eration II EMS, the split between first classand all other classes of mail would change from61/39 in 1985 to essentially an even split

Several computer runs

Sensitivitywere performed to

determine the- sensitivity of the base caseresults to changes in key variables and/orassumptions.

Three-Percent UnderlyingGrowth Rate

As discussed in chapter 3, the growth inUSPS mail volume since World War II hasaveraged over 3 percent compounded annual-

(51/49) in 2000. Given the revenue/cost struc-ture of USPS, this change could have signifi-cant implications for USPS revenues, rates,and competitive posture, as will be discussedlater.

Analysesly. A 3-percent growth rate can be justified onthe several grounds summarized earlier intable 10. Some researchers believe that a3-percent rate should be defined as the baselinerather than 2 percent. Figure 5 shows theresults of the market penetration analysis forhigh but plausible Generation II EMS growth,assuming a 3- rather than a 2-percent underly-ing growth rate in the mainstream. The dif-ferences from the base case are significant.USPS-delivered mail would peak at about 134

Figure 5.—Market Penetration for High But Plausible Generation II EMS Growth Alternative(assuming 30/’ growth in underlying millstream)


Ch. 4—Market Penetration Results 4 1

billion pieces in 1990 and decline to slightlymore than 110 billion pieces in 2000. Thus, ifthe underlying growth rate in the mainstreamequals or exceeds 3 percent annually, USPS-delivered mail volume would exceed 1981levels at least through 2000. In other words,with a 3-percent underlying growth rate, theUSPS-delivered mail volume (conventionalplus Generation II EMS) would not drop

below the estimated 1981 volume of 110 billionpieces until 2000. Even the conventional mailvolume would not drop below 110 billion piecesuntil roughly 1994. The relative breakdown ofmail by class would not be significantly dif-ferent for an underlying growth of 3 percentas compared with 2 percent, but the volumesfor all types of mail would be significantlyhigher, as indicated in figure 6.

Figure 6.—Breakdown of Mail by Class for High But Plausible Generation II EMS Growth Alternative(assuming 3°/0 growth in underlying mainstream)

80

70

60

u)co.-—--nc 40.-

20

10

0

. Volume of mail(in billions of pieces)

Type of mail 1985 1990 1995 2000

Conventional first-class . . . . . . . . . . . 74.43 72.16 56.20 40.51Conventional othera . . . . . . . . . . . . . . . 48.03 51.78 51.94 53.47Generation II EMS first-class . . . . . . . 2.06 8.25 16.26 15.65

. Generation II EMS otherb . . . . . . . . 0.66 2.02 2.19 1.17

USPS - delivered totalsFirst-class. ... . . . . . . . . . . . . . . . . 76.49 80.41 72.46 56.16Other classes ., . . . . . . . . . . . . . . ....48.69 53.80 54.13 54.64

aSecond, third, fourth classes, and mmcellaneousbTh lrd class

.

1985 1990 1995 2000

Year


42 ● Implications of Electronic Mail and Message Systems for the U.S. Postal Service——.. ———.-——

Other Sensitivity Runs

In order to fully test the sensitivity of thebase case results to changes in key assump-tions, several other sensitivity runs were con-ducted. The results are summarized in figure7. The projected USPS-delivered mail volumeswould be higher than the base case if: 1) EFTgrowth rates were cut in half (half of what wereassumed for the base case); 2) the underlyinggrowth in the mainstream was 3 percent ratherthan 2 percent (discussed earlier); 3) 5-percentpenetration of Generation III EMS was de-layed 3 years; or 4) Generation II EMS stimu-lated 100 percent additional traffic. If two ormore of these changes from the base case oc-curred simultaneously, the projected USPS-delivered mail volume would be even higherthan shown in figure 6.

On the other hand, the projected USPS mailvolumes would be lower than the base case if:1) Generation III penetration was acceleratedby 3 years; 2) USPS did not deliver 100 per-cent of industry Generation II EMS hardcopyoutput; 3) the underlying growth in the mail-stream was 1 percent rather than 2 percent;or 4) a large percentage of second- and third-class mail was lost to alternative delivery serv-ices. Again, if two or more of these changesfrom the base case occurred at the same time,the projected USPS-delivered mail volumeswould be even lower than indicated in figure 7.

OTA’s qualitative evaluation of the likeli-hood of various changes is summarized intable 9. With respect to changes that would

reduce mail volume compared to the base case,OTA concluded that a l-percent underlyingmainstream growth rate, a doubling of the in-itial EFT growth rate (from 20 to 40 percent),and an acceleration of the year of 5-percentGeneration III penetration (from 1987 to 1984)were all unlikely, as was a significant reduc-tion in USPS delivery of industry GenerationII hardcopy output (short of a major revisionin the Private Express Statutes). OTA did con-clude that significant diversion of second- andthird-class mail to alternative delivery serviceswas possible, although probably not at the

rate assumed in the sensitivity run shown infigure 7.

With respect to changes that would increaseprojected mail volume compared to the basecase, OTA concluded that while a 4-percentunderlying mainstream growth rate was un-likely, a 3-percent rate was quite possible,given growth trends during periods of eco-nomic prosperity. However, the current uncer-tainty in the short- and long-term economicoutlook suggested to OTA that a 3-percent un-derlying growth rate assumption would haveto be considered somewhat optimistic. OTAalso concluded that reductions in the base caserates of development for EFT and GenerationIII EMS were possible, due to marketing andcompetitive (and, in the case of EFT, regula-tory) uncertainties. On the other hand, tech-nology per se does not appear to be a limitingfactor, and the sales of home computers, com-puter games, and small business computersare indicative of rapid development. As for thestimulation of additional Generation II EMSvolume, OTA could not determine whether theexperience with all-electronic technologies(e.g., telephone) is applicable. Some stimula-tion of additional messages, although prob-ably considerably less than the 100-percentstimulation assumed in the sensitivity runshown in figure 7, seems possible. This andtwo other sensitivity runs are discussed inmore detail below.

One-Hundred-Percent Stimulation of Gen-eration II- EMS Traffic.--The base case as-sumed that Generation II EMS traffic wouldbe diverted on a one-for-one basis from the con-ventional mainstream; that is, Generation IIEMS volume is subtracted from the conven-tional mail volume. In actual practice, Genera-tion II EMS systems might stimulate addi-tional traffic, rather than just diverting con-ventional mail traffic.

Experience with other electronic communi-cation services suggests that the availabilityof Generation II EMS may indeed stimulatedemand for new messages not presently sentthrough the mail at all. Figure 7 shows the

Ch. 4—Market Penetration Results ● 4 3—

Figure 7.—Sensitivity Analyses of Market Penetration Projections Assuming High But PlausibleGeneration II EMS Growth

130 —

\

30/. growth in underlying mainstream . . . . . . . . . . . . . . . . . . . 125.17 134.20 126.59 110.80

1000/. stimulation of additional Generation II EMS . . . . . . . . 118.28 127.26 121.68 101.97

EFT growth rates cut in half . . . . . . . . . . . . . . . . . . . . . . . . . . . 115.76 119.90 112.10 100.22

Generation Ill EMS delayed in 3 years . . . . . . . . . . . . . . . . . . 117.03 123.19 118.17 98.93

20/. growth in underlying mainstream . . . . . . . . . . . . . . . . . . . 115.67 118.21 106.20 88.52

Generation Ill EMS accelerated by 3 years . . . . . . . . . . . . . . . 112.99 109.22 94.37 80.62

USPS does not deliver industry Generation II . . . . 113.26 109.16 90.74 75.10

10/0 growth in underlying mainstream . . . . . . . . . . . . . . . . . . . 107.47 104.01 88.95 70.58

Second-/third-class mail diversion to alternative delivery. 103.31 98.95 82.98 62.59

SOURCE Office of Technology Assessment


Table 9.—Sensitivity Analyses

USPS-delivered mail volume would be reduced compared tothe baseline projections if:

●

●

●

●

●

Growth in underlying mainstream were 1 percent—Unlikely, except in event of economic depression.EFT growth rate doubled from 20 percent (in 1985) to 40percent—Unlikely.Generation Ill EMS were accelerated by 3 years from1987 (for 5 percent penetration) to 1984—Unlikely.Significant reduction in USPS delivery of private sectorGeneration II hardcopy output occurred—Unlike/yunder current USPS interpretation of Private ExpressStatutes (PES); however, FCC and some private firmsbelieve that hardcopy output falls within the jurisdic-tion of the Communications Act, not the PES.Significant diversion of second-/third-class mail toalternative delivery occurred—Possible; some diver-sion known to be occurring, but second-class has re-mained essentially constant over last 10 years andthird-class has increased by 52 percent. USPS rate in-creases may accelerate use of alternative delivery.

USPS-delivered mail volume would be increased comparedto the baseline if:

●

●

b

●

●

Growth in underlying mainstream were 4 percent—Unlikely over the long-term, although short-termgrowth spurts of 4 percent are possible.EFT growth rate were halved from 20 percent (in 1985)to 10 percent— Possible due to marketing, competitive,and regulatory uncertainties.Generation Ill EMS were delayed by 3 years from 1987(for 5 percent penetration) to 1990–Possible due tomarketing and competitive uncertainties; however,sales of home computers, computer games, and smallbusiness computers look very strong.Stimulation of additional Generation II EMS volumeoccurred—Possible given that other electronic tech-nologies (e.g., telephone, computer conferencing) havegenerated additional message volume; however,whether experience with all-electronic technologiesapplies to hybrid forms (such as Generation II EMS) isunknown.Growth in underlying mainstream were 3 percent—Quite possible given the historical growth trends dur-ing periods of relative economic prosperity.


results of the market penetration analysis forhigh but plausible Generation II EMS growth,assuming 100-percent stimulation of EMStraffic. This means that for each messagediverted from conventional mail to GenerationII EMS, a new Generation II EMS messageis generated.

Under this assumption, USPS-deliveredmail would peak at about 127 billion pieces in1990 and decline to a little over 100 billionpieces in 2000. USPS-delivered mail volumewould exceed present levels through about1998, although the conventional mail volume

would drop below the present level by 1990.Generation II EMS volume would grow muchfaster and sooner, and would outpace Genera-tion III EMS at least through about 1995. Bycomparison, in both the base case and the3-percent underlying growth case, GenerationIII EMS would overtake Generation II EMSas early as 1990. Overall, a 100-percent stim-ulation of Generation II EMS traffic wouldresult in a higher projected USPS-deliveredmail volume than the base case, but not ashigh as the 3-percent underlying growth case.There is, however, a question as to whetherthe 100 percent EMS stimulation assumptionis realistic.

Generation III EMS Three Years Sooner.—Generation III EMS involves end-to-end elec-tronic service; that is, electronic delivery ofmail as well as electronic sending and trans-mission. Electronic delivery requires that bothsenders and receivers of mail have the neces-sary terminal equipment. In developing themarket penetration model, OTA made a num-ber of assumptions about the growth of Gen-eration III. For example, OTA projected thatin 1987 home computer terminals (or theirequivalent) would achieve a 5-percent share ofmail segments involving the household aseither sender or receiver. While this wasOTA’s best estimate based on economic, mar-ket, and technological conditions at the timeof the study, the timing and rate of home com-puter development is a subject of considerabledebate.

In order to test the sensitivity of the base-line market penetration results to GenerationIII, the model was run with all Generation IIItiming estimates advanced by 3 years. Thatis, 3 years were subtracted from all estimatesof the year of 5-percent penetration for a par-ticular Generation III technology and marketsegment. For example, the time to 5-percentpenetration for home computer terminal pen-etration of household-to-household corre-spondence was changed from 1987 to 1984.

Under the base case, Generation II volumewould be greater than Generation III volumethrough about 1990. With 100-percent EMS

Ch. 4—Market Penetration Results ● 4 5—— — —.

stimulation, Generation II volume would begreater through about 1996. But if GenerationIII came 3 years sooner than assumed in thebase case, Generation II would never exceedGeneration III. Even in the peak year (1995)for Generation II, Generation III EMS vol-ume would be more than four times larger. By2000, Generation 111 would be about 56 billionpieces (or messages) and rising rapidly, whileGeneration II would be about 9 billion piecesand declining.

As a result, the USPS-delivered mail volumewould be less compared to the base case, sincethere would be less Generation II EMS hard-copy delivery to offset reductions in conven-tional mail delivery. As a consequence, by1990 total USPS-delivered mail volume wouldfall below the current 1981 level of 110 billionpieces. By 2000, USPS-delivered mail volumewould be down to about 81 billion, a reductionof about 27 percent from 1981. Thus, accelerat-ing Generation III creates a worse case (interms of USPS mail volume) than the basecase.

Second- and Third-Class Mail Losses toAlternative Delivery.–As discussed earlierand presented in figure 4, first-class volumefor the base case declines significantly as apercentage of total USPS-delivered mail. Thisis because first-class mail is most susceptibleto diversion to EFT or Generation III EMS.Other classes of mail, primarily second andthird classes, show very little decline over thenext 20 years. The reduction in first-class mailmight lead to a substantial increase in costs(and rates) for other classes of mail, since theseother classes would have to cover a larger per-centage of USPS fixed institutional costs.Rate increases could in turn lead to additionallosses of second- and third-class mail.

In order to test the sensitivity to suchlosses, OTA conducted a run of the marketpenetration model assuming a 3-percent an-nual reduction in second-class mail and a2-percent annual reduction in third-class mail.While these assumptions are fairly extreme,fourth-class mail has been declining annuallyby an average of 3 to 4 percent over the lastfew years. In contrast, third-class mail has in-creased significantly in recent years. However,both second- and third-class mailers are in-creasing their use of alternative means ofdistribution. For example, some third-classmailers are shifting to newspaper inserts.These are identical in purpose, content, andappearance to items commonly carried as bulkthird-class mail and are much cheaper on a perpiece basis than bulk third class. In the 1980rate case filings before the Postal Rate Com-mission, many mailers indicated that they areclose to the limit in terms of absorbing highermail rates. The rates for second- and third-class mail have already risen by about 400 per-cent since 1970 as a result of steps requiredby the Postal Reorganization Act of 1970 tobring rates for all classes of mail in line withcosts.

The impact of these assumptions is dra-matic. USPS-delivered mail volume (conven-tional plus Generation II EMS) would startdeclining right away, and by 2000 would fallto about 63 billion pieces, about 43 percentbelow the 1981 mail volume. Conventionalmail would decline to about 50 billion pieces,more than 50 percent below the current 110billion pieces. Thus, this level of second- andthird-class diversion clearly leads to the worst-case scenario with respect to USPS mail vol-ume.

Comparison of Alternative Generation IIEMS Growth and Timing Estimates

Up to this point, all market penetration rate of Generation II EMS development. Inresults have been for the baseline EMS alter- other words, for the base case as well as thenative, which assumes a high but plausible various sensitivity runs, the Generation II

.

46 ● Implications of Electronic Mail and Message Systems for the U.S. Postal Service——

EMS growth and timing parameters havebeen held constant while other variables (e.g.,EFT growth rate, Generation III 5-percentpenetration date, underlying mainstreamgrowth rate) have been changed. In order totest the sensitivity of the projected mailvolumes to changes in the baseline GenerationII EMS assumptions, computer runs were con-ducted for each of the four Generation II EMSalternatives defined in chapter 3 (table 7):

1. baseline alternative-high but plausibleGeneration II EMS growth;

2. very high Generation II EMS growth;3. moderate growth; and4. slow growth.

The Generation II EMS market projectionsfor these four alternatives are presented infigure 8. Results are shown for both 2- and3-percent underlying mainstream growth rates.The market projections developed by RCA(under contract to USPS for the electronicmessage service system concept, known asEMSS) are also included in figure 8 for com-parison purposes.

As expected, projected volumes for the highbut plausible Generation II growth alternativefall in the middle when compared to the veryhigh, moderate, and slow alternatives. Some-what surprisingly, however, volumes for thehigh but plausible alternative are considerablybelow the RCA projections. If this alternativeis indeed a high but plausible market develop-ment scenario, as assumed by OTA, then itwould appear that the RCA projections repre-

sent a very high (i.e., optimistic) marketdevelopment scenario, that they have ignoredcompetition with Generation III EMS serv-ices, or both. The size of the Generation IIEMS market takes on considerable impor-tance with respect to the actual deploymentand pricing of any USPS EMS offering.

Based on the figure 8 results, it would ap-pear that a conservative estimate (assumingslow to moderate Generation II growth anda 2-percent underlying mainstream growth)would place the year 2000 Generation II EMSvolume in the range of 7 billion to 14 billionpieces, rather than the RCA estimate of 25billion. Likewise, a conservative estimatewould place the 1995 volume in the range of3 billion to 10 billion pieces, rather than the23 billion RCA estimate. If Generation IIEMS growth actually followed the slowgrowth path, volume is projected to reach only40 million pieces in 1985, equivalent to thevolume of Mailgrams for fiscal year 1980. Onthe other hand, if Generation II EMS growsvery rapidly, the projected volume would ex-ceed the RCA estimates until the late 1990’s.If a 3-percent underlying mainstream growthrate is assumed, the projected year 2000 vol-ume of about 23 billion pieces for the high butplausible Generation II EMS alternative is es-sentially the same as the RCA estimate.

In sum, the projected mail volumes are verysensitive to the assumptions implicit in thefour alternatives considered for Generation IIEMS development.

Ch. 4—Market Penetration Results ● 4 7

Figure 8.—Generation II EMS Market Projections

/

/

\

Underlying Generation II EMS volumemainstreamgrowth (billions of pieces)

Chapter 5

Revenue/Cost Modeland Results

Contents


USPS Revenue/Cost Model for First-Class Conventional Mail . . . . . . . . . . 51

USPS Revenue/Cost Model for First-Class Electronic Mail . . . . . . . . . . . . 52

Results of the Revenue/Cost Analysis for First-Class Mail. . . . . . . . . . . . . 53

FIGURES

Figure No. Page9. Contribution of USPS-Delivered First-Class Mail to USPS

Fixed Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5410. Contribution of USPS-Delivered First-Class Mail to USPS Fixed Costs

for High But Plausible Generation II Growth Alternative . . . . . . . . . . 55

Revenue/Cost Model and Results

Introduction

This chapter describes the revenue/costmodel used by OTA to project the impact ofU.S. Postal Service (USPS) involvement inelectronic mail and message systems (EMS)on its finances and outlines the results of theanalysis. OTA originally intended to prepareprojections of USPS revenues and costs for allclasses of mail under the four alternatives con-sidered for the years 1985, 1990, 1995, and2000. However, the USPS revenue and cost

structure proved to be too complex, and theavailable baseline data too ambiguous, tomake complete projections. Instead, OTA hasfocused on the financial impacts for first-classmail which, according to the results of themarket penetration analysis discussed inchapter 4, will be affected most by EMS. Therevenue/cost model for conventional mail isdescribed first, followed by the model for elec-tronic mail.

USPS Revenue/Cost Model forFirst-Class Conventional Mail

USPS revenue and cost relationships are un-usually complex for several reasons. First,postal costs vary not only with the volume andweight of mail delivered, but also accordingto class of mail, number of route stops, speedof service standards, size and shape of mail,and whether (and how) the mail is ZIP-codedand presorted. Costs also may vary betweenurban, suburban, and rural routes, althoughUSPS does not collect cost data based on maildestination.

Second, cost analysis is complicated by theproblem of how to assign joint and commoncosts properly to the different services theysupport; for example, allocation of deliveryroute costs by class of mail.

Third, there is a problem in determiningwhat costs are variable with volume changesover various time periods. Thus, some coststhat appear fixed in the short run (e.g., 1 to3 years) may be variable (i.e. adjustable tomeet volume requirements) over a 10- or 20-year period.

On the revenue side, postal revenues dependon the volume of mail sent by customers ineach of the many service categories estab-

lished by USPS, and on the rates in each cate-gory. Revenue projections are further compli-cated by the need to consider the impact ofinflation and public subsidies on rates, and inturn the impact of rates on mail volumes ineach service category. There is a feedbackprocess, but its exact nature is unknown. Thatis, changes in rates may have a significant ef-fect on mail volume, which in turn affects mailrates 1, 2, or 3 years later (in the next rate-setting cycle).

In order to simplify the revenue/cost analy-sis for the purposes of this study, OTA devel-oped a USPS revenue and cost model basedon the following assumptions:

● Percentage Fixed v. Variable Costs. Tothe extent that a significant fraction ofUSPS costs are fixed, declining volumeswould cause an increase in the averagecost per piece of mail. This higher costwould have to be recovered by increasesin rates or postal subsidies or offset bycost reductions through service cutbacks.OTA assumed a USPS fixed cost of about36 percent based on the revenue and costanalysis used in the 1980 rate case before

51

.

52 ● Implications of Electronic Mail and Message Systems for the U.S. Postal Service—. —

the Postal Rate Commission (PRC). Theprimary fixed costs were estimated byPRC to be $5.8 billion for USPS institu-tional costs (e.g., headquarters, postmas-ters, inspection service) and $1.8 billionfor service-related fixed costs that couldbe assigned to various mail classes.’

● Revenue and Cost Per Piece. The 1980PRC rate case was also used as the basisfor determining revenue and variable costper piece. For first-class mail, the perpiece revenue and variable costs were 20¢and 13¢, respectively.2 The 20¢/piece rev-enue estimate assumed an 18¢ first-classstamp.

● Economies of Scale. OTA assumed thatUSPS is still operating with economies ofscale, so that mail volume reductionswould tend to increase the per piece costof the remaining mail. If mail volumesreached or exceeded the optimal capacityof the system, then volume reductionsmight actually reduce rather than in-crease the per piece cost.

● Inflation. Clearly, inflation will causepostal costs to rise, and presumably rateincreases will be necessary to keep upwith inflation (to the extent that increasedcosts are not offset by improved produc-tivity). For the purposes of this analysis,future revenues and costs are expressed

‘Postal Rate Commission, Opinion and Recommended Deci-sion, docket No. R-80-1, p. 222.

‘Ibid., app. G, schedule 1, p. 1.

●

●

●

in “constant dollars. ” Changes, too, areexpressed in so-called “real” revenues or“real” costs—net of changes due to infla-tion.Public Service Subsidy. For the purposesof this analysis, the postal public servicesubsidy level was held constant at the$692 million level assumed by PRC in the1980 rate case.3 At the present time, thereare no proposals to increase the subsidy;in fact, the Omnibus Budget and Recon-ciliation Act of 1981 has reduced theauthority for such appropriations to zeroby fiscal year 1984.Productivity. In terms of costs, any over-all productivity improvements with re-spect to conventional mail were assumedto be offset by increases in the cost of cap-ital and increases in real wages. Produc-tivity gains due to the introduction ofEMS were considered as part of the rev-enue/cost model for electronic mail.Use of the Model By using the 1980 esti-mates of per piece first-class mail cost andapplying this to future projections ofUSPS volumes for conventional first-classmail, future costs were calculated in 1980dollars. Likewise, by using the 1980 esti-mate of per piece first-class revenue andapplying this to projected mail volumes,future revenues were calculated in 1980dollars.

‘Ibid.

USPS Revenue/Cost Model forFirst-Class Electronic Mail

In addition to a projected volume of conven-tional mail, USPS will deliver some volume ofelectronic mail (defined as Generation II EMShardcopy). Thus, it was also necessary to de-velop revenue and cost assumptions for USPSelectronic mail services. The cost consists oftwo parts: the cost for the USPS electronicportion of the system (including printing andenveloping), and the USPS mainstream cost ofdelivering the hardcopy.

For the mainstream portion, OTA assumeda cost displacement of 5¢/first-class piece,based on 1980 PRC estimates of the costdisplacement for Mailgram.4 That is, themainstream cost of Generation II would be 8¢,

41bid; according to Frank Heselton, USPS Manager of Rev-enue and Cost Analysis, the 8@/piece Mailgram cost includesonly delivery and administrative costs. When the cost of op-erating Mailgram teleprinters is included, the per piece cost in-creases to about 24$.

Ch. 5—Revenue/Cost Model and Results ● 53

5¢/piece less than the mainstream cost of con-ventional mail. The 5¢/piece cost displacementis also reasonably consistent with estimatesmade by RCA for a USPS electronic system(specifically the electronic message service sys-tem (EMSS) concept).’

For the electronic portion, OTA did not in-dependently verify either the RCA estimatesfor EMSS or the USPS estimates for electron-ic computer-originated mail (E-COM). There-fore, cost estimates were developed only for

‘RCA Government Communications Systems Division,EMSS System Analysis Task AB, VOL II, (%st and ServiceImpact of System Decentralization, October 1977.

USPS delivery of Generation II first-classhardcopy output.

For average revenue per piece of GenerationII EMS hardcopy output delivered, OTA as-sumed that the “markup” of per piece revenueover the per piece cost for EMS must be thesame as the markup for the correspondingclasses of conventional mail. Analysis of the1980 PRC rate case indicated that the averageper piece revenue level for first-class mail wasroughly 50 percent higher than the per piecevariable cost. This 1.50 factor was used to esti-mate a 12¢/piece revenue for USPS deliveryof Generation II EMS hardcopy output.

Results of the Revenue/CostAnalysis for First-Class Mail

Given the first-class mail volume projectionsfrom chapter 4 and using the revenue and costmodels (for both conventional and USPS deliv-ery of Generation II developed above), thefinancial impacts on USPS for first-class mailcan be projected.

Figure 9 summarizes the results for theyears 1995 and 2000 for each of the four Gen-eration II EMS alternatives under the baselineassumptions (2-percent underlying mainstreamgrowth). The results are also shown for eachalternative under the alternative revenue/costassumption alone and in combination with the100-percent EMS stimulation assumption.

The tabular data in figure 9 gives the mailvolumes for conventional first-class and Gen-eration II EMS first-class with USPS deliveryof industry hardcopy output. The revenuesand costs for these volumes are indicatedalong with the contribution of each to USPSfixed costs. USPS is not allowed to make aprofit overall, but individual classes and sub-classes of mail do make varying contributionsto fixed costs. First-class mail historically hasmade the largest contribution of any class ofmail, and thus the continuing ability of first-class mail volumes to generate a substantialcontribution appears to be very important to

overall USPS financial stability. In fiscal year1980, the first-class mail contribution to USPSfixed costs was about $4.2 billion, based on anactual volume of 60 billion pieces and assum-ing 20¢/piece revenue and 13¢/piece variablecost.

Basically, the results indicate that by 2000,for the baseline assumptions, USPS-deliveredfirst-class mail is projected to contribute be-tween $1.25 billion (for the slow GenerationII EMS growth alternative) to about $1.5 bil-lion (for the very high, high, and moderategrowth alternatives) less to USPS fixed coststhan in fiscal year 1980. Thus, in 2000, for thehigh but plausible Generation II EMS growthalternative, the first-class mail contribution isprojected to be about $2.76 billion, which is$1.44 billion less than the contribution in 1980.If a 3-percent underlying mainstream rate isassumed, in 2000 the net reduction in first-class mail contribution for the high but plausi-ble growth alternative would be less but stillsignificant. As shown in figure 10, under the3-percent growth assumption, the first-classmail contribution is projected at $3.46 bil-lion, which is about $750 million below the1980 contribution. With a l-percent underly-ing mainstream growth, the first class mail

97-918 f) - 82 - 5


Figure 9.–Contribution of USPS-Delivered First-Class Mail (conventional plus Generation II EMS) to USPSFixed Costs (assuming 2°/0 underlying mainstream growth)

9

8

7

6

5

4

3

2

1<

0

-';.1.

1 I/)

j ~ '0 ..(~ "0 .~~ '0 -j I/)

c:

~ ~ :~

:0 .!:

~.-:: I/)

iii :~

0 0

"0 CD

~ CJ) a..

;<:"J CJ) ::::>

.9 c:

~~, 02 -;

~ .Q

-~~ 0;:

C 7~~ 0

1 0

0(ij ,) E w

I/) .;:tt!'-

.~ I/) ca

<;> ~: '§

u::

1995 2000



Figure 10.—Contribution of USPS-Delivered First. Class Mail to USPS Fixed Costs for High But PlausibleGeneration II Growth Alternative (assuming 1%, 2%, and 3% underlying mainstream growth)

s1980 contributionof first-class mail

s

II

to USPS fixed costs

—— ——_L —

,> j 1% 2% 3% 1% 2% 3%Conventional first-class with 20¢/piece revenue, 13¢/piece cost. . ’ ”

Volume (billions of pieces) . . . . . . . . . . . . ....39.49 47.15 56.20 25.81 32.37 40.51Revenue ($ billions) . . . . . . . . . . . . . . . . . . . . . . 7.90 9.43 11.24 5.16 6.47 8.10Variable cost ($ billions) . . . . . . . . . . . . . . . . . . 5.13 6.13 7.31 3.36 4.21 5.27Contribution to fixed costs ($ billions) . . . . . . . 2.77 3.30 3.93 1.80 2.26 2.83

>* “;Generation II EMS first-class with 12¢/piece revenue, 8¢/piece cost

Volume . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.42 13.64 14.78 9.98 12.51 15.65 “ ‘ ‘Revenue. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.37 1.64 1.77 1.20 1.50 1 . 8 8

!. Variable cost. . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.91 1.09 1.18 0.80 1.00 1.25‘ . . . . Contribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.46 0.55 0.59 0.40 0.50 0.63

. .- Total first-class contribution to fixed costs. . . . . 3.23 3.85. . . .’ 4.52 2.20 2.76 3.46F. ,. k:” :“ : ‘ . .’ W’ ‘ ‘‘. ,.

SOURCE. Off Ice of Technology Assessment,

contribution would be $2.2 billion, a full $2 would appear to be quite significant consider-billion below the 1980 level. ing that the total public appropriation (public

service subsidy plus revenue forgone) to USPSHow significant is a $0.75 billion, $1.25 bil- for fiscal year 1980 was about $1.6 billion. The

lion, or $1.50 billion reduction in first-class fiscal year 1980 revenue forgone appropriationmail contribution to USPS fixed costs? It (to offset revenue losses from mail service pro-

56 ● lmplications of Electronic Mail and Message Systems for the U.S. Postal Service

tided at reduced rates, primarily nonprofit sec-ond-class mail, nonprofit bulk rate third-classmail, library materials, and free mail for theblind and handicapped) alone was $0.782 bil-lion. And the incremental cost of delivery 6days per week compared to 5 has been esti-mated by USPS to be about $0.65 billion.

In other words, the projected reduction infirst-class mail contribution in 2000 is roughlyequivalent, under the baseline assumptions,to the combined 1980 public service and reve-nue forgone appropriation, or to the revenueforgone appropriation plus the cost of Satur-day delivery, or the equivalent. Under the3-percent mainstream growth assumption, thereduction in first-class mail contribution isroughly equivalent to the 1980 public servicesubsidy, or the revenue forgone appropriation,or the cost of Saturday delivery.

In general, as long as Generation II firstclass costs less per piece than conventionalfirst class but has the same markup of revenueover cost per piece (50 percent), the first-classmail contribution to USPS fixed costs will de-cline as Generation II volume increases (as-suming Generation II EMS costs USPS lessthan conventional.) Thus, as indicated in figure9, the first-class contribution is greater for theslow Generation II EMS growth alternativethan for the high or very high growth alterna-tives.

The financial contribution of Generation IIEMS could be increased if the rate for USPSdelivery of EMS hardcopy output were thesame as the rate for conventional mail deliv-ery. Up to this point in the analysis, OTA as-sumed that all cost savings from EMS wouldbe passed on directly to the EMS user (i.e.,whoever is paying the postage). Thus, the rela-tionship between EMS first-class mail perpiece revenue and cost was assumed to be thesame as for conventional first-class mail. Inother words, cost savings from EMS were re-flected in lower EMS rates rather than in lowerrates for other classes of mail or the mail-stream as a whole. As a result, USPS couldnot obtain any greater return (“markup” or

contribution to fixed costs) from EMS thanfrom conventional mail.

Under the alternative revenue/cost assump-tion tested by OTA, EMS first-class rateswere assumed to be the same as conventionalfirst-class mail rates. In this scenario, USPSwould, in effect, be pricing Generation II EMSfirst-class mail to contribute a higher percent-age (or markup) to fixed costs than would con-ventional mail. Thus, a revenue per piece of20@ was used for EMS instead of 12¢. Asshown in figure 9, when using the alternativerevenue/cost assumption, the high, very high,or moderate growth alternatives would resultin an additional first-class mail contributionof roughly $1 billion in 2000 (compared to theprojected contribution under the baseline as-sumptions). This level would still be a few hun-dred million dollars below the current first-class mail contribution to fixed costs. Notethat the analysis also assumed no loss in EMSvolume due to the higher USPS rates. If EMSusers were sensitive to the increase from 12to 20¢/piece (for USPS delivery of GenerationII hardcopy output), then the increased reve-nue from higher rates might be offset by re-duced revenue from lower volume.

If 100-percent EMS stimulation is also as-sumed (each Generation II message stimulatesan additional new message), the first-class mailcontribution to fixed costs in 2000 would besignificantly higher than the 1980 contribu-tion, by as much as $1 billion for the high andmoderate growth alternatives and $1.5 billionfor the very high growth alternative. Theseprojections are included in figure 9. Thisscenario would enable USPS to use net rev-enues from Generation II EMS to help keepdown overall rates for conventional mail, evenin the face of the declining conventional mailvolumes projected for 2000. However, as notedin chapter 5, the 100-percent EMS stimulationassumption is considered speculative at thistime.

Under current USPS pricing policies, to theextent that private firms transmit GenerationII messages and present them to USPS for


local delivery, these messages will be delivered veloping, and/or transmission) as well asas first-class mail at the same rates as for con- physical processing and delivery, USPS wouldventional mail. Thus, the alternative revenue/ establish rates to cover the costs of electroniccost assumption would apply, and the contri- services plus some markup over costs. Thisbution to USPS fixed costs would be 12¢/piece would provide an additional contribution torather than 4¢. To the extent that USPS pro- fixed costs. OTA has not estimated or ana-vides electronic services (e.g., printing, en- lyzed this contribution.

.

Chapter 6

Implications for Postal Rates,Service Levels, and

Labor Requirements

.—— .

Contents


Postal Rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61

Service Levels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63

Labor Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

TABLES

Table No. Page10. Projected First-Class Mail Rate Increases or Decreases,

11.12.13.

14.

15.

16.

Years 1995 and 2000 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62USPS Service Levels, Fiscal Year 1980 . . . . . . . . . . . . . . . . . . . . . . . . . . 63Structure of the USPS Labor Force, Fiscal Year 1980 . . . . . . . . . . . . . 65Changes in USPS Labor Force Requirements for High But PlausibleGeneration II EMS Growth Alternative, Year 2000, As a Function ofUnderlying Mainstream Growth Rate and Labor productivityImprovement per Year. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68Projected USPS Year 2000 Labor Force Requirements by EmployeeG r o u p , Assuming High But Plausible Generation II EMS Growth,Z-Percent Underlying Mainstream Growth, and 1.5-Percent LaborProductivity Improvement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68Projected USPS Year 2000 Labor Force Requirements by EmployeeGroup, Assuming High But Plausible Generation II EMS Growth,Z-Percent Underlying Mainstream Growth, and 3-Percent LaborProductivity Improvement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69Projected Year 2000 USPS Labor Force Reductions Assuming2-Percent Underlying Mainstream Growth and 1.5-Percent ProductivityImprovement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

FIGURE

Figure No. Page11. Sensitivity Analyses of Projected Changes in USPS Labor Force

Requirements for 1995 and 2000 Assuming High But PlausibleGeneration II EMS Growth and 1.5-Percent Annual LaborProductivity Improvement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

Chapter 6

Implications for Postal Rates,Service Levels, and

Labor Requirements

Introduction

Postal rates, service levels, and labor re-quirements are integrally related. By law, theU.S. Postal Service (USPS) is required toachieve break-even operations; that is, reve-nues should equal costs as nearly as possible.Thus, postal rates are established at the levelsneeded to generate revenues sufficient to coverprojected costs over a given period of time (2to 3 years in a typical ratesetting cycle). Theprojected costs are based on assumptionsabout USPS mail volumes and service levels.The major cost component is labor, which in

PostalWhile OTA was not able to do a revenue/

cost analysis for all classes of mail, the resultsof the analysis for first-class mail suggest thatunder the baseline high but plausible Genera-tion II electronic mail and message system(EMS) growth alternative, postal rates wouldhave to increase in constant dollars (net of in-flation) after 1995, assuming service levelswere not reduced. As shown earlier in figure4 (based on the market penetration analysis),the breakdown of mail by class for the highbut plausible Generation II EMS growth alter-native indicates that through the early 1990’sthe split between first-class mail and otherclasses would be roughly the same as in 1980.Reductions in conventional first-class mailwould be largely offset by increases in Genera-tion II EMS hardcopy output delivered byUSPS.

However, by 1995 the net reduction in totalUSPS-delivered first-class mail would, for the

fiscal year 1980 accounted for about 85 per-cent of total USPS costs.1 For the current levelof service, about 60 percent of labor costs arefixed; that is, are required to maintain theservice level regardless of variations in mailvolume.2

IAnnual Report of the Postmaster Genera~ fiscal 1980, p. 24.‘USPS fiscal year 1980 Revenue and Cost Analysis; also see

Robert W. Anthony, et al., Strategy for Decisions: AmericanPostal Workers Union and the Electronic Information R.evolu-tioq The George Washington University Program of Policy Sci-ence and Technology, Washington, D. C., Mar. 1, 1980, pp. 55-56.

Ratesbaseline high growth alternative, require asmall rate increase to maintain the first-classmail contribution to USPS fixed costs, basedon the chapter 5 revenue/cost analysis (seefigs. 9 and 10). By 2000, the required rate in-crease could be more substantial, on the orderof 18 percent. If the underlying mainstreamgrowth turned out to be 1 percent rather than2 percent, a rate increase of more than 30 per-cent might become necessary. On the otherhand, for a 3-percent underlying mainstreamgrowth, the required rate increase in 2000would be about 7 percent. For illustrative pur-poses, an 18-percent increase in first-class mailrates would translate roughly into an increaseof 3.5¢/piece in the conventional first-classmail rate and a 1.5¢/piece increase in the ratefor delivery of Generation II EMS first class.These rate increases would recover the pro-jected $1.5 billion shortfall in first-class mailcontribution to fixed costs in 2000 for thebaseline high growth alternative.

61

—.————


Projected rate increases (or decreases) undervarious assumptions are summarized in table10. As shown, using the alternative/revenuecost assumption (where revenue/piece is thesame for conventional and Generation II hard-copy fist-class delivery, as would be the caseunder current USPS pricing policies), a smallrate decrease is projected for 1995. If the100-percent EMS stimulation assumption alsoapplied, for the high Generation II growthalternative, rate decreases of about 16 and 9percent are projected for 1995 and 2000, re-spectively. Under these assumptions, the pro-

Table 10.—Projected First-Class Mail RateIncreases or Decreases, Years 1995 and 2000

1995 2000

High but plausible Generation II EMS growth alternativeUnderlying mainstream growth rate

1 percent . . . . . . . . . . . . . . . . . . . . . . . + 10.50/0 + 31.40/0

Underlying mainstream growth rate2 percent , . . . . . . . . . . . . . . . . . . . . . . + 3.2

Underlying mainstream growth rate3 percent ... , . . . . . . . . . . . . . . . . . . . – 2.5

Assume alternative revenue/costfigures. . . . . . . . . . . . . . . . . . . . . . . . . – 6.1

Assume alternative revenue/costfigures and 100 percent EMSstimulation . . . . . . . . . . . . . . . . . . . . . – 15.9

Very high Generation ii EMS growth alternativeUnderlying mainstream growth rate

2 percent . . . . . . . . . . . . . . . . . . . . . . . + 6.2Assume alternative revenue/cost

figures. . . . . . . . . . . . . . . . . . . . . . . . . – 9.9Assume alternative revenue/cost

figures and 100 percent EMSstimulation . . . . . . . . . . . . . . . . . . . . . – 23.7

Moderate Generation ii EMS growth alternativeUnderlying mainstream growth rate



figures and 100 percent EMSstimulation . . . . . . . . . . . . . . . . . . . . . —1 1.8

Slow Generation ii EMS growth alternativeUnderlying mainstream growth rate



figures and 100 percent EMSstimulation . . . . . . . . . . . . . . . . . . . . . – 4.3

+ 18.1

+ 7.4

+ 4.9

- 9 . 2

+ 20.4

+ 2.6

– 14.3

+ 18.1

+ 5.0

– 9.1

+ 14.6

+ 8.4

+ 0.1( + ) = projected first-class mail rate increase( -) = projected first-class mail rate decreaseUnless otherwise indicated, 2-percent underlying mainstream growth rate isassumed.

SOURCE: Office of Technology Assessment; based on data from figs. 9 and 10.

jected rate decreases are even larger for thealternative representing very high GenerationII EMS growth. Even the slow growth alter-native would not require a rate increase in2000.

In comments to OTA, the Department ofJustice (DOJ) has expressed the view thatUSPS could not successfully charge a differentrate for delivery of Generation II as comparedto conventional first-class mail, either legallyunder the Postal Reorganization Act’s rateset-ting requirements or as a practical operationalmatter. Both the Postal Rate Commission andDOJ reviewers believe that OTA’s alternativerevenue/cost assumption is the most likelyrate basis, absent a change in USPS pricingpolicies.

All of the foregoing projections assumeother variables are held constant, includingvolume. If first-class volume was sensitive torate increases, a volume reduction could resultwhich might in turn necessitate further rateincreases, and so on. A 20- to 30-percent in-crease (net of inflation) in first-class rates couldbe enough to adversely affect the competitiveposition of first-class mail. If second- andthird-class volumes declined significantly, per-haps due to competition from alternative de-livery services, additional pressure on rateswould be generated. Also, all of the rate pro-jections are in constant 1980 dollars and donot reflect increases due to inflation. And therevenue/cost analysis in chapter 5 assumedthat productivity improvements with respectto conventional mail would offset increases inthe cost of capital and real wages. Should thereal cost of capital or labor or both exceed theinflation rate, further upward pressure onpostal rates would be experienced. In addition,the revenue/cost analysis assumed significantcost displacement for USPS delivery of Gen-eration II EMS hardcopy output. If the costdisplacement turned out to be less, or if otherkinds of USPS EMS services resulted in a netreal cost increase, the rate projections couldbe substantially different. Finally, the rev-enue/cost analysis assumed that currentUSPS service levels would be maintained.

Ch. 6—implications for Postal Rates, Service Levels, and Labor Requirements ● 63—

However, USPS could choose to reduce serv-ice levels and labor requirements rather than

increase rates in the face of declining USPS-delivered mail volume.

Service LevelsBased on projected mail volumes for the

baseline high Generation II growth alter-native, it appears that USPS might have toreduce current service levels significantly after1995 in order to avoid real rate increases. Afterthat time, the shortfall in the first-class mailcontribution to USPS fixed costs could be-come large enough to warrant considerationof service cutbacks rather than rate increases,particularly if rate increases would furtherreduce USPS mail volume. On the other hand,if the alternative revenue/cost assumption heldup, and if Generation II EMS could contributesignificantly more per piece to USPS fixedcosts than could conventional mail, then majorreal rate increases (net of inflation) or servicecutbacks could be forestalled until past 2000.This would also be the case if the underlyingmainstream growth was 3 percent rather than2 percent.

Present USPS service levels are summarizedin table 11. Since the USPS commenced opera-tions on July 1, 1971, the number of days ofdelivery per week has remained the same. Thenumber of post offices, branches, and stationshas declined at a very slow rate. For example,over the last 5 years, the number of post of-fices declined at a rate of 1.4 percent per year.3

In contrast, the number of city delivery pointshas increased by about 21 percent since 1971,and the number of rural delivery points byabout 50 percent.4 Since USPS is obligated bylaw to provide mail service to all business andresidential addresses in the United States, thenumber of delivery points has expanded alongwith growth in population and in populatedareas. USPS has instituted some measures,such as the use of cluster mailboxes, to limitthe increase of delivery points in newly popu-

3AnnuaJ Report of the Postmaster Gener~ fiscal 1980, p. 31and Annual Report of the Postmaster Gener4 fiscal 1979, p. 31.

‘Annual Report of the Postmaster General fiscal 1980, p, 10.

Table 11.— USPS Service Levels, Fiscal Year 1980

Service criteria Level of service

Days of delivery per week . . . . . . . . . 6Number of post offices . . . . . . . . . . . 30,326Number of branches and stations . . 9,160Number of city delivery points . . . . . 68.5 millionNumber of rural delivery points . . . . 14.7 millionOvernight delivery of local area

mail . . . . . . . . . . . . . . . . . . . . . . . . . 95 percent on time2-day delivery of 600-mile radius .

mail . . . . . . . . . . . . . . . . . . . . . . . . . 86 percent on time3-day delivery of cross-country

mail . . . . . . . . . . . . . . . . . . . . . . . . . 87 percent on time

SOURCE: Annua/ Report of the Postmaster Genera/, F/sea/ 1980, pp. 10, 11,31.

lated areas such as suburban or exurban resi-dential developments.

While OTA did not study USPS service lev-els in detail, there is some evidence to suggestthat the mail system is nearing capacity. Inother words, there may be limits to the volumeof conventional mail that can be handledwithout sacrificing quality of service. For ex-ample, a USPS task force concluded that 5-daydelivery “would have a negative impact onservice, including overtime, inconsistentdelivery, delayed deliveries, equipment short-ages, inadequate space to store accumulatedmail, and inadequate vehicle capacity.”5 Also,since 1977 when total mail volume reachedabout 92 billion pieces, first-class mail deliveryperformance has declined for all but local areamail. Ontime (overnight) delivery for local areamail has remained quite stable at about 95 per-cent, but ontime delivery (2-day) for 600 mileradius mail has decreased from 90 to 86 per-cent since 1977. Ontime (3-day) delivery forcross-country mail has dropped from about 91to 87 percent.6 The long-term performancetrends since 1973, the year when comparable

‘U.S. General Accounting Office, Implications of ElectronicMail for the Postal Service’s Work Force, Feb. 6, 1981, pp.34-35.

‘Annual Report of the Postmaster General fiscal 1980, p. 11.

64 ● Implications of Electronic Mail and Message Systems for the U.S. Postal Service—— .—-——— .— .——

delivery data first were collected, indicate thatpossibly USPS has reached its peak in termsof improvement in national delivery perform-ance with the current mail system.7 While per-formance is dropping off (for other than localarea first-class mail), delivery work hours arecreeping up again after declining for manyyears. More specifically, delivery work hoursdecreased by about 8 percent from 1970 to1978 while total USPS delivery points in-creased by 20 percent during that period. Butsince 1978, delivery work hours have increasedby about 4 percent while delivery points roseby 6 or 7 percent.8

● EMS have several implications for USPSservice levels. On the one hand, the projectedreductions in USPS-delivered first-class mailfor the baseline high Generation II EMSgrowth alternative would reduce the year 2000first-class contribution to USPS fixed costs byabout $1.5 billion and, in turn, could translateinto $1.5 billion worth of service reductions inorder to avoid rate increases. OTA did notanalyze in detail what kinds of service cutswould save $1.5 billion. However, USPS of-ficials have estimated that delivery 5 days aweek would save about $650 million (1980dollars). A 1975 General Accounting Officestudy estimated that closing 12,000 small andrural post offices would save $100 million(1975 dollars),9 and a 1976 congressional studyprojected a saving of $1.1 billion (1977 dollars)for delivery 3 days a week.10 Any of these serv-ice changes undoubtedly would be contro-versial.

‘Ibid.‘Ibid., p. 12.Comptroller General of the United States, $100 Million Cbuld

Be Saved Annualy in Postal Operations in Rural AmericajWashington, D. C., U.S. General Accounting Office, 1975.

‘“U. S., Congress, House Committee on Post Office and CivilService, The Necessity for Change, 94th Cong., 2d sess., Dec.10, 1976, p. 40.

On the other hand, Generation II EMS serv-ice could help to improve USPS performance.To the extent that excessive mail delays nowoccur in processing and sorting the largevolumes of conventional mail, Generation IIEMS could help to relieve these strains sinceGeneration II would bypass many of the proc-essing and sorting steps now required. Pre-sumably, Generation II EMS would also speedup delivery of mail sent outside the local areasince electronic transmission would be muchfaster than physical transport. A fullydeployed nationwide Generation II EMS serv-ice could reasonably be expected to providel-day delivery nationwide 95 percent of thetime. Cost savings associated with the reducedsorting, processing, and transportation re-quirements of Generation II EMS were re-flected in the 5¢/piece cost displacement (com-pared to conventional mail) assumed in therevenue/cost analysis (ch. 5).

Finally, if Generation II EMS volume washigher than the baseline and/or if GenerationII EMS made a greater per piece contributionto USPS fixed costs than conventional mail,then Generation II EMS could generateenough revenue to offset the need for servicecutbacks and would contribute to improvedUSPS performance at current service levels.

Even if service cutbacks became necessary,Generation II EMS could help USPS maintainservice to particular geographic areas or typesof customers where full service conventionalmail might be considered too expensive or nolonger cost effective. For example, location ofGeneration II EMS terminals in smaller ormore remote post offices could reduce oreliminate long distance transportation coststhat might become prohibitively high, forcingthe closure of such offices.

Labor RequirementsUSPS is a labor-intensive organization. As are primarily determined by three factors—

noted earlier, labor represents about 85 per- USPS service levels, labor productivity, andcent of total USPS costs. Labor requirements mail volumes. By making assumptions about

—

Ch. 6—lmplications for Postal Rates, Service levels, and Labor Requirements ● 6 5

service levels and labor productivity, it ispossible to estimate the labor requirements forprojected future mail volumes.

As a first step in the analysis, OTA esti-mated variable and fixed percentages for eachmajor group of USPS employees. The resultsare summarized in table 12, along with thetotal number of employees in each group asof fiscal year 1980. The variable percentage isthe labor component that varies directly withmail volume. The fixed percentage is the laborcomponent that is necessary to maintain thecurrent service levels. The variable and fixedpercentages in table 12 were derived by OTAdirectly from the USPS fiscal year 1980 Rev-enue and Cost Analysis. The variable laborpercentage is based on the direct and indirectvariable attributable cost from the USPS anal-ysis, expressed as a percentage of total ac-crued costs. The fixed labor percentage is thesum of specific fixed attributable costs plusall other institutional costs for each employeegroup. The variable and fixed labor percent-

Table 12.—Structure of the USPS Labor Force,Fiscal Year 1980

Number of Var iable FixedE m p l o y e e g r o u pa e m p l o y e e s p e r c e n t a g e p e r c e n t a g e

H e a d q u a r t e r s e m p l o y e e s 2,798 0 1000/0Regional and other field

units . . . . . . . . 6,228 0 100I n s p e c t i o n s e r v i c e . . . 5,242 0 100P o s t m a s t e r s . . . . . . 28,967 190/o 81Post office supervisors

and technicalpersonnel . . . . . 36,481 52 48

Subtotal . . . . 79,716Post office clerks and

mail handlers . . . . . . . 303,560 86 14City delivery carriers and

vehicle drivers . . . . 193,730 50 50Rural delivery carriers . . . . 53,069 27 73Specia l del ivery

messengers . . . . . . . 2,502 39 61Bui ld ing and equipment

maintenance personnel 29,409 45 55Vehic le maintenance

f a c i l i t y p e r s o n n e l . 4,837 29 71

Total . . . . . . . . . . . . . . . . 666,823 6 1 % 3 9 %

alncludes fu1l- and part-time employees.bB=ti on Annual Report of the Postmaster General, Fiscal 1980, P. 31, and data

supplied by F. L. Ben Kinney, USPS Manager of Financial Planning.cB~ed on USPS , Fyao Revenue and Cost Analysis, Cost Se9menfs and Com-

ponents Workpaperj pp. 742.


ages were reviewed with USPS and found tobe reasonable.

The overall cost split for the entire USPSlabor force was calculated to be 61 percentvariable and 39 percent fixed. This fixedpercentage is somewhat higher than the 36percent fixed for total USPS costs used in theOTA revenue/cost analysis (ch. 5). The 36 per-cent reflects the lower fixed percentages of thecost components for transportation (primari-ly air, rail, and highway) and building occupan-cy (rents, fuel, and utilities). The 61/39 splitfor labor is reasonably consistent with actualdata on the relationship between USPS laborwork hours and mail volume changes collectedduring 1979 and 1980.11

As a next step, OTA calculated labor pro-ductivity index values at 5-year intervals from1980 to 2000. An average productivity im-provement of 1.5 percent per year was as-sumed as the baseline. USPS labor productivi-ty improved by roughly 3 percent annuallyduring the 1970’s, and a goal of 3 percent im-provement per year has been announced. Butthis does not appear to be realistic in view ofthe fact that most productivity improvementfrom automation and mechanization has al-ready been realized. Even the expanded ZIPcode program, known as ZIP + 4, wouldrealize a cumulative labor productivity im-provement of only 2.3 percent, according toUSPS estimates.

12 For comparison purposes,labor force requirements were also calculatedfor 3.0- and O. O-percent productivity improve-ment per year.

Employee groups with 100 percent fixedcosts would not vary with mail volume. A totalof only 14,268 employees, or about 2.1 percentof the total USPS labor force, fall into thiscategory. Included here would be head-quarters, regional, and inspection serviceemployees. This assumes that current servicelevels are maintained. If, for example, a signifi-cant number of major post offices were closed,then some portion of the costs for these em-ployee groups would become variable. In this

1lRobert W. Anthony, et al., op. cit., pp. 55-56,12* Ch. 2, footnote ~.

—..


case, the fixed percentage for postmasterswould be lower than the 81 percent shown intable 12.

Costs for the 303,560 clerks and mail han-dlers (full- and part-time) would vary sig-nificantly with conventional mail volume. Theclerks and mail handlers represent the oneemployee group with very limited participa-tion in a Generation II EMS service. Genera-tion II would bypass many of the traditionalmail sorting and processing functions per-formed by clerks and handlers, but would stillrequire physical delivery (whether by city orrural carriers). The cost split for the clerks andmail handlers is 86 percent variable/14 percentfixed. Most other employee groups would varysignificantly with total USPS-delivered mailvolume (conventional plus Generation IIEMS). This would include supervisory, main-tenance, and technical personnel plus the city,rural, and special delivery employees. A totalof 320,028 employees, or about 48 percent ofthe total USPS labor force, are in this cat-egory. This includes part-time and casual em-ployees who function in part to help the USPSaccommodate to fluctuations in mail volume,for example, at peak holiday mailing periods.

The equation used by OTA to calculate laborrequirements, given the variable labor cost,labor productivity, and projected mail volume,is provided in appendix C. In order to simplifythis analysis, clerks and mail handlers wereassumed to vary with total USPS-deliveredmail volume, not just with conventional mail.This will tend to understate the projectedchange in requirements for clerks and mailhandlers and in the total USPS labor force.

Figure 11 presents the projected overallUSPS labor force changes for the years 1985,1990, 1995, and 2000 under the baseline highbut plausible Generation II EMS growth alter-native, and assuming 1.5-percent annual laborproductivity improvement. The projected in-crease or decrease, expressed as a percentagechange in the total USPS labor force, is shownfor each of the various sensitivity assumptionsused in chapter 4 to develop the USPS-de-livered mail volumes on which the labor re-

quirement calculations are based. (See fig. 7for the corresponding mail volume projec-tions.) The projected labor force reductionranges from 2.7 to 22.8 percent in 1995, andfrom 13.8 to 34.3 percent in 2000. The basecase (assuming 2-percent underlying mail-stream growth) falls roughly in the middle ofthe range, with a 12.2-percent labor forcereduction projected for 1995 and a 23.3-per-cent reduction projected for 2000.

For the base case, this translates into areduction of 81,352 employees by 1995 and155,370 employees by 2000. By comparison,between 1971 and 1980 the USPS labor forcedeclined by roughly 65,000 employees. Thus,the projected rate of reduction under the basecase (high but plausible Generation II EMSgrowth, 2-percent underlying mailstreamgrowth, 1.5 percent per year labor productivi-ty improvement) over the next two decadeswould be higher than the rate of reduction ex-perienced during the last decade. However, asshown in figure 11, if the underlying mail-stream growth is 3 percent rather than 2 per-cent, the labor force reduction in 2000 wouldbe 13.8 percent or 92,022–significantly lessthan the rate of reduction experienced in the1970’s. On the other hand, if the underlyingmainstream growth rate turned out to be 1 per-cent or if USPS did not deliver industry Gen-eration II hardcopy output, the labor forcereduction would be about 30 percent, whichtranslates into 200,000 employees.

Of course, if 3-percent-per-year labor pro-ductivity improvement is assumed, the pro-jected labor force reductions would be evenhigher or, if zero labor productivity improve-ment is assumed, labor force reductions wouldbe lower. The sensitivity of labor force re-quirements to both labor productivity im-provement and underlying mainstream growthrate is illustrated in table 13 for the highGeneration II growth alternative. As shown,with 3-percent annual growth in the underly-ing mainstream and a zero-percent annual laborproductivity improvement, no reduction-andindeed a small increase—in the labor force re-quirement is projected for 2000. At the otherextreme, with l-percent underlying mail-

Ch. 6—Implications for Postal Rates, Service Levels, and Labor Requirements ● 67

Figure Il.–Sensitivity Analyses of Projected Changes in USPS Labor Force Requirements for 1995 and 2000Assuming High But Plausible Generation II EMS Growth and

1.5-Percent Annual Labor Productivity Improvement

I 1 Ji 1990 1995

Year

Sensi t iv i ty assumpt ionsChange in USPS labor force

(assume 20/. underlying mainstream(percent of 1980 total employees)

growth un less o therwise ind icated) 1985 1990 1995 2000

3% growth in underlying mainstream . . . . . . . . . . . . . . . . . . . + 5.7 + 5.6 – 2.7 – 13.8100°/0 stimulation of additional Generation II EMS . . . . . . . . + 2.0 + 2.1 – 5.0 – 17.7EFT growth rates cut in half . . . . . . . . . . . . . . . . . . . . . . . . . . . + 0.7 – 1.5 – 9.4 – 18.3Generation Ill EMS delayed by 3 years . . . . . . . . . . . . . . . . . . + 1.4 + 0.1 – 6.6 - 1 8 . 92°/0 growth in underlying mainstream . . . . . . . . . . . . . . . . . . . + 0.6 – 2.4 – 12.2 - 2 3 . 3Generation Ill EMS accelerated by 3 years . . . . . . . . . . . . . . . – 0.8 – 6.8 – 17.6 - 2 6 . 7USPS does not deliver industry Generation II . . . . . . . . . . . . . – 0.7 – 6 . 8 – 19.2 – 29.010/0 growth in underlying mainstream . . . . . . . . . . . . . . . . . . . – 3.7 – 9.4 – 20.1 – 30.9Second-/third-class mail diversion to alternative delivery. . . – 6.0 -12.2 – 22.8 – 34.3

SOURCE Off Ice of Technology Assessment

——


Table 13.—Changes in USPS Labor ForceRequirements for High But Plausible Generation IIEMS Growth Alternative, Year 2000, As a Functionof Underlying Mainstream Growth Rate and Labor

Productivity Improvement per Year

Labor productivity

Underlying mainstream improvement per year

growth rate per year 0% 1 .5% 3.0%

3% . . . . . . . . . . . . . . . . . + 2.7% – 13.8% – 25.6%2% . . . . . . . . . . . . . . . . . – 10,1 – 23.3 – 32.91 0/0 . . . . . . . . . . . . . . . . . – 20.4 – 30.9 – 38.6NOTE. Changes In labor force requirement expressed as percentage In.

crease ( + ) or decrease ( – ) from 1980 USPS labor force

SOURCE: Office of Technology Assessment

stream growth and 3-percent productivity im-provement, a very substantial 38.6-percentreduction in labor force is projected in 2000.However, only about half of this reductionwould be due to declining mainstream volume,with the other half due to labor productivityimprovement.

Labor requirements could be calculated forevery separate employee group under variousassumptions in any future year specified. OTAestimated detailed labor force requirements in2000 for the base case (high but plausible Gen-eration II EMS growth alternative, 2-percentunderlying mainstream growth, 1.5-percentlabor productivity growth) and for the base

case modified to assume a 3-percent annuallabor productivity improvement. Actually, asindicated in table 13, in terms of overall pro-jected labor force changes, the base case isroughly the same as the 1-percent mailstream/O-percent labor productivity and 3-percentmainstream/3-percent labor combinations. The2-percent mainstream/3-percent labor produc-tivity is about the same as the 1-percent mail-stream/1.5-percent labor productivity combi-nation.

As summarized in tables 14 and 15, the pro-jected net reduction in the total USPS laborforce for these two cases is about 158,000 and223,000 employees, respectively, with a re-maining labor force in 2000 of roughly 506,568and 452,813 employees.

The projected labor force reductions are notspread evenly among all employee groups.Almost two-thirds of the reductions would befrom the post office clerks and mail handlersgroup--99,900 and 141,000, respectively, forthe two cases shown in tables 14 and 15. Thisis because the work of clerks and mail handlersis more directly related to mail volume thanany other employee group. In addition, USPSdelivery of Generation II EMS probablywould not require many of the sorting and

Table 14.—Projected USPS Year 2000 Labor Force Requirements by Employee Group, Assuming High ButPlausible Generation II EMS Growth, 2-Percent Underlying Mainstream Growth, and 1.5-Percent Labor

Productivity Improvementa

Percentage Number of Employeeschange reduction from year 2000

Employee group year 2000b fiscal year 1980 total

Headquarters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0 0 2,800Regional and other field units . . . . . . . . . . . . . . . . . . . . . . 0 0 6,220Inspection service . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0 0 5,240Postmasters. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . –7.3”!0 –2,120 26,900Post office supervisors and technical personnel. . . . . . . – 19.9 –7,260 29,220Post office clerks and mail handlers . . . . . . . . . . . . . . . . . --32.9 –99,900 204,000City delivery carriers and vehicle drivers . . . . . . . . . . . . . – 19.1 –37,000 157,000Rural delivery carriers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . – 10.3 –5,470 47,600Special delivery messengers . . . . . . . . . . . . . . . . . . . . . . . -- 14.9 –373 2,130Building and equipment maintenance personnel . . . . . . -- 17.2 –5,060 24,400Vehicle maintenance facility personnel . . . . . . . . . . . . . . – 11.1 –537 4,300

Totals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . - – 158,000 510,000aRou@lY ~qulvalent tO s.p~rceflt mailstreafn growth/3-percent productivity improvement and l-percent mainstream growthlo-percent pro-

bductlvity improvement.Based on total year 2000 USPSdellvered mail volume of 88.5 bllllon.

NOTE: All numbers rounded to three significant figures.SOURCE: Off Ice of Technology Assessment.

Ch. 6—implications for Postal Rates, Service Levels, and Labor Requirements ● 69

Table 15.—Projected USPS Year 2000 Labor Force Requirements by Employee Group, Assuming High ButPlausible Generation II EMS Growth, 2-Percent Underlying Mainstream Growth, and 3-Percent Labor

Productivity Improvementa

Percentage Number of Employeeschange reduction from year 2000

Employee group year 2000b fiscal year 1980 total

Headquarters employees . . . . . . . . . . . . . . . . . . . . . . . . . . 0 0 2,800Regional and other field units . . . . . . . . . . . . . . . . . . . . . . 0 0 6,220Inspection service . . . . . . . ., ... , ... , , . . . . . . . . , . . . . . 0 0 5,240Postmasters. . . . . . . . . . . . . . . . . . . . . ., . . . . . . . . . . . . . . – 10.20/0 –2,960 26,000Post office supervisors and technical personnel . . . . . . . – 28.0 – 10,200 26,300Post office clerks and mail handlers . . . . . . . . . . . . . . . . . –46.4 – 141,000 163,000City delivery carriers and vehicle drivers . . . . . . . . . . . . . – 27.0 –52,300 141,000Rural delivery carriers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . – 14.6 – 7,750 45,300Special delivery messengers . . . . . . . . . . . . . . . . . . . . . . . – 21.0 –525 1,980Building and equipment maintenance personnel . . . . . . – 24.3 – 7,150 22,300Vehicle maintenance facility personnel . . . . . . . . . . . . . . – 15.6 –755 4,080

Totals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . – 223,000 444,000

aRoughly qe uivalent to 1-percent mainstream growthll .5-percent productivity improvement.bBased on total year 2000 USPS-delivered mail volume of 88.5 billion.

NOTE: All numbers rounded to three significant figures.

SOURCE Office of Technology Assessment

processing functions now performed by clerksand mail handlers. A lesser but still significantpercentage (almost one-quarter) of the totalemployee reductions would be in the city de-livery carrier group.

Overall, the clerks and mail handlers couldbe reduced by about 33 to 46 percent of their1980 complement, respectively, for the twocases in tables 14 and 15. Post office super-visors could be reduced by about 20 to 28 per-cent, and city delivery carriers by 19 to 27percent.

The projections in tables 14 and 15 do notreflect any addition of employees for operationof USPS Generation II EMS facilities. RCAestimated that 5,000 new USPS jobs wouldbe created if a USPS Generation II EMS serv-ice were fully deployed.13 OTA did not in-dependently verify this estimate. However, itseems clear that, qualitatively, any increasein USPS employees for Generation II opera-tions would be very small compared to the pro-jected employee reductions.

—.19The RCA study conducted for USPS estimated that 5,000

new USPS jobs would be created through full deployment ofthe EMSS. See RCA, IWctronic Message Service System:Growth and Economic Analyses, 1977.

Up to this point, all the discussion and anal-ysis of projected labor force requirements hasbeen in terms of the high but plausible Genera-tion II EMS growth alternative. OTA con-ducted additional sensitivity runs to deter-mine if the labor force changes would besignificantly different under very high, mod-erate, and slow growth alternatives. The re-sults are summarized in table 16. Basically,the net reduction in the overall USPS laborforce would not change unless the 100-percentEMS stimulation factor applies. That is, ifeach Generation II EMS message stimulatesa new Generation II message, then the overalllabor force reduction would be somewhat lessfor all alternatives. The labor force reductionwould then be smallest (–15.3 percent) for thevery high Generation II growth alternativeand largest ( – 20.4 percent) for the slowgrowth alternative.

Whether or not these labor force reductionscould be handled through attrition dependslargely on future USPS retirement, quit, andnew hire rates. In recent years, retirementshave averaged about 4 percent of the full-timelabor force and about two-thirds of all separa-tions.” However, over the last 10 years, net

“U.S. Department of Labor, “The Labor Impact of Insti-tuting Electronic Mail Systems in the United States PostalService” pp. 9-10, paper prepared for the 1979 Presidential)Review Memo on USPS electronic mail policy.

97-918 0 - 82 - 6

. . ..


Table 16.—Projected Year 2000 USPS Labor Force Reductions Assuming 2-PercentUnderlying Mainstream Growth and 1.5. Percent Productivity Improvement

USPS mail volumein billions of pieces

Conventional Generation II Labor forceEMS alternative mail mail Total reduction

High but plausible growth . . . . . . .Very high growth . . . . . . . . . . . . . .Moderate growth . . . . . . . . . . . . . .Slow growth . . . . . . . . . . . . . . . . . .High but plausible growthb . . . . . .Very high growthb, . . . . . . . . . . . . .Moderate growthb . . . . . . . . . . . . .Slow growthb. . . . . . . . . . . . . . . . . .

75.1.- . -- . -- - .

69.975.281.875.169.975.281.8

1 3 . 4

18.613.36.7

26.937.226.713.4

88.5 – 23.3 ”/088.5 – 23.388.5 – 23.388.5 – 23.3

102.0 – 17.5107.1 – 15.3101.9 – 17.595.2 – 20.4

aExPressed as a percentage of 1980 employee levelsbAs~umes loo-percent” EMS stimulation.

SOURCE: Off Ice of Technology Assessment.

separations (retirements plus quits minus newhires) have averaged slightly under 1 percent.This would be adequate to absorb the pro-jected labor reductions for the baseline alter-native (assuming 2-percent mainstream growthand 1.5-percent productivity improvement).But any significant drop in the retirementand/or quit rates would mean fewer new hires,restricted promotion opportunities, and lim-ited upward mobility. If the labor force reduc-tion was higher than the baseline, a furthercutback in new hires would be necessary. Theadjustment could be quite difficult for groupsthat would be affected most, such as clerksand mail handlers. All of these conditionscould adversely affect employee morale andcomplicate future contract negotiations, anddeserve serious attention by USPS manage-ment and labor unions.

There is some evidence to suggest that re-tirement rates may decline in future years. Asof 1980, the age distribution of the USPSworkforce shows a bimodal distribution withpeaks at about age 58 and age 33. This meansthat a peak has been passed in terms of thenumbers of employees reaching retirementeligibility (age 55). Of course, many employeeswait until age 62 or 65 to retire. The agedistribution suggests that, for the next 17years or so, decreasing although still signifi-cant numbers of USPS employees will be

reaching retirement eligibility each year.15 Inaddition, the removal of mandatory age 70retirement restrictions may encourage someemployees to work longer, and the poor stateof the economy could discourage retirementsand quits. Whether or not this poses a problemwould depend on the actual retirement andquit rates in the late 1980’s and early 1990’s,the time when the need for significant laborforce reductions (due to declining USPS-de-livered mail volume) is first likely to be felt.

Finally, as noted earlier, the impact of laborforce reductions would fall unevenly on thevarious employee groups. For example, themail handlers would be hit especially hard,since to a substantial extent their currentskills would not be needed in processing elec-tronic mail. This might be one employee groupwhere retraining opportunities for possibleEMS jobs might be emphasized. Another re-lated concern is that, as of late 1978, the mailhandlers as a group had one of the highestpercentages of black employment, more thandouble the USPS-wide average.16 Thus, thepossibility exists that labor reductions mightfall disproportionately on black and perhapsother minority employment. This possibilitywarrants further study.

“Ibid., p. 8.“Ibid., p. 11.

Chapter 7

Implications for theTelecommunication and

Computer Industries, EMSPrivacy and Security, and

USPS Long-Term Viability

Contents


Telecommunication and Computer Industries . . . . . . . . . . . . . . . . . . . . . . . . 73Fairness of USPS Role in EMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73Legality of USPS Role in EMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76Competition Between Generations II and III . . . . . . . . . . . . . . . . . . . . . . 77Innovation and Standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78

EMS Privacy and Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79

USPS Long-Term Viability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

Chapter 7

Implications for theTelecommunication and Computer

Industries, EMS Privacy andSecurity, and USPS Long-Term

Viability

IntroductionAlthough the primary emphasis of this electronic mail and message systems (EMS)

study is on the U.S. Postal Service (USPS) on the telecommunication and computer in-mailstream and on rates, service, and labor, dustries, EMS privacy, and the long-term via-a discussion of the effects of a USPS role in bility of the Postal Service is included.

Telecommunication and Computer IndustriesA major concern expressed in regulatory

and judicial proceedings by a number of pri-vate sector telecommunication firms, andmore recently by data-processing and comput-er firms, has been that a USPS role in EMSwould constitute unfair and perhaps even ille-gal competition with private industry.

Fairness of USPS Role in EMS

Private firms argue that USPS has the fol-lowing advantages: 1) the Private ExpressStatutes (PES), which protect certain mailservices from competition; 2) exemption fromincome taxes; 3) access to the U.S. Treasuryfor investment funds; 4) public funds appropri-ated by Congress; and 5) a cost and ratesettingprocess that is complex and difficult to under-stand, which makes cross-subsidies possiblebetween different classes of mail.

On rebuttal, USPS has pointed out that thePES protect only letter mail from competition,and then only if it is carried over postal routesand is not time sensitive. A number of compet-itive alternatives to the USPS letter delivery

services exist that are legal and apparentlyviable. These alternatives include private spe-cial messenger services; electronic messagealternatives such as telephone, telegraph,telex, and privately offered “electronic mail”services; and certain kinds of media advertis-ing (by newspaper, radio, or television) whenserving as a substitute for first-class or third-class advertising mail or direct mail solicita-tions.1

Competitive alternatives to USPS nonlettermail include local and regional private deliveryservices and successful nationwide deliveryservices such as United Parcel Service (UPS),Federal Express, and Purolater. In some mar-ket segments, such as surface delivery of par-cels, the competitive impact on USPS has beensevere. For example, in 1957, USPS deliveredabout 64 percent of total parcel volume andUPS about 36 percent. By comparison, in1977, USPS delivered only about 23 percent

‘See USPS Marketing Services Division, Competitors and(%mpetition of the USPS, vol. XII, September 1978 and up-dated yearly.

73


of total volume while UPS delivered 77percent.2

Thus, the so-called “postal monopoly” pro-vided to USPS under the PES is limited toonly a few of the many classes of mail serviceoffered by USPS. Available evidence suggeststhat even among these protected servicesUSPS market power is eroding in the face ofcompetitive alternatives, both electronic andnonelectronic.3 Nonetheless, the Departmentof Justice (DOJ), among others, believes thatthere is a significant chance that a USPS EMSoffering (such as E-COM) could be subsidizedby revenues from the USPS monopoly on de-livery of first-class letter mail.

If E-COM were priced artificially belowwhat it costs, DOJ believes that E-COMmight be used at the expense of both conven-tional first-class mailers and the taxpayers (tothe extent that USPS continues to receivepublic funds appropriated by Congress). DOJalso argues that underpricing of E-COM mightdiscourage other firms from offering a similarservice, thereby decreasing competition. Ingeneral, the DOJ position is that the existingregulatory structure and oversight process donot provide adequate safeguards against theimpacts of E-COM that could be anticompeti-tive and discriminatory.

With respect to taxes, as an independentFederal Government agency USPS is not le-gally subject to Federal or State income taxes.Whether or not this is a real competitive ad-vantage is a matter of dispute. USPS arguesthat under the Postal Reorganization Act andcurrent ratesetting procedures it is effective-ly prohibited from making a profit; thus, evenif it were subject to income taxes USPS gener-ally would pay none because it would have notaxable net income. Also, while USPS does notpay property taxes on USPS-owned property,some State and local property taxes are paidindirectly when USPS is the lessee rather thanthe owner. Finally, USPS points out that itdoes not benefit from tax advantages (such asaccelerated depreciation) available to private

‘Ibid., p. 25.‘Ibid.

firms, and that any advantage from tax ex-empt status is more than offset by the costsof service and regulatory requirements im-posed on it by the Postal Reorganization Act.4

Still, to the extent that private EMS firms areprofitable and pay income and other taxes,such taxes represent a cost not incurred byUSPS. Some firms believe that nonprofitstatus offers USPS a price advantage overprivate competitors.

On the subject of access to the U.S. Treas-ury for investment purposes, the Postal Reor-ganization Act authorizes USPS to issue andsell obligations not to exceed $10 billion out-standing at any one time. As of September 30,1980, USPS long-term debt totaled about$1.84 billion, consisting of $250 million inPostal Service bonds and $1.59 billion in notespayable to the Federal Financing Bank.5

USPS observes that “private industry obtainsvarious forms of financial aid including loansfrom the Treasury. Moreover, private firms donot operate under the same statutory or prac-tical limits on their borrowing authority asdoes the USPS.”6 In addition, the Secretaryof the Treasury has the option not to pledgethe full faith and credit of the U.S. Govern-ment for USPS bond issues if the Secretarydetermines that it would not be in the publicinterest, although the Secretary has never ex-ercised this option. Clearly, USPS competitorsdo not have comparable access to the U.S.Treasury for purposes of long-term borrowing,a factor that becomes even more significantwhen money is tight and interest rates arehigh.

USPS also receives annual appropriationsfrom the U.S. Government as authorized bythe Postal Reorganization Act. In fiscal year1980, the annual appropriations totaled $1.6billion, which included $828 million for publicservice costs and $782 million for revenue for-gone due to free and reduced rates for certainmail services.7 Some USPS competitors have

‘Sept. 18,1980, letter to OTA from Charles R. Braun, USPSAssistant General Counsel, pp. 11-12.

‘Annual Report of the Postmasta General fiscal 1980, p. 21.‘Letter from Braun, op. cit., p. 11.7Annual Report of the Postmaster General fiscal 1980, p. 24.

Ch. 7—Telecommunication and Computer Industries, EMS Privacy and Security, and USPS Long-Term Viability ● 75

argued that this constitutes an unfair publicsubsidy to the USPS. However, the revenueforgone subsidy is intended to reimburse theUSPS for the revenue given up or “forgone”as a result of providing mail service free (forthe blind and handicapped) or at a reduced rate(e.g., for library materials, nonprofit bulk mail,and classroom publications), as required by thePostal Reorganization Act.

Likewise, the public service subsidy is in-tended to reimburse USPS “for public servicecosts incurred by it in providing a maximumdegree of effective and regular postal servicenationwide.”8 In any event, the public servicesubsidy was reduced from $828 million in fis-cal year 1980 to approximately $468 millionin fiscal year 1981. * The fiscal year 1982 con-tinuing resolution provided a public service ap-propriation of about $221 million, and the Om-nibus Budget and Reconciliation Act of 1981reduced the public service authorization to$100 million for fiscal year 1983 and $0 for fis-cal year 1984. The phasing out of the publicservice subsidy minimizes or eliminates anycompetitive advantage this may have givento USPS. DOJ and some private firms haveexpressed concern that there is nothing to pre-vent a future administration and Congressfrom reinstituting the subsidy, thereby pos-sibly resulting in a significant USPS advan-tage over private industry in any competitiveactivity.

The USPS cost and ratesetting process isadmittedly complex, which has led some pri-vate firms to be concerned about possible hid-den cross-subsidies. These firms are particular-ly concerned about cross-subsidies from con-ventional mail services to EMS services; thatis, the use of revenues from conventional mailto subsidize EMS costs which would keepdown the rates for EMS services. OTA has notindependently verified USPS costs and reve-nues by class of mail. However, the Postal Re-organization Act generally prohibits cross-subsidization between classes of mail and in-

cludes the requirement that “each class of mailor type of mail service bear the direct and in-direct postal costs attributable to that classor type plus that portion of all other costs ofthe Postal Service reasonably assignable tosuch class or type.”9

In addition, all USPS rate requests are sub-ject to usually extensive and lengthy hearingsconducted by the Postal Rate Commission(PRC) at which all postal rates (for all classesof mail and service) normally are considered.Many USPS competitors and mail users par-ticipate in these hearings, along with USPS,PRC and the Officer of the Commission(charged with representing the interests of thegeneral public), and occasionally other Govern-ment agencies (such as, in the E-COM proceed-ing, the Departments of Commerce and Jus-tice). Given the statutory requirements andthe adversary regulatory process in which allinterested parties are represented (and whichitself is subject to judicial review), postalcross-subsidies would seem to be rather dif-ficult to hide. Nonetheless, there is no absoluteguarantee against cross-subsidies since the al-location of indirect and institutional costs isalways somewhat arbitrary (in any organiza-tion), and the statutory criteria included in thePostal Reorganization Act may not be neces-sarily as applicable or appropriate now as theywere when it was enacted in 1970. Some pri-vate firms are concerned that the rates initial-ly set for E-COM service do not fully reflectthe actual costs, and that at the present timethere may be a hidden cross-subsidy of E-COMby other classes of mail. OTA has not inde-pendently evaluated this concern. However,the public record indicates that the PRC ap-proved the E-COM rate of 26¢ based on an es-timated capital cost of $7.4 million and firstyear volume of 12.5 million messages (240,000per week). In comparison, the actual capitalcost of E-COM is apparently close to $39 mil-lion, with volume averaging about 25,000 mes-sages per week for the first 6 months of 1982.

81 bid., p. 20.*$1,25 b~on continuing appropriation ta USPS for fiscal year

1981 less $782 million revenue forgone subsidy.’39 USC 3622(b)(3). Institutional costs are also apportioned

by class or service based on statutory criteria.

.———.—


Legality of USPS Role in EMS

From a strictly legal perspective, some pri-vate firms have argued that a USPS role inEMS (other than delivery of hardcopy output)is beyond the mandate of the Postal Reorgani-zation Act of 1970, and is in direct conflictwith Federal Communications Commission(FCC) regulations promulgated pursuant tothe Communications Act of 1934 and with Of-fice of Management and Budget regulationsconcerning Federal Government procurementof goods and services from and competitionwith the private sector.

The Postal Reorganization Act mandatesUSPS to “provide prompt, reliable, and effi-cient service, “ “give the highest considerationto the requirement for the most expeditiouscollection, transportation, and delivery of im-portant letter mail,” and “emphasize the needfor facilities and equipment designed to create. . . a maximum degree of convenience for effi-cient postal services . . . and control of coststo the Postal Service.”10 Thus, USPS views itsuse of electronic equipment and technology asconsistent with the Postal Act and as a sim-ple extension of its prior use of, for example,automated sorting machines to carry outpost-al policy as defined in the act. From thisperspective, EMS technology would be consid-ered, along with the stagecoach, pony express,railroad, truck, and airplane, as another stepin a long succession of new technologies usedto expedite the provision of postal services. Noprivate firm or Government agency has suc-cessfully challenged the USPS interpretationon legal grounds. In its original Opinion andRecommended Decision, PRC supported theuse of EMS technology by USPS although itdiffered with USPS in the application of thattechnology. More specifically, the PRC recom-mended that USPS provide only the printing,enveloping, and hardcopy delivery functionsand not the telecommunication function.11

‘“Public Law 91-375, sec. IOl(a), (e), and (g).“U.S. Postal Rate Commission, Opinion and Recommended

lkbion on Ekxtronic Mad Classification Proposal docket No.MC78-3, Dec. 17, 1979, pp. 278ff.

The PRC Recommended Decision was basedsubstantially on its finding that “the generalobligation imposed on regulatory agencies toconsider and promote competitive policies ap-plies to this Commission."12 The PRC decisionwas also based on the clearly procompetitivepolicy of FCC and the fact that the FCC as-serted jurisdiction over the original USPSE-COM proposal, primarily on the groundsthat it included telecommunication transmis-sion functions to be provided by a telecom-munication firm (Western Union), which wassubject to FCC jurisdiction under the Commu-nications Act of 1934.13 PRC concluded thatcompetition would be best served if USPS didnot provide telecommunications. This also per-mitted PRC to avoid both a possible regula-tory impasse with FCC and any direct conces-sion of FCC jurisdiction over postal servicesper se.

USPS subsequently appealed the FCC rul-ing which asserted jurisdiction over E-COM.However, the appeal was dismissed and theFCC ruling vacated as moot by the court inview of the PRC Recommended Decision andUSPS cancellation of the Western Union con-tract.* Thus, the court did not rule on themerits, and the legal jurisdiction of FCC overUSPS involvement in EMS remains unclear.

However, regulatory developments since theFCC ruling on E-COM suggest that so-called“enhanced services” such as electronic mailmay not be subject to active FCC regulationunder title II of the Communications Act. Inother words, as long as USPS does not ownand operate its own telecommunication trans-mission system and uses telecommunicationservices of firms who are regulated as pro-viders of so-called “basic services, ” the USPSEMS offerings would not necessarily be regu-lated by FCC.14 The applicable FCC decision,known as Computer II, is still underregulatory reconsideration and judicial chal-lenge, and also may be affected by congres-

]Z1bid., ~. 52 (caps and underlining removed).“Ibid., pp. 36-51.*SW rela~d discussion in ch. 3.14Dec. 8, 1980, letter to USPS from Philip L. Verveer of FCC.


sional revision of the Communications Act. Asnoted in chapter 2, S. 898 as passed by theSenate would limit FCC jurisdiction to USPSEMS services involving USPS leasing of tele-communications from private firms, and thenonly to establishing costs of the telecom-munication portion of the EMS service and toassuring that such service is offered by a sep-arate organizational entity within USPS.

More recently, various computer service anddata-processing firms, among others, have ex-pressed concern that a USPS role in Genera-tions I or II EMS, let alone Generation III,could violate Federal Government policy (ex-pressed, for example, in OMB Circulars A-76and A-121) that the Government should notproduce goods and services otherwise avail-able from the private sector and should notcompete with private firms, except as a lastresort. USPS has procured the computer andelectronic equipment for E-COM from privatefirms, but USPS operation of E-COM is con-strued by some to constitute a computer-basedelectronic message processing service.15 It isunclear whether or not this is any differentfrom USPS owning (through purchase fromprivate manufacturers) and operating its ownfleet of nearly 120,000 vehicles as it doesnow. 16

At present, the use of computers and mes-sage processing in the E-COM service is inter-nal to USPS, and serves to convert the elec-tronic input to hardcopy output within a givenpostal facility. As long as USPS is not in-volved in telecommunication or electronic de-livery, there is no direct competition with pri-vate Generation III EMS providers. However,various Generations I and II electronic mailproviders and computer service bureaus be-lieve that USPS to some extent is competingwith them. These firms have suggested severalalternatives (discussed in ch. 8) to establishwhat would be, in their judgment, a coopera-tive rather than competitive relationship. Asmentioned in chapter 2, H.R. 4758, introduced

‘We Association of Data Processing Service Organizations,position paper on “Government Provision of Electronic Mes-sage Services, ” Feb. 16, 1982.

‘“Annual Report of the Postmaster General fiscal 1980, p. 14.

in the 97th Congress and the subject of exten-sive hearings by the House Government Oper-ations Subcommittee on Government Infor-mation and Individual Rights,17 would appearto have the effect of prohibiting USPS fromproviding data-processing or telecommunica-tion services to non-Federal entities or personsunless explicitly authorized by statute.

Competition BetweenGenerations II and III

Other implications for the telecommunica-tion and computer industries are also difficultto assess. The market penetration projectionssuggest that by 2000, even with 100-percentGeneration II EMS stimulation, GenerationIII EMS volume (end-to-end electronic mail in-cluding electronic delivery) would exceed Gen-eration II EMS volume. The sum of electronicfunds transfer (EFT) (a form of Generation III)plus Generation III EMS would exceed Gener-ation II EMS by the mid-1990’s. Under thehigh but plausible Generation II EMS growthalternative, as shown in figure 3, GenerationIII EMS would surpass Generation II EMSby 1990, and EFT plus Generation III EMSwould exceed Generation II EMS as early as1985. However, Generation II EMS volumecould still be substantial in 2000 and beyond,even though the Generation II market sharewould be declining.

To some extent, perhaps until 1990, Genera-tion II EMS would compete with GenerationIII EMS for relative but not absolute marketshare. In developing the market penetrationmodel, OTA assumed that development ofGeneration III EMS and EFT would be large-ly independent of the USPS role in EMS (shortof a role in electronic delivery or ownership andoperation of telecommunications). * Prelimi-nary review of the initial responses to theUSPS invitation for telecommunication indus-try participation in E-COM indicates that

“See statements of Philip M. Walker of GTE-Telenet andWilliam D. English of Satellite Business Systems before theOct. 5, 1981, hearings of the House Government OperationsSubcommittee on Government Information and IndividualRights.

*See app. A.


none of the major domestic Generation IIIEMS providers (such as Tymnet, GTE-Tele-net, or Satellite Business Systems) indicateda desire for dedicated access to E-COM facil-ities. Whether this is because of continuinguncertainty over E-COM rates and serviceand/or because they perceive E-COM as irrele-vant to, or as a competitive threat to, theirown marketing and product developmentplans is not known.

The firms that did respond are primarilythose providing international EMS service(such as ITT World Communications andWestern Union International) seeking to ex-pand into the domestic EMS market; those al-ready providing a domestic EMS service (suchas Dialcom, Inc., and Graphnet, Inc.) who pre-sumably see E-COM as a way to increase orat least maintain their own shares of thedomestic market; or those with telecommuni-cation expertise who wish to enter anew mar-ket (such as TRT Telecommunications Corp.).TRT officials see E-COM “as a unique oppor-tunity to participate in a new venture whichhas a very large market.”18

Of the firms who actively opposed USPS in-volvement in EMS, Graphnet is the only one(as of December 1981) that has indicated aninterest in dedicated access to E-COM. Ofcourse, other firms will be able to gain accesson a dial-up basis. One major firm with dial-up access, Western Union, has recently madeapplication for dedicated access. Various localtelephone companies have apparently ex-pressed an interest, although AT&T has not.Even firms that are primarily in the Genera-tion III business could supplement their serv-ice through dial-up access to the GenerationII E-COM.*

laRichmd Yden, TRT Telecommunications Vice President,as quoted in Michael Selz, “Electronic Mail Service Promises2-Day Delivery, ” The Tampa !lYibunq Dec. 10, 1981, p. 6-B.

*For ~mple, sa~~ Business SYSIX?mS, TYmnet, GTE! ‘dAT&T among others, already offer or have plans to develop na-tionwide networks with Generation III electronic mail capabil-ity.

Innovation and Standards

Some private firms have expressed concernthat the entry of USPS into the EMS marketmight stifle innovation and possibly leadtoward adoption of technically inferior stand-ards in the telecommunication industry. Whilethese concerns were justified to some extentwith respect to the original USPS proposal forE-COM, the protracted regulatory proceed-ings before PRC and FCC have had the effectof significantly upgrading the E-COM technol-ogy. In essence, the regulatory process in thiscase had the effect of mandating improvementin the E-COM design.

The final USPS provisions for interconnec-tion between telecommunication providers andE-COM facilities at the 25 serving post offices(SPOs) appear to be substantially responsiveto comments received from private firms. ’gUSPS is providing four different interconnec-tion arrangements for those firms desiringdedicated access to E-COM facilities, and twodifferent arrangements for dial-up access ateach SPO. The interconnection arrangementsare summarized in appendix D. The dedicatedaccess interconnection arrangements appearto meet the technical needs of most firms thatprovided comments to USPS. In addition,USPS has stated that it “will consider pro-viding additional arrangements as necessaryand feasible” and “will consider requests toaccommodate user-provided interface circuitboards.”20 The dial-up access arrangementswill permit dial-up access via any public tele-phone network.

The E-COM interconnection arrangementsuse technical standards that are currently ac-cepted in the industry. Thus, it appears thatwhile USPS cannot realistically be expectedto be a source of new standards, it is not like-

%ee USPS, “Telecommunications Connection Arrangementsfor Postal service Electronic Computer Orginated Mail (E-COM)Service and Invitation for Capacity Planning Cooperation,”Federal Registw, vol. 46, No. 199, Thursday, Oct. 15, 1981.

*“Ibid., p. 50879.

Ch. 7—Telecommunication and Computer Industries, EMS privacy and Security, and USPS Long-Term Viability ● 79

ly to impose inferior standards on the industryas long as current regulatory oversight andsafeguards are maintained.

With respect to technology, the picture isless clear. The selection of technology (e.g.,computers and printers) for E-COM was madeby RCA under contract to USPS, not byUSPS itself. Some private firms have ques-tioned whether the best state-of-the-art tech-nology was selected, particularly with respectto the printing equipment. USPS believes thatthe initial E-COM technical configuration wasthe best possible using off-the-shelf productswith proven reliability, and given the natureof the E-COM market. USPS also believes thatit can stimulate innovation in some areas, suchas advanced high-volume printing and envel-oping technology, where USPS is one of thelargest users. However, given the relativelylimited expertise of USPS in telecommunica-tion and computer technology, and the rapid

EMS PrivacyThe subject of privacy with respect to EMS

includes two components of interest. One isthe legal protection afforded such services, andthe other is the technical vulnerability of suchsystems to interception of information, andthe willingness and ability of system providersto secure these systems against such intercep-tion.21

USPS is required by law to maintain “oneor more classes of mail for the transmissionof letters sealed against inspection.”22 First-class mail, priority mail, express mail, and in-ternational letter mail are “sealed against in-spection.

The sender’s choice of the class of mail serv-ice determines whether the contents are

“This section of the report is based in part on a 1980 USPSmemorandum regarding “Applicability of Mail Privacy LegalProtections to Electronic Mail. ” For additional general discus-sion see chs. 7 and 8 of the OTA report C%mputer-Based i’Va-tional Information Systems: Technology and Publik PolicyIssues, OTA-CIT-146, Washington, D. C., September 1981.

2239 USC 3623(d),

rate of private sector development, themainstream of technological innovation ap-pears to be beyond the scope of the USPS re-search and development capability, present orplanned.

As for market innovation, USPS believesthat a Generation II offering like E-COM willhelp stimulate innovation by private firms. In-deed, as mentioned earlier, some of the smallerfirms that applied for dedicated access viewE-COM as an opportunity to get into the EMSmarket and compete with the larger, morewell-established firms. But various telecom-munication carriers and computer servicebureaus have stated that their EMS serviceinnovation will be stifled unless the relation-ship with USPS is cooperative rather thancompetitive, and have proposed alternatives(discussed inch. 8) that they believe will en-courage maximum innovation in the EMSmarket.

and“sealed

Securityagainst inspection. ” Generally it does

not matter whether the mail is physicallysealed. The message contents of a postcard arelegally or constructively sealed, as are the con-tents of letters opened in the dead letter of-fice to determine to whom they might be deliv-ered. The effect of being “sealed against in-spection” is to prohibit the mail from beingopened without a warrant, or any use or dis-closure of information obtained in the courseof opening a sealed letter without a warrant.The Supreme Court has held that mail whichis sealed against inspection must be consid-ered as though it had been retained in thesender’s home. It cannot be opened withoutthe consent of the sender or addressee, exceptunder the authority of a search warrant issuedby a court upon probable cause.23

Access to addressor-addressee information,however, is possible without a warrant. Access

23Exp=& Jac~oa 96 U.S. 727, 732-733 (1877); uni~ s~~sv. Van Jkeuwen, 397 U.S. 249, 251-252 (1970).


to the information contained on the envelopeof a letter can be secured by convincing desig-nated postal officials that the “mail cover”--as this access is called-is needed to locate afugitive, to obtain evidence of a crime, or toprotect the national security. According toUSPS, the number of mail covers is decliningand the amount of mail that comes under amail cover is very small.

The privacy protection afforded to mail con-tent and addressee-addressor informationpassed through a telecommunication systemis less clearly established. First, it does not ap-pear that the postal statutes apply in fullmeasure to information when it is in electronicform, perhaps even if the electronic system isoperated by or for USPS and/or if the informa-tion is ultimately to be printed out and deliv-ered as first-class mail. This is because of adistinction between information in tangible orcorporeal form and information that exists in“incorporeal” form. It is only certain that thepostal statutes apply when the letter is a tan-gible object.

Electronic communication is afforded ameasure of protection by other statutes, butthe degree of protection is presently somewhatless than that afforded by the postal statutes.Further, the application of these other statutesis confused. Section 605 of the Communica-tions Act of 1934 prohibits the unauthorizeddisclosure of any communication by wire orradio. However, the enforcement of legal pro-tections is more difficult when mail is in elec-tronic form than in physical form and underthe direct supervision and control of USPS.

It seems clear that the postal privacy lawsand regulations would apply when an elec-tronic message is printed out in hardcopy format a postal facility, or printed out elsewhereand deposited into the USPS mainstream. Thehardcopy output when delivered over postalroutes would remain fully protected as longas it remains in the mainstream. Thus, thephysical delivery of hardcopy output and theprinting and enveloping of the EMS outputat postal facilities would be protected.

However, the telecommunication portion ofthe EMS service when provided by privatefirms appears to be subject only to the Com-munications Act and not to the Postal Reorga-nization Act. Therefore, for services likeE-COM, unless the electronic input of mes-sages to USPS were considered to be an inte-gral part of the service and under USPS juris-diction, there would seem to be no obviousbasis for applicability of the Postal Act.

On the other hand, where USPS providesthe telecommunication (as well as the printing,enveloping, and delivery), the electronic por-tion of a Generation II service could conceiv-ably be protected. USPS notes that the “ques-tion of whether these (postal) laws would applyto the electronic portion of any electronic mailservices offered by the Postal Service has nev-er been authoritatively tested, ” but finds that“there is little in the laws, however, to sug-gest that they would not.”24 In other words,according to USPS the electronic signals ap-parently could be construed to represent mailin postal custody even during transmissionover the telecommunication portion of a postalEMS service. However, some independent pri-vacy analysts dispute this interpretation andbelieve that the postal laws do not apply tothe telecommunication portion of an electronicmail system.

With respect to security, legal safeguardsmay offer less than total protection if messagecontents can be intercepted by third partieswith little risk of detection. Thus, securitymeasures are intended to safeguard messagestransmitted through electronic systems to pro-tect against eavesdropping. At the presenttime, it is left to the telecommunication car-riers to determine the degree of physical andelectronic protection to be provided. Some car-riers offer the user the option of encryptingdata, and it maybe that a market or a require-ment for such protected communication willevolve.

z4usps memorandum, “Mail Privacy,” op. cit., pp. 5-6.


The nature of an EMS service raises somesecurity concerns beyond those encounteredin conventional mail delivery. Because mes-sages may be stored for some time (1 week inthe case of E-COM), there is the potential foraccess to an historical file of traffic. Also, EMSsystems could easily produce extensive dataon sender-addressee patterns. Finally, thehardcopy output of EMS systems maybe vul-nerable to unauthorized inspection at the pointof printing and/or enveloping. None of thesetypes of security intrusion can be performedas easily or as efficiently in the conventionalmail system.

In these respects, electronic mail is more vul-nerable than conventional mail. Any EMS al-ternative that involves telecommunications,whether offered by USPS or by private firms,faces the threat of interception and monitor-ing of telephone, microwave, and/or satellitetransmissions. While available data encryp-tion technology can help to secure telecommu-nication systems, most transmissions at pres-ent are unencrypted and therefore intercep-table. To the extent EMS services includegrowing volumes of sensitive personal, busi-

ness, and financial information, the incentivesto intercept such messages would increase.Some security experts have recommendedthat USPS provide, in cooperation with tele-communication carriers, an ‘‘electronicallysealed” message service that offers protection(through encryption) at least equivalent tothat of conventional first-class mail.

Electronic mail is also vulnerable to securi-ty threats at the electronic switching and com-puter locations (including printing and en-veloping functions). The security of conven-tional mail is protected by sealed envelopes,diligent monitoring of postal employee activ-ities, locked delivery and route mailboxes, and,as discussed earlier, a variety of postal stat-utes that provide criminal sanctions for unlaw-ful intrusion by postal employees or privateparties. Additional new security measures willbe necessary at switching and computer cen-ters involved in providing electronic mail.25

“For further discussion of privacy and security, see chs. 4and 5 of the OTA background paper on Selected ElectronicFunds Transfer Issues: Privacy, Security, and Equity,OTA-BP-CIT-12, March 1982.

USPS Long-Term ViabilityThe results of the OTA analysis indicate

that, regardless of what role USPS plays inGeneration II electronic mail, reductions inconventional mail volume due to diversion toGeneration III EMS and EFT could reach sig-nificant levels by 2000. The threat to conven-tional mail could come even sooner if Genera-tion III EMS services (all electronic) developfaster than currently anticipated, if the under-lying growth in the mainstream is less than thehistorical average, or if diversion of second-and third-class mail to alternative (nonelec-tronic) delivery services increases significantlybeyond current levels.

Moreover, almost surely by 2000, probablyby 1995, and perhaps as early as 1990, Genera-tion III EMS and EFT are likely to catchupto and pass Generation II while it begins to

decline. At that point, the volume and revenue“cushion” from Generation II EMS would bereduced, and significant rate increases and/orservice and labor force reductions would belikely in order for USPS to maintain a break-even operation without increased publicsubsidies.

Should Congress concern itself about thispossibility now? While the market penetrationprojections could change somewhat given dif-ferent assumptions, the only kinds of changesthat could radically alter the projections wouldbe a growth rate in the underlying mainstream50 or 100 percent above the historical average,or a significant delay in the development andintroduction of Generation III EMS and EFTservices. Neither of these seems very likely inview of aggressive private sector Generation


III EMS activity and the continuing economictrends that work in favor of electronic mailand against paper-based mail, especially forfirst-class letter mail.

For an important institution the size ofUSPS, 15 or 20 years is not an excessive leadtime for planning an orderly transition. It canalso be argued that changes are taking placeso fast in the so-called “communications revo-lution” that by the time USPS actually experi-enced significant mail diversion, it would bemuch more difficult to adjust if steps are nottaken in advance to avoid a crisis situation.For example, while normal attrition may beable to accommodate any necessary labor forcereductions over the next 10 or 15 years, afterthat time necessary reductions may becomerather severe, particularly for volume-sensitivegroups of employees such as mail handlers,clerks, and part-time employees. Maintaininggood employee morale and career continuitywould be difficult at best under these circum-stances.

As another example, significant mail diver-sion could undermine the ability of USPS tocarry out its primary mandate “to providepostal services to bind the Nation togetherthrough the personal, educational, literary,and business correspondence of the people.”26

Reductions in USPS-delivered mail volumecould generate severe financial pressureswhich would force service and labor cutbacks.This could translate, for instance, into reduc-tions in the number of days of delivery, num-ber of collection points, or number of post of-fices. Such cutbacks could seriously disadvan-tage some postal customers who may not haveaccess to satisfactory electronic alternatives,or whose mailing needs are not amenable toelectronic transmission.

On the other hand, the projections in chap-ster 4 indicate that USPS still is likely to havea significant though reduced volume of con-ventional nonelectronic mail in 2000—perhaps70 billion to 75 billion pieces.27 Thus, it would

‘“Public Law 91-375, sec. IOl(a).~~~ver~ studie9 have concluded that a significant volume

of paper-based mail will continue almost indefinitely. See HenryB. Freedman, “Paper’s Role in an Electronic World,” The fitur

seem that, with orderly planning, enough ofthe basic USPS infrastructure could be main-tained to provide adequate conventional mailservices, although at reduced service levels.Also, new ways might be identified for USPSto carry out some of the elements of its statu-tory mandate, such as to “provide prompt,reliable, and efficient service to (postal)patrons in all areas and to render postal serv-ices to all communities,”28 and to “provide amaximum degree of effective and regular post-al services to rural areas, communities, andsmall towns where post offices are not self-sus-taining.” 29

Generation II EMS might be able to helpUSPS maintain adequate service levels torural and less populated areas that would beunable to sustain cost-effective conventionalmail service at present levels under reducedmail volume. Indeed, in some rural and veryremote areas where even Generation II EMSdelivery might be cut back, USPS might con-sider contracting with Generation III EMSproviders to assure regular electronic deliveryto individual homes and offices or, at a min-imum, perhaps to provide self-service elec-tronic hardcopy printers in post offices orother public locations. Of course, GenerationIII EMS providers may find it profitable toprovide such services on their own withoutany USPS involvement. These possibilitieswarrant further research.

Generation II EMS might also help USPSmaintain reduced rates for certain classes ofmail, such as educational and nonprofit mail-ings, that have been partially subsidized bythe annual revenue forgone appropriation frompublic funds. Even if the revenue forgone sub-sidy is reduced, the cost advantages of Genera-tion II EMS over conventional mail might per-mit the continuation of a lower rate to thosemany nonprofit and educational organizationsthat depend on the mail for their livelihood.

is~ October 1981, pp. 11-16; and Robert W. Anthony, et al.,An Exploratory Amemment of timputer Asskted Makeup andImaging $w~ The George Washington University Programof Policy Studies in Science and Techology, Washington, D. C.,Jan. 31, 1980.

tafiblic Law 91-375, %. lol(a).“public Law 91-375, Sec. IOl(b).


It also seems likely that Generation II EMSservice could help to meet the needs of smallmailers of all kinds, even though the servicemay be of greatest absolute economic benefitto large mailers. For example, as of February1982, over 90 business mailers had applied toUSPS for technical certification to use E-COMservice, including such high volume mailers asShell Oil Co., Equitable Life Assurance Co.,and Merrill Lynch Pierce Fenner & Smith.However, at least three of the telecommunica-tion carriers that have applied for dedicatedaccess (ITT World Communications, TRTCommunications Corp., and Netword, Inc.) in-tend to offer batch mailings from low-volumemailers to meet the 200-message minimumvolume requirement for E-COM use. A spokes-man for ITT World Communications has indi-cated that “ITT would allow customers tomail as few as 25 letters per mailing” and thenconsolidate orders to meet E-COM volume re-quirements. 30

In the longer term, it is possible that aUSPS Generation II EMS capability, perhapscombined with a scaled-down version of the

‘OSelz, “Electronic Mail, ” op. cit.

USPS infrastructure and delivery network,could be used to provide other Federal Govern-ment services. For example, with proper inter-connection and technical enhancements, aGeneration II system might be used by USPSto provide printing and delivery of variousGovernment forms and documents.31 This kindof role would, of course, compete to some ex-tent with functions now carried out by theGovernment Printing Office, the NationalTechnical Information Service, and other Fed-eral agencies, but might prove to be more costeffective in the long run. This role might evenbe extended to include provision of abstractsof and indexes to Government forms, docu-ments, and other kinds of Government infor-mation. These possibilities, too, deserve fur-ther study.

‘] See Robert W. Anthony, Lynne Filderman, Henry Freed-man, and Henry H. Hitchcock, Strategy for Decisions: APWUand the Electronic Information Revolution, The GeorgeWashington University Program of Policy Studies in Scienceand Technology, Washington, D. C., Mar. 1, 1980; and AlfredM. Lee and Arnim H. Meyburg, The Impact of ElectronicMessage Zhnsferon USPS Operations, Working Paper No. 3,Cornell University Program on Science, Technology, and Socie-ty, September 1980. Also see “Electronic Computer OrginatedMail, ” Technology Watch vol. 2, No. 2, December 1981, p. 2.

—. —

Chapter 8

CongressionalPolicy Considerations

Contents

PageIntroduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

Provide a Clear Direction for USPS Involvement in EMS . . . . . . . . . . . . . 89Impact on USPS-Delivered Mail Volume . . . . . . . . . . . . . . . . . . . . . . . . . . 90Impact on USPS Finances. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90Impact on USPS Labor Force . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90Space in SPOs for Carrier Output Equipment . . . . . . . . . . . . . . . . . . . . . . 90Electronic Transmission to SPOs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91Electronic Delivery to Recipient . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91Interconnection and Standard Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . 92Transmission Facilities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92Marketing of EMS Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92Performance Standards for EMS Services . . . . . . . . . . . . . . . . . . . . . . . . . 93

Reduce or Eliminate Further Regulatory and Judicial Delay . . . . . . . . . . . 93Applicability of Private Express Statutes . . . . . . . . . . . . . . . . . . . . . . . . . 93Regulatory Jurisdiction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94USPS Subdivision for EMS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95

Privacy Protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 96

Maintain Oversight and Initiate Planning on USPS Long-Term Viability 97

Chapter 8

Congressional Policy Considerations

Introduction

Congressional authority over the role of theU.S. Postal Service (USPS) in electronic mailand message systems (EMS) derives in thefirst instance from the U.S. Constitutionwhich vests in Congress the power to establishpost offices and post roads (sec. 8, clause 7)and to regulate commerce among the severalStates (sec. 8, clause 3).

USPS operates as an independent Federalagency pursuant to the Postal ReorganizationAct enacted by Congress in 1970. It is sub-ject to policy direction by the USPS Board ofGovernors and to regulatory review of mailrates and classifications by the Postal RateCommission (PRC), both established by thePostal Act. Private EMS firms operate pur-suant to the Communications Act of 1934 andregulations promulgated by the Federal Com-munications Commission (FCC) under authori-ty granted by the Communications Act.

Historically, technology has been used toimprove and speed up the processing, trans-portation, and delivery of mail. Thus, therailroad, plane, truck, and automated sortingmachine have been integrated into postal op-erations. The Postal Act reiterates andstrengthens the mandate to use new facilitiesand equipment to improve the convenience, ef-ficiency, and cost effectiveness of mail service.Electronic message technology is viewed fromthis perspective as one more step in an evolu-tionary process of keeping the “post officesand post roads” up to date and competitive.

Over the last three decades, there has beena continuing revolution in computer and com-munication technology, a gradual deregulationof the telecommunication industry (the com-puter industry being essentially unregulated),and a proliferation of new and old firms offer-ing or planning to offer EMS services. Con-gress has generally encouraged this deregula-tory process, and continues to work to revise

the 1934 Communications Act to bring it inline with current technological, economic, andcompetitive realities. During the last decadeand a half, there has also been a concerted ef-fort by Congress, exemplified by the 1967“Brooks Act” amendment to the FederalProperty and Administrative Services Act andthe Paperwork Reduction Act of 1980, to im-prove the Federal Government’s managementand procurement of data-processing and re-lated telecommunication equipment and serv-ices, and to rely on the private sector for pro-vision of such equipment and services on acompetitive basis wherever possible.

In essence, technological advances havereached the point where these three original-ly independent congressional policy directionsare increasingly in conflict. USPS stands nearthe center of this conflict. Privately offeredEMS services can and ultimately will competewith and divert a significant portion of theUSPS conventional mainstream, based on themarket penetration analysis in chapters 3 and4. Absent any USPS participation, this pros-pect would likely lead to significant rate in-creases and/or service and labor force reduc-tions by the year 2000. It seems clear thatUSPS can benefit from participation in pro-viding EMS services, and indeed it can beargued (see ch. 7) that USPS has the statutoryauthority to participate as long as the finaloutput is in hardcopy letter form deliveredover postal roads. But even a minimal USPSrole—delivery of EMS hardcopy output—is ofconcern to some private firms if such deliveryis considered subject to the Private ExpressStatutes (PES). These firms believe that de-livery of hardcopy output is ancillary to com-munication services subject to the Communi-cations Act as well as the Postal Act.

A larger USPS role that involves messageprocessing and computer-based printing and

87


enveloping, as well as hardcopy delivery, ap-parently troubles portions of the computer in-dustry (particularly computer service bureaus)because of concern over potential competitionfrom USPS and the belief that the industryis willing and able to provide such services.Should the USPS role extend to the telecom-munication portions of an EMS service, thenmany telecommunication carriers would viewUSPS as a direct competitor. Even with re-spect to E-COM, where the USPS role does notinclude telecommunication, some carriers areconcerned that it was designed to accom-modate future functions (e.g., magnetic com-puter tape input) that are not presently au-thorized. These carriers believe that any USPSinvolvement in telecommunication, whetherdirectly or by resale, would be subject to thejurisdiction of the Communications as well asthe Postal Act, and would constitute the en-try of a Federal agency into competition withprivate industry. The computer service bu-reaus apparently feel the same way withrespect to USPS provision of message proc-essing.

Reaching a consensus on a role for USPS inEMS has been further complicated by jurisdic-tional conflicts between PRC and FCC, PRCand USPS Board of Governors, FCC andUSPS, and the Departments of Commerce andJustice and USPS. These conflicts have cometo a head over E-COM, resulting in legal ac-tions brought in Federal court by USPSagainst FCC and PRC, and by Justice againstUSPS. Various parties, especially telecom-munication value-added carriers, have filedbriefs in these judicial proceedings, and beforeFCC and PRC in regulatory proceedings onE-COM, raising substantive issues of rateset-ting, potential cross-subsidization, and pri-vacy, among others.

On the other hand, many of the private tele-communication and computer firms who havebeen adversaries of USPS also believe that fulldevelopment of Generation II EMS depends

on a major role for USPS, but they disagreewith USPS on what that role should be. Vari-ous mailer organizations, consumer groups,and postal labor unions see a USPS role inEMS as essential to USPS long-term viabili-ty and to maintaining, or at least minimizingany reductions in, mail services that are vitalto a large part of the U.S. population. Theypoint to the critical role of USPS in providinga universal, low-cost, nondiscriminatory na-tionwide communication service that is stat-utorily mandated by Congress.

Based on interviews with many of the stake-holders and USPS, as well as a comprehensivereview of the historical record, OTA has con-cluded that, absent congressional action, thecontroversy over the USPS role in EMS is like-ly to continue. Although the U.S. DistrictCourt of Appeals has denied a Justice petitionto block E-COM, further regulatory proceed-ings are anticipated and additional legal ac-tions are possible. With continuing uncertain-ty over the future of E-COM, and in generalof the USPS role in EMS, the prospects fora successful USPS entry into domestic EMSservices are uncertain. Some firms have in-dicated to OTA that they are reluctant tomake any major commitments until they arecertain what role USPS is going to have.Meanwhile, many of the carriers continue toput much of their research and developmenteffort into Generation III EMS, which wouldcompletely bypass USPS. In addition, USPSis unable to establish effective working rela-tionships with many private carriers and po-tential Generation II EMS users, given thecontinuing adversarial atmosphere.

Should Congress wish to take action, thereare several major possibilities. Congress could:1) provide a clear direction for USPS involve-ment in EMS; 2) reduce or eliminate furtherregulatory and judicial delay; and 3) maintainoversight and initiate planning on long-termUSPS viability. These possibilities are dis-cussed below.

Ch. 8—Congressional Policy Considerations ● 8 9

Provide a Clear Direction for USPSInvolvement in EMS

There are essentially nine major alternativesfor a USPS role in EMS. Variations on eachare possible.

1. USPS would deliver the hardcopy printedoutput of industry EMS services when desiredand at the discretion of industry. USPS wouldnot otherwise participate in EMS. This rolewould presume that hardcopy output is an-cillary to communication services subject tothe Communications Act and outside thescope of the Postal Act or PES.

2. USPS would deliver the hardcopy printedoutput of all industry EMS services when con-veyed over postal roads (routes served byUSPS), with exceptions for time sensitive let-ters. USPS would not otherwise participate inEMS. This role would be based on currentUSPS interpretation of PES.

3. USPS would deliver the hardcopy printedoutput as in 2 above, but would also permitthe location of carrier Generation II EMS ter-minal equipment on USPS premises. Thiswould be similar to the current role of USPSin Western Union’s Mailgram service, exceptthat equipment from several carriers, not justWestern Union, would be located on premises.These carriers would then be permitted to in-terconnect with USPS facilities.

4. USPS would deliver the hardcopy outputfrom industry EMS as in 2 above, and wouldalso provide printing and enveloping portionsof EMS when desired by and to meet the speci-fications of industry (within reason). Here,USPS would offer a range of options with re-spect to number of pages, paper style and for-mat, envelope logo, and possibly inserts tomeet varied needs of carriers and their cus-tomers. All carriers (defined as in E-COM toinclude all message-processing companies)would be permitted to interconnect with USPSfacilities either on a dedicated or dial-up basis.This would be similar to E-COM, except thatcarriers would be able to retain their individualidentity (through use of logo envelopes and

possibly letterhead paper) and meet a widerrange of mailer needs (through variable letterlengths and possibly inserts).

5. USPS would deliver the hardcopy outputfrom industry EMS and would provide print-ing and enveloping portions of EMS on astandardized basis available to all carriers andmailers (within reasonable limits). This wouldbe similar to the current role of USPS inE-COM. All messages would be no more than2 pages in length, be printed on identicalpaper, and use E-COM logo envelopes. Otherthan standardized business reply envelopes,inserts are not possible. Carriers would be per-mitted to interconnect as in 4 above.

6. USPS would provide printing, enveloping,and telecommunication portions of EMS andphysically deliver hardcopy output. USPSwould lease telecommunication facilities fromprivate industry, and would also provide in-terconnection for industry carriers. This wouldbe similar to the USPS role in E-COM if thereis a “demonstrated need” for USPS provisionof telecommunication as well as printing, en-veloping, and delivery.

7. USPS would provide printing, enveloping,the telecommunication portions of EMS, andphysical delivery, as in 6 above, plus electronicdelivery if there is a “demonstrated need” forcertain geographical areas that can no longersustain conventional mail service at com-parable levels. USPS would lease or contractfor telecommunication and electronic deliveryfacilities on a competitive basis from privateindustry, and would also provide interconnec-tion for industry telecommunication carriers.

8. Combination of 4 and 6.

9. Combination of 4 and 7.

All of these alternatives are technicallyfeasible. In evaluating each, Congress maywish to take into account the following con-siderations.


Impact on USPS-DeliveredMail Volume

The market penetration results (ch. 4) in-dicated that USPS-delivered mail volume (con-ventional plus Generation II hardcopy output)is one key consideration. USPS-delivered vol-ume is in part a function of the rate of Genera-tion II EMS growth and the degree of stimula-tion of the Generation II EMS market. Thefaster the rate of growth (and the earlier thetake-off) and the greater the stimulation ofnew message traffic, the larger the GenerationII EMS volume (and hence USPS-deliveredvolume), assuming that USPS delivers Gen-eration II EMS hardcopy output. There is cur-rently little consensus on the extent to whichthe various alternatives would contribute toGeneration II EMS growth and volume.

Impact on USPS Finances

The revenue/cost results (chs. 5 and 6) in-dicated that EMS cost displacement and con-tribution to USPS fixed costs are also keyconsiderations. The greater the EMS cost dis-placement (avoidance of conventional mail-stream costs) and contribution to USPS over-head, the less likely the need for service (andlabor) reductions. Again, there is lack of agree-ment between the USPS and major stakehold-ers. While Mailgram apparently provides botha substantial cost displacement and contribu-tion to fixed costs, it is not clear whetherE-COM would do likewise at current rates andin its present configuration. All parties, in-cluding USPS, agree that the RCA cost esti-mates prepared for the electronic messageservice system (EMSS) in 1977 and the orig-inal E-COM cost estimates prepared for PRCin 1978 are now outdated. A comprehensivecost review of E-COM is needed.

Impact on USPS Labor Force

Based on the chapter 6 labor requirementsanalysis, the size of the USPS labor force isdetermined principally from the volume ofUSPS-delivered mail and labor productivity.There is general agreement that USPS partic-

ipation in EMS would generate only a relative-ly small number of new jobs. However,through higher mail volumes it could offset orat least defer significant labor reductions thatwould otherwise be necessary.

There are an estimated 200 persons (125 op-erations, 50 maintenance, 25 marketing andadministrative) currently working on E-COM,and a fully deployed service (at 150 servingpost offices (SPOs) compared to the current25) is estimated to require perhaps 2,000persons.

In contrast, the additional volume fromUSPS delivery of industry hardcopy outputunder the baseline assumptions would requireabout 38,000 employees more than wouldotherwise be necessary. The additional mailvolume from 100-percent Generation II EMSstimulation and high but plausible GenerationII EMS growth would require 39,000 employ-ees more than otherwise would be needed, fora total of about 77,000 employees. Put dif-ferently, under the baseline assumptions, theyear 2000 USPS total labor force reduction isprojected at 29 percent assuming no USPSparticipation in EMS (i.e., industry delivers itsown Generation II EMS hardcopy output), butat only 17.5 percent for USPS-delivery ofGeneration II EMS output coupled with highbut plausible Generation II EMS growth and100-percent stimulation (see fig. 11 and table16, ch. 6). Absent the 100-percent stimulation,the labor force reduction is projected at 23.3percent.

Space in SPOs for CarrierOutput Equipment

A continuing issue is whether and howUSPS should provide space on USPS premisesfor carrier equipment. For the Western UnionMailgram service, USPS has agreed to locateand operate Western Union printers in 144SPOs. Other telecommunication carriers have,in the past, asked USPS for a similar arrange-ment. USPS indicates that it is willing to con-sider any serious proposals along these lines,but that none have been received. The carriers


believe that USPS is not receptive to such pro-posals, and question why Western Unionshould receive special accommodation. It isnot clear whether physical constraints andeconomies of scale would permit the locationof equipment from a large number of carriers,or that a large number of carriers would findsuch an arrangement to be cost effective. Thisquestion deserves further study.

In E-COM, USPS does not provide space inSPOs for carrier hardcopy output devices. In-stead, USPS owns and operates its own print-ing equipment and provides access facilitiesto enable carriers to interconnect directly withSPOs. USPS has purchased the interconnec-tion equipment and leases it to carriers desir-ing dedicated access at a monthly charge.1

Some carriers are not happy with the relative-ly limited capabilities of the E-COM equip-ment. According to USPS, the selection of E-COM equipment was made by RCA, and wasjudged to be the best technology available offthe shelf that met USPS requirements. Con-gress could request an independent review ofthe technology selected.

Electronic Transmissionto SPOs

The role of USPS in the telecommunicationportion of EMS has proven to be controver-sial. USPS originally proposed to initiate itsE-COM service using a telecommunicationnetwork leased from Western Union. That is,USPS would have provided telecommunica-tion as well as printing, enveloping, anddelivery. In December 1979, PRC recom-mended an alternative plan under which USPSwould not own or operate a telecommunicationnetwork, but all telecommunication carrierswould be permitted to interconnect withE-COM at the designated SPOs. In April1980, on remand from the USPS Board ofGovernors, the PRC explicitly recognized the

‘The monthly charge per SPO ranges from $102 to $412, de-pending on the type of equipment. See “TelecommunicationConnection Arrangements for Postal Service Electronic Com-puter Originated Mail (E-COM) Service and Innovation forCapacity Planning Cooperation, ” Federal Register, vol. 46, No.199, Oct. 15, 1981, p. 50882.

authority of USPS to contract with a telecom-munication carrier to transmit messages elec-tronically on behalf of USPS.2 However, PRCconditioned this authority on a showing ofdemonstrated need, a term that has not beenclearly defined but presumably implies a situa-tion where the needs of E-COM users could notbe met adequately through the telecommuni-cation services of private carriers. The Gover-nors have accepted this condition.3 However,some carriers are concerned that the ambigui-ty of this condition could be used in the futureto, in their opinion, improperly and unjusti-fiably permit USPS to contract with a carrier(or carriers) to provide electronic transmissionon behalf of USPS. Congress may wish to clar-ify the definition of demonstrated need.

Electronic Deliveryto Recipient

USPS has not proposed, nor have the Gover-nors or PRC considered, any EMS servicewhereby USPS would provide electronic de-livery of mail directly to the recipient. USPShas stated repeatedly that it “will not provide‘Generation III’ services which transmit mes-sages all the way through telecommunica-tions.” 4 However, some consumer advocatesand researchers have suggested that electronicdelivery might be justified to maintain USPSservice levels in geographic areas where con-ventional mail service could no longer be main-tained at present levels. Congress may wishto examine what, if any, conditions would con-stitute a demonstrated need for USPS involve-ment in electronic delivery (presumably bycontract with private Generation III EMSfirms). Any USPS role in Generation III wouldhave to be carefully defined to avoid either theappearance or reality of competition withprivate firms and the substantial controver-sy and opposition that would likely generate.

‘Further Recommended Decision, docket No. MC78-3, PostalRate Commission, Apr. 8, 1980.

‘I bid., pp. 4-9; Decision of the United States Postal ServiceBoard of Governors, docket No. MC78-3, Feb. 22, 1980,

4Mar. 12, 1981, letter and position paper from Edward E.Horgan, Jr., of USPS to Sen. Barry Goldwater.


Interconnection andStandard Interface

For E-COM, USPS provides interconnectionfor message input from qualifying telecom-munication carriers and users (i.e., mailers) us-ing a standard interface. Mailers may estab-lish an account directly with USPS and prepayUSPS for E-COM delivery, while arrangingseparately with a qualified telecommunicationcarrier for transmission of messages to SPOs.Alternatively, mailers may choose to deal onlywith a qualified telecommunication carrierthat acts as an agent for E-COM service.These carriers must establish an account withUSPS and prepay USPS for delivery.5

In E-COM, USPS is offering both dial-upand dedicated access. As explained by USPS,“the dial-up access facilities will permitcustomers to connect to any SPO by meansof any public telephone network, using which-ever telecommunication carriers the customerschoose. . . Dedicated access is designed forthose who wish to have exclusive access toE-COM.”6 USPS provides two standard inter-faces for dial-up access and four standard in-terfaces for dedicated access.7

At the present time, carriers appear to bereasonably satisfied with these interfaces froma technical point of view, and USPS has in-dicated a willingness to consider other inter-faces proposed by carriers. However, some car-riers are not happy with the allocation of in-terconnection lines (or ports) at SPOs betweendedicated and dial-up access, or with the totalnumber of lines available. The total numberof lines is limited by the present E-COM tech-nology. The allocation between dedicated anddial-up access is a management decision. Con-gress may wish to review whether technicalmodifications could permit more total lines(and at what cost), as well as alternative alloca-tion schemes.

“’Telecommunication Connection, ” op. cit., p. 50875.‘Ibid., p. 50879.‘Ibid.

Transmission Facilities

USPS has not proposed, nor have the Boardof Governors or PRC considered, any EMSservice whereby USPS would own transmis-sion facilities. USPS originally proposed tocontract for the use of transmission facilities,as noted above. Given the wide range of pri-vately offered transmission services and therapid change in that industry, it seems unlike-ly, even in the case of demonstrated need toprovide the telecommunication portion ofEMS, that USPS would buy rather than leaseor contract for telecommunication transmis-sion lines.

Marketing of EMS Services

Historically, USPS has been granted the au-thority (under applicable law and regulation)to market services filed and approved underthe Domestic Mail Classification Schedule.While some private firms have objected toUSPS marketing of EMS service, E-COM inits current form is considered to be a subclassof first-class mail under the Domestic MailClassification Schedule. Accordingly, USPShas already initiated marketing efforts to iden-tify customers for E-COM.8 Even if USPSwere authorized to provide telecommunica-tions, such EMS service would most likely befiled and approved as one or more subclassesof first-class mail (and other classes of mailwhere electronic transmission may be ap-propriate), and thus could be marketed byUSPS.

However, the question of how aggressivelythe participating telecommunication carrierswould market their portion of E-COM, or anyother USPS EMS service where the identityof individual firms is not retained, is an openone. Some firms have proposed the use of en-velopes (and possibly paper) with the companylogo, rather than or in addition to the stand-ardized E-COM envelopes. By maintainingg theindividual identity of participating carriers,these firms would, in theory, have greater in-centive to develop the Generation II market.

‘USPS News Release No. 53, Oct. 19, 1981, p. 3.


Performance Standards forEMS Services

The current E-COM service is designed toachieve guaranteed 2-day delivery. Since 2-daydelivery is already guaranteed for conven-tional mail deposited within 600 miles ofdestination, the time advantage of E-COM isprimarily for cross-country mail where normaldelivery is 3 days. Of course, mailers mayrealize benefits other than time; for example,reduced printing and enveloping or computerprocessing costs. Still, some carriers andmailers have argued that l-day delivery, as isavailable with Mailgram, would be preferable.

According to USPS, E-COM could achieve1-day guaranteed delivery if the number ofSPOs equipped with E-COM facilities were ex-panded from the current 25 to about 150. Thetotal capital cost is estimated at about $250million (roughly six times the current cost ofE-COM system design and implementation),a substantial investment but considerably lessthan the $1.5 billion to $2.0 billion originallyestimated by RCA for a nationwide electronicmail service system (EMSS). However, it is

questionable whether E-COM volumes wouldsupport this investment.

In 1978, USPS projected a volume of 230million E-COM messages 5 years out. A morerecent Opinion Research Corp. survey commis-sioned by USPS projected a market of 500million messages per year now and 1 billionmessages per year 5 years out. This latter pro-jection falls somewhere between the moderateand high OTA projections.

Because of the uncertainty of such projec-tions, an incremental approach to expansionappears to be warranted. For example, E-COM(or some other alternative) could be expandedin a small number of selected origin-destina-tion pairs (e.g., Washington, D. C.-San Fran-cisco, New York-Los Angeles) to test thefeasibility of and market for l-day guaranteeddelivery. An incremental approach would ap-pear to require more flexibility in the USPSdecisionmaking process (including regulatoryreview) than is presently the case. Congressmay wish to consider some changes in thePostal Reorganization Act to provide moreflexibility.

Reduce or Eliminate Further Regulatory andJudicial Delay

The most important action Congress cantake to reduce delay is to provide clear direc-tion for USPS involvement in EMS, as dis-cussed earlier. A note of caution is in order.If the direction set out is not well understoodand reasonably clear and does not reflect asubstantial consensus, further regulatorydisputes and litigation could result.

Additionally, Congress could: 1) clarify theapplicability of PES to delivery of hardcopyoutput; 2) delineate the division of regulatoryjurisdiction between PRC and FCC; 3) man-date a separate USPS entity for any EMS of-fering; and 4) establish standards for protec-tion of privacy for EMS services involvingUSPS.

Applicability ofPrivate Express Statutes

PES9 restrict the delivery of letters by

organizations other than USPS. In general,the private carriage of letters is prohibited.PES give USPS the exclusive right to carry“letters” over postal routes, with importantexceptions. One exception is an administrativesuspension for “extremely urgent” letters,which permits private carriage if at least $3or twice the applicable postage is charged or,in certain cases, if delivery is completed within

’39 U. SC. $$601-606; 18 U.S.C. jf 1693-1699, 1724. USPSregulations implementing PES require that letters to be private-ly delivered over postal roads must be covered, sealed, dated,and stamped.

94 ● Implications of Electronic Mail and Message Systems for the U.S. Postal Service——

a matter of hours. Many electronic mail serv-ices might qualify under this exception.

USPS regulations define a “letter” as a“message in writing” addressed to a “par-ticular person,” and include electronicallytransmitted messages when in hardcopy formand delivered over postal routes.10 FCC,among others, has claimed that PES do notextend to physical delivery of hardcopy out-put from electronic communications, on thegrounds that such delivery is incidental toelectronic communications as defined in theCommunications Act of 1934 and therefore isnot subject to PES.11 In addition, FCC haschallenged a USPS proposal to redefine a pre-viously granted exemption for telegrams.USPS had proposed to limit the exemptionsto telegrams “as commonly sent in the pastby other members of the public.” Other formsof hardcopy output from electronic communi-cations would not be exempt.12 Although thisproposal was subsequently withdrawn, USPShas taken the general position that its long-standing exemption for telegrams does notapply to other types of hardcopy output fromelectronic transmission.

Some private firms dispute the USPS posi-tion, but to date no party has successfullychallenged the legality and applicability ofPES to the delivery of hardcopy output as de-fined by USPS regulations. On the other hand,PES do not apply to end-to-end electronic com-munication of messages, according to recentstatements by USPS officials and USPS in-terpretation of PES. “The PES will remain ap-plicable only to ‘hardcopy’ letters.”13 “Mes-sages transmitted by wire or wireless or elec-tronically between sender and addressee arenot letters since the PES apply only to cor-poreal messages physically carried on postroutes." 14 However, as discussed earlier,“messages so transmitted which are convertedto physical form and carried over a post road

before delivery are letters,”15 and thus are sub-ject to PES.

Nonetheless, some private firms questionnot only the applicability of PES to Genera-tion II hardcopy delivery, but are also con-cerned that USPS may attempt to extend PESto Generation III EMS. Congressional clari-fication may be needed.

Regulatory Jurisdiction

Over the last 3 years, the question of whichregulatory bodies have what jurisdiction overvarious USPS proposals for offering EMSservice has been considered extensively inregulatory proceedings. In two court actions,USPS has challenged the extent of appropri-ate jurisdiction as asserted by both PRC andFCC.

By declaratory ruling, FCC asserted authori-ty under section 2(a) of the CommunicationsAct of 193416 to regulate parts of the originalUSPS plan for E-COM. This plan called forUSPS to contract for transmission and otherservices on a sole source basis with a singlecommon carrier (Western Union) .17 FCC ap-parently based its assertion on the groundsthat Western Union was already subject toFCC regulation, and furthermore, to the ex-tent that USPS offered electronic communica-tion services, it was a “person” within themeaning of the Communications Act and wastherefore itself subject to FCC jurisdiction.18

USPS petitioned the U.S. Court of Appeals forthe District of Columbia for a review of theFCC action arguing that, under the Postal Re-organization Act of 1970, PRC is the ap-propriate regulatory body and further thatUSPS is not a “person” subject to the jurisdic-tion of FCC. On October 14, 1980, the courtdismissed the USPS appeal and vacated theFCC ruling as moot for two reasons: 1) “thecontract between Western Union and the

1039 CCFOR. 152 (197o) and USPS order 71-10”IIMm. 12, 1979, letter to Louis A. Cox, USPS Gener~

Counsel, from Robert R. Bruce, FCC General Counsel, pp. 2-3.12 43 F~. R,ego 60,616 (1978); 45 Fed. Wg. 59,871 (1980)”l~Horgm letter, Op. cit.14usps, ~n~~mtatjon of PES, 1973 report, P. 7.

151bid.1047 U.S.C. $152(a).l~post~ Rate commission, docket No. MC78-3.lsFeder~ communications Commission Common Ctl.ITier

Docket No. 79-6; in the matter of request for declaratory rul-ing and investigation by Graphnet Systems, Inc., concerningthe proposed E-COM service.


Postal Service . . . that was objectionable tothe FCC has been cancelled, and 2) PRC itselfrejected several features of the Postal Serviceelectronic mail system proposal found objec-tionable by the FCC.”19 Thus, the court did notrule on the merits of the case and the legaljurisdiction of FCC over USPS involvementin the telecommunication portion of EMS re-mains unclear.

USPS also petitioned the court for reviewof that part of the PRC Final RecommendedDecision that designates E-COM as an “ex-perimental” subclass of first-class mailauthorized only through October 1, 1984.Basically, USPS claimed that PRC far ex-ceeded its authority and sought to exercise apower reserved to the USPS Board of Gover-nors.20 In June 1981, the court ruled for USPSand remanded the matter back to PRC for fur-ther consideration.21

However, two issues were still in dispute.First, PRC believed it was proper to reviewthe entire E-COM decision, not just the “ex-perimental” designation which was the sub-ject of the court proceeding. A number of com-munication carriers and others (including theDepartments of Commerce and Justice) whofiled statements with PRC took the positionthat the court, in effect, vacated the PRCRecommended Decision in toto, and thatUSPS was not authorized to proceed withE-COM on January 4, 1982. USPS maintainedthat the court’s remand, and therefore PRC’sreconsideration, extended only to the questionof “experimental” designation and that USPSwas otherwise authorized to initiate E-COMservice in January. In December 1981, PRCsuspended the proceedings, leaving the legalstatus of E-COM uncertain. On January 4,1982, USPS started E-COM service. In April1982, the U.S. Court of Appeals for the Dis-trict of Columbia denied a Department ofJustice petition to block implementation ofE-COM.

Absent clear direction from Congress, itseems likely that USPS entry into EMS willprecipitate continued regulatory (and relatedjudicial) conflicts. Congress, through legisla-tion or otherwise, could clarify regulatoryjurisdiction over USPS involvement in EMS.For example, this might take the form of theamendment to S. 898 which stipulates thatFCC shall establish costs for the telecom-munication portion of any USPS EMS serviceand shall assume that any such telecommu-nication services are offered by a separateorganizational entity within USPS. Apartfrom these two provisions, the amendment toS. 898 states that FCC shall not regulateUSPS.22 As another example, Congress mightclarify-through an amendment to the PostalAct—the extent of PRC jurisdiction over aUSPS role in EMS.

USPS Subdivision for EMS

As discussed in chapter 7, a number of pri-vate firms and other parties have expressedconcern that USPS involvement in EMSwould constitute unfair competition betweenan independent Government agency and theprivate sector. This concern focuses in part onthe possibility that USPS might use public ap-propriations or revenues from other USPSservices to cross-subsidize EMS services. InJuly 1979, the White House proposed the crea-tion of a separate USPS subdivision for EMSservice in order to make cross-subsidies easierto detect and prevent. The original WhiteHouse proposal suggested “a separate entityfor accounting and ratemaking purposes.”23

H.R. 2813 would require USPS to establish byregulation “a separate organizational unit . . .to provide for the management of all electronicmail service of the USPS.”24 S. 898 would re-quire “a separate organizational entity” forany telecommunication services offered byUSPS.25

“Oct. 14, 1980, Order, U.S. Court of Appeals.‘“Decision of the Governors of the U.S. Postal Service, Aug.

15, 1980.21654 F 2d, 108 D.C. Court of Appeals 1981.

ZZ&n=e9sjona] Record-f% mate, Oct. 7, 1981, p. S.1 1211.ZSAdministration policy Statement, The White House, JUIY

19, 1979.Z4H.R. 2813, 97th Cong., 1st sess., Mar. 25, 1981, P. z.‘Sbgressiomd R.ecord-&ma@ Oct. 7, 1981, S.11211.

.

96 ● Implications of Electronic Mail and Message Systems for the U.S. Postal Service———

USPS has already established a separate Of-fice of E-COM Operations and implementeddetailed cost accounting procedures which, ac-cording to USPS, are more stringent than any-where else in the organization. USPS has in-itiated a complete review of E-COM costs todate, recognizing that some costs have beenhigher than initially estimated and that rate

adjustments may be necessary so that costsare fully covered over a given period of opera-tion and projected volume. However, Congressmay wish to consider stronger safeguards (e.g.,outside audit) and a greater degree of organiza-tional separation to prevent cross-subsidiza-tion and allay private sector fears.

Privacy ProtectionPrivacy protection in a USPS EMS service

is a continuing issue. First raised in theoriginal PRC consideration of the E-COM pro-posal, a 1982 National Research Council (NRC)report26 has amplified the privacy and securi-ty concerns discussed in chapter 7. To quotefrom the NRC report: “Electronic mail pre-sents potentially serious problems of securi-ty and privacy protection. The processing,storage, and transmission of large amounts ofdata, which are functions central to electronicmail, offer an attractive target for anyoneseeking access to individual and corporateinformation. ’27

OTA has not conducted a thorough reviewof E-COM security and privacy. However, pre-liminary discussions with USPS indicate thatwhile some protections are in place, additionalsecurity measures appear to be necessary. TheE-COM equipment is apparently physicallysecure, but the technical configuration makesit possible for the operator to read the hard-copy printouts before being enveloped. Oper-ators are instructed not to read the contents,and unauthorized personnel are not permittedin the E-COM facilities when printers are inoperation. Nonetheless, the potential for secu-rity breaches does exist.

A second potential problem is that the user(carrier or mailer) account numbers are printedon the outside of E-COM envelopes, thus guar-

‘Nationa.l Research Council, Assembly of Engineering, Com-mittes on Review of U.S. Postal Setice Plannin g for ElectronicMail Service Systems, Review of Electronic Mail ServiceSystems Planning for the U.S. Postal Service, NationalAcademy Press, Washington, D. C., 1981.

2TIbid., p. xi.

anteeing dissemination in a physically visiblemanner of one of the two pieces of informa-tion needed to use E-COM. The account num-ber, together with an access code and familiari-ty with the E-COM technical interconnectionstandards, would permit unauthorized use ofE-COM. A third potential problem is that allincoming messages are stored for 1 week incomputer memory or on magnetic tape in theE-COM computers. While this archiving maybe necessary in case of errors in message con-version or transmission, it also could presentanother target for security violations. Thissecurity risk is heightened by the fact thatcomputers at each of the 25 E-COM locationsare interconnected electronically to the USPSmanagement operations center in Wilkes-Barre, Pa. The purpose of the management in-formation system is to validate account num-bers and access codes and keep track ofmessage volume by account. However, it maybe technically possible to tap the archivedmessages via the management informationsystem which apparently uses dedicated, butnot otherwise secure, leased telephone lines.

Congress may wish to mandate an independ-ent review of E-COM security to ensure thatthe necessary security measures are either inplace or implemented shortly. Since it appearsthat the postal statutes do not at present ex-tend to the electronic transmission portion ofGeneration II EMS, or at least it is not clearthat the statutes apply, Congress may wishto consider the possibility of amending thePostal Act and/or Communications Act to pro-vide additional statutory protection, and con-sider the use of data encryption to provide ad-ditional technical protection.

Ch. 8—Congressional Policy Considerations ● 9 7—

Maintain Oversight and Initiate Planning onUSPS Long-Term Viability

While the immediate focus is on E-COM,providing a clear direction for USPS involve-ment in EMS and resolving current regulatoryproblems and delays, EMS issues are likely tobe with Congress for many years. Issues willbe driven by the impact of EMS on USPS, therole of USPS in EMS, and the broader impactof EMS on American society and the publicat large. For a discussion of these broader im-pact areas, see the related OTA report onComputer-Based National Information Sys-tems (1981).

As the historical (and legal) distinctions be-tween conventional and electronic mail areblurred by technological advances, Congresswill be called on to maintain oversight and in-itiate planning on the long-term viability ofUSPS for all the reasons cited in chapter 7.

At present, it is difficult for USPS to con-duct effective long-range planning with re-spect to EMS, since this requires good work-ing relationships with private telecommunica-tion and computer firms, many of whom havebeen and/or are adversaries of USPS. If someclearer consensus can be reached on the direc-tion and limits of USPS involvement in EMS,perhaps a more constructive relationship withthe private sector can develop.

USPS reports that the EMSS concept is es-sentially on hold, and that a new or modified,and more incremental rather than total sys-tems, approach to planning may be adopted.Given the dynamic nature of the telecommu-nication and computer industries, USPS canhardly be expected to develop the best conceptfor its own role without the flexibility to testand try out various alternatives, on a limitedbasis. In most successful private firms, the in-troduction of any major new product or serv-ice is preceded by a long series of research,development, and market testing of several op-tions to hopefully arrive at the one that is mostcompetitive and cost effective. At present, reg-ulatory and institutional constraints make itvery difficult for USPS to experiment. In anyrealistic sense, E-COM should be viewed as anexperiment, designed to be modified as operat-ing and market experience identifies areas forimprovement or change. In a more supportiveclimate, USPS might conduct some joint tech-nical and market tests with various privatefirms in different parts of the country. Theresults could then help guide the evolutionfrom Mailgram and E-COM to a long-termpartnership with the private sector that recon-ciles the statutory mandate of both the Postaland Communications Acts to the ultimate ben-efit of the American public.

Appendixes

Appendix A

Supplemental Detail on theMarket Penetration Model

Narrative Discussion funds transfer (EFT) and Generations II and IIIelectronic mail and message service (EMS). Table

Appendix A includes the following. Figure A-1 A-1 provides complete detail on the baseline mail-shows the structure of the model with the diver- stream.sion parameters (P, CY , to) indicated for electronic

I I

P a

tt.

SOURCE Office of Technology Assessment.

101

97-918 0 - 82 - 8


Table A-1 —Baseline Mainstream, 1977, Billions of Pieces

franked mail

The major mail segments in table A-1 wereregrouped to combine some of the smallest cate-gories, and to further divide some of the larger cat-egories. For example, household/household cor-respondence was separated into letters and greet-ing cards, since the potential for electronic han-dling of letters maybe significantly greater thanthat of greeting cards. Also, nonhousehold/non-household correspondence was separated into in-tracompany and intercompany categories, sincethe potential for electronic handling of intracom-pany mail maybe developing significantly fasterthan for intercompany mail. The miscellaneouscategories, merchandise, and segments with avolume of less than 1 billion pieces per year werecombined into one expanded miscellaneous cat-egory for each class of mail.

The mainstream segments resulting from this re-grouping are listed in table A-2, along with themail class (first, second, third, fourth, other) and1977 baseline mail volume for each segment. Thosesegments judged to be susceptible to penetrationby EFT and/or EMS are marked by an “X” intable A-2.

Table A-3 provides further detail on the EFT di-version parameters (P, CY , to) for each mainstreamsegment judged to be susceptible to EFT.

Table A-4 provides complete detail on the Gen-eration II and Generation III EMS diversionparameters. In order to simplify the analysis, the26 mainstream segments listed in table A-2 wereconsolidated into 12 segments shown in table A-4.For the purposes of table A-4, the analysis focusedon the type of mail content and sender/receiverpairs, rather than on different classes of mail. Theparameters that determine the growth and timingof the projected Generation II logistic substitutioncurve for each mailstream segment are summa-rized in the upper half of the cells in table A-4. Foreach mainstream segment (column), the values ~and t. for Generation II are listed in the row op-posite the technology that controls growth andtiming for that segment. For example, for Genera-tion II household-household (H-H) greeting cards,a = 0.2 and to = 1995 is projected, based on the

estimated availability of cost effective advancedelectronic printers. The parameters that determinethe growth and timing of the projected GenerationIII logistic substitution curve for each mainstreamsegment are shown in the lower half of the cellsin table A-4. Again, the values are shown in therow opposite the controlling technological devel-opment.

App. A—Supplemental Detail on the Market Penetration Model ● 103

Table A-2.—Mailstream Segments

Mail 1977 Penetration by:

Mainstream segment class volume a EFT EMS

H-H letters . . . . . . . . . . . . . . . . . . . . . . . . .H-H greeting cards . . . . . . . . . . . . . . . . . . .H-NH correspondence . . . . . . . . . . . . . . . .NH-H bills . . . . . . . . . . . . . . . . . . . . . . . ... ,NH-H financial statements. . . . . . . . . . . . .NH-H other nonadvertising. . . . . . . . . . . . .NH-H other nonadvertising. . . . . . . . . . . . .NH-H other nonadvertising. . . . . . . . . . . . .NH-H correspondence . . . . . . . . . . . . . . . .NH-H advertising . . . . . . . . . . . . . . . . . . . . .NH-H advertising . . . . . . . . . . . . . . . . . . . . .NH-N H advertising . . . . . . . . . . . . . . . . . . .NH-N H advertising . . . . . . . . . . . . . . . . . . .NH-NH intracompany correspondence. . .NH-N H intercompany correspondence. . .NH-NH other nonadvertising . . . . . . . . . . .NH-NH bills and statements . . . . . . . . . . .H-NH negotiable instruments . . . . . . . . . .NH-H negotiable instruments . . . . . . . . . .NH-NH negotiable instruments . . . . . . . . .NH-N H legal/financial instruments . . . . . .Miscellaneous . . . . . . . . . . . . . . . . . . . . . . .Miscellaneous . . . . . . . . . . . . . . . . . . . . . . .Miscellaneous . . . . . . . . . . . . . . . . . . . . . . .Miscellaneous . . . . . . . . . . . . . . . . . . . . . . .Miscellaneous . . . . . . . . . . . . . . . . . . . . . . .

11111123113131131111112340

3.43.2

2.66.08.39.51.51.17.22.33.03.22.02.38.46.51.81.51.92.11.02.50.70.9

— x— x— xx xx x— x— x— x— x— x— x— x— x— x— x— xx xxxx

abllllons of pieces of mailH = HouseholdNH = Nonhousehold1 = First.class2 = Second-class3 = Third-class4 = Fourth. classo = Other class

SOURCE. Off Ice of Technology Assessment

Table A-3.—EFT Mail Diversion Parameters

Penetration Growth1977 volume potential constant Time when 5 percent

Mainstream segment (billions) (P) (a) diversion occurs t.

NH-H bills . . . . . . . . . . . . . . . . . . . . . . . . . 8.9 0.9 0.20 1985NH-H financial statements. . . . . . . . . . . 2.6 0.9 0.20 1985NH-NH bills, statements, etc. . . . . . . . . 8.4 0.9 0.20 1985H-N H negotiable instruments . . . . . . . . 6.5 1.0 0.20 1985NH-H negotiable instruments . . . . . . . . 1.9 1.0 0.20 1985NH-NH negotiable instruments . . . . . . . 1.5 1.0 0.20 1985

SOURCE: Office of Technology Assessment; see OTA, selected Hecfrorrlc Funds Transfer Issues: Secur/ty, Privacy, and Equity, OTA-BP-CIT-12, March 1982, for further discussion of EFT trends and developments which are generally consistent with the EFT diversionparameters.


Table A-4.—EMS Technology Assumptions and Diversion Parameters by Mail Content and Sender/Receiver Pair———.

From-to N-N Intra N-N Inter N-N N-N N-N N-H H-H H-H H-N “ N H N-H N-H

Other non- Other non- BiIIs andContent Corr Corr advertising Bills Advertising Advertising Cards Corr Corr Corr advertising statements

PenetrationGeneration II 100% 100 ”/0 70°/0 100% 100 ”/0 100 ”/0 100 ”/0 100 ”/0 100% 100 ”/0 70 ”/0 100 ”/0

Generation Ill 100 ”/0 100 ”/0 70 ”/0 100% 100 ”/0 30% 300/o 100 ”/0 100% 100 ”/0 7 0 % 100 ”/0—

Early electronic II 30% 1983 30 ”/0 1983 300/o 1983 30% 1983 30% 1983 30 ”/0 1983printers Ill

Advanced II 20 ”/0 1995 20 ”/0 1995 20 ”/0 1995

EDP and office IIautomation Ill 1983 20 ”/0 1984 20 ”/0 1984 20 ”/0 1985 20 ”/0 1987

Home computer II 30% 1987 30% 1987

terminals--- —

Ill 20 ”/0 1987 41)0/0 1987 40% 1987 @ O / o 1987

Viewdata/ IIteletext Ill 4 0 % 1985 20 ”/0 1985

.

Inexpensive HC IIreceiver

—Ill 20 ”/0 1990.

N = NonhouseholdH = Household a tm

= Initial rate of growth.,

(r

t o= Year of 5% diversion Key to Entries: Generation II EMS

SOURCE: Office of Technology Assessment. Generation Ill EMS

Table A-5 shows the actual procedure used bythe computer program to obtain the overall diver-sion results. The computer program applied anunderlying growth rate to each mainstream seg-ment, and then calculated the portion of each seg-ment diverted to EFT, Generation II EMS, andGeneration III EMS. These diversion estimateswere calculated using the logistic growth curve(described in app. B) for each mainstream segment,with the parameters P (penetration potential), ~(growth constant) and t0 (time of 5-percent penetra-tion), as specified earlier in tables A-3, A-4, andA-5. The diversion estimates for each mailstreamsegment were then added together to give overallestimates for residual conventional mail volumeand for the volumes of mail diverted to EFT, Gen-eration 11 EMS, and Generation 111 EMS.

Diversion estimates were calculated for theyears 1985, 1990, 1995, and 2000. As explained inchapter 3, the results of the computer runs wereadjusted upward by 10 percent (multiplied by afactor of 1.10) to compensate for the difference be-tween the projected and actual growth rate in themainstream for the years 1977-81.

Table A-5.—Procedure Used by Computer Programfor Market Penetration Projections

2.

3.

4.

5.

6.

calculated.Compute a “UG” factor for the underlying growth in themainstream relative to 1977. For most runs the assump-tion was 2 percent compounded growth. Hence the “UG”factor is 1.02 raised to the power (t-1977).Compute EFT penetration for each of the segments intable A-2 which are penetrated by EFT. First computethe penetration fraction f using the logistic substitutionformula in appendix B, and the values of ~ and t. fromtable A-3. Then multiply the 1977 volume x the “UG”factor x the penetration potential P (from table A-3) x f.This yields the volume diverted to EFT in year t for eachmainstream segment.Reduce the 1977 volumes for segments affected by EFTby the amount of EFT diversion before computing EMSdiversion.Compute diversion to Generation Ill EMS for each mail-stream segment affected by EMS just as in step 3 abovefor EFT, except use the reduced 1977 volumes for EFTimpacted segments, and use w tO, and P for GenerationIll from table A-4. The penetration potential P is con-tained in the third row of table A-4 (marked “PENETRA-TION”). The a and t. values for Generation Ill are in thelower half of the cells in table A-4.Compute diversion to Generation II EMS for each seg-ment in the same manner as in step 5 above, using a andt. from the upper half of each cell in table A-4. Thenreduce the computed Generation II volume by the com-puted Generation Ill volume. If the Generation III volumeexceeds the Generation II volume, then Generation IIvolume is zero.Add results across mainstream segments for each classof mail to get diversion totals by class of mail.

7.

—SOURCE Office of Technology Assessment

Appendix B

Logistic Substitution Process

The logistic substitution process is one math-ematical formulation that can be used to describethe encroachment of one technology on the marketof another. In the field of systems ecology a mul-titude of models have been used for this purpose.However, the more complex models require datathat are significantly more detailed than wereavailable for this analysis, and for all their com-plexity have no more inherent validity than a sim-ple model such as logistic substitution. The mostimportant considerations for any model used forthis kind of analysis are that it be based on sup-portable data and that it make sense.

The logistic substitution curve (fig. B-1) presentsa reasonable macroscopic model of the encroach-ment of a new technology into the market of anestablished, mature technology. When the newtechnology becomes available, it penetrates themarket slowly at first due to relatively high costand limited consumer acceptance. As time goes on,cost declines as volume of use builds, leading toan accelerating growth. Growth remains gradual,however, since consumer acceptance can be gained

Figure B-1.— Logistic Substitution Growth Process

NOTES P = Maximum penetration potentialf = Market share (fraction of total market) for new technology at

time t= Growth constant

Time when 5% penetration occurst“ = Time when 50°/0 penetration occurs

SOURCE, Off Ice of Technology Assessment and Fred B, Wood, et al , USPSand the Cornrrrun/caf/ons l?evo/uf/on /rnpacf, Opt/ens arrd /ssues,George Washington Unwerslty, Mar. 5, 1977.

only with time. As the available market ap-proaches saturation, the rate of growth declines.

In mathematical terms, the key variable is themarket share of the new technology, expressed asa fraction f of the total potential market. Ultimate-ly, the entire potential market will be penetrated,and f will equal 1 or the maximum penetration po-tential P, whichever is less. Initially, the marketshare fraction f is very small, but grows with timeat a rate proportional to the market share itself.Thus, in the early stages of growth, the growth off per unit of time is expressed as ~ X f, where CYis a constant for the particular technology andmarket being considered. The factor Q will bereferred to as the “growth constant” for the par-ticular substitution process. It is a measure of howquickly the technology will penetrate the market.For example, with CY = 0.4 (per year), the marketshare will rise from 5 to 75 percent in 10 years,whereas with CY = 0.2 such a change in marketshare will require 20 years.

The market share at any time can be computedfrom the equation in figure B-1 if the growth con-stant Q and the time of 50-percent market pene-tration t are specified. Alternatively, some otherpoint on the curve can be specified along with thegrowth constant ~. OTA chose to specify a par-ticular logistic substitution curve by specifyingthe growth constant, a , and the time (calendaryear) at which 5-percent penetration of the avail-able market occurs. This 5-percent penetrationtime is designated to. The market share f can thenbe expressed in terms of a and tO as follows:

The logistic substitution process has been usedonce before to model the penetration of mail byEMS technology.l However, in that 1977 studythe mainstream was considered a single market and

‘Fred B. Wood, et al., USPS and the Communications Revolution:Impacts, Options and Issues, George Washington University, Mar. 5,1977.

105


the substitution of EMS a single process. For thepurposes of this study, the mainstream was con-sidered to be many different submarkets withvarying susceptibility to several different EMStechnologies.

The competition between Generation II andGeneration III EMS technologies can be describedas follows. As EMS technology develops, Genera-tion III systems eventually will most likely be lessexpensive to use than Generation 11 systems be-cause Generation III employs electronic, ratherthan postal carrier, delivery. Also, Generation IIIEMS technology will most likely advance to thepoint where it will be able to accommodate any for-mat or display capability that can be handled byGeneration II. In other words, any mainstream seg-ment that can be diverted to Generation II EMScan eventually be diverted to a Generation IIIsystem as well. Furthermore, the switch to Gen-eration III EMS will most likely occur when theeconomics (to the user) favor Generation 111 overGeneration II EMS, and at that point GenerationIII systems will start to take away market sharefrom Generation II systems.

Thus, in the market penetration model, penetra-tions of conventional mail by Generation II EMSand Generation III EMS were calculated separate-ly using parameters derived from an assessmentof relevant technologies.* The volume resulting forGeneration III was then subtracted from the re-sulting Generation 11 volume, producing a netGeneration II volume figure. The net GenerationII and Generation 111 volumes were subtractedfrom the conventional mail volume to obtain theresidual conventional volume.

Figure B-2 illustrates the relationships involvedin a market segment where a portion of the totalmessage volume is judged suitable for penetrationby EMS. First, the total market in this segmentis shown as one that grows at a constant percent-age rate. The potential market for EMS—the por-tion of the market which is physically suitable for

● See table A-4.

Time

NOTES: P = Maximum penetration potential

f2 = Fraction of total market for EMS at time t

13 = Fraction of total market for Generation Ill at time tz - f3 = Fraction of total market for Generation II at time t

SOURCE: Office of Technology Assessment,

eventual transmission by EMS systems—is rep-resented as a constant fraction P of the totalmarket. Initially, Generation II EMS begins topenetrate the potential market, capturing a mar-ket share f2 of the total market. Later, GenerationIII begins to penetrate. At this point, the fractionf2 represents the total EMS market share, andGeneration III growth comes at the expense ofGeneration II. Hence f3, the result of a separatesubstitution process, represents the market sharefor Generation III, while the market share forGeneration II becomes f2 – f3. For the assump-tions used in this study, total substitution of Gen-eration 111 for Generation II does not occur in anymainstream market segment within the 20 yeartimeframe of the market penetration projections.

Appendix C

Equation Used to CalculateLabor Requirements

labor requirement in year x expressed as a fraction of labor required in 1980.

mail volume projected for a specified future year x.

mail volume in 1980, the base year.

labor productivity index for future year x, where LP1980 = 1.0.

variable cost component of total labor cost or of total cost for a specific employeegroup.


107

—.

Appendix D

E-COM InterconnectionArrangements

Dial-Up Access Arrangements

1. Binary synchronous, compatible with theIBM 2780 and 3780 terminals using EBCDICcharacter code sets, operating in half-duplex modeat 2400 bits per second with Bell System 201Ccompatible modems and at 4800 bits per secondwith Bell 2088 compatible modems.

2. Asynchronous, using the Texas Instruments700 series convention for data block transmissionand the 96 character ASCII subset as defined byANSI Standard X3.4—1977, currently used in awide variety of message and data terminals, oper-ating at 300 and 1200 bits per second full-duplexwith Bell System 212A compatible modems.

Dedicated AccessArrangements

1. Packet switched X.25, with LAP data linkprotocol, binary synchronous framing using the

ASCII character set, and full-duplex operations at2400, 4800, or 9600 bits per second.

2. DEC (Digital Equipment Corporation)DDCMP serial synchronous byte oriented line pro-tocol, using the ASCII character set, with full-duplex operation at 2400, 4800, 9600, or 56,000bits per second.

3. Binary synchronous, compatible with theIBM 2780 and 3780 terminals using EBCDICcharacter code sets, operating in half-duplex modeat 2400 bits per second with Bell System 201Ccompatible modems and at 4800 bits per secondwith Bell System 208B compatible modems, andat 9600 bits per second with Bell System 209Acompatible modems.

4. Asynchronous, using the Texas Instruments700 series convention for data block transmissionand the 96 character ASCII subset as defined byANSI Standard X3.4—1977, currently used in awide variety of message and data terminals, oper-ating at 300 and 1200 bits per second full-duplexwith Bell System 212A compatible modems.

‘USPS; see Federal Register, vol. 46, No. 199, Thursday, Oct. 15, 1981,p. 50879.

108

Index

American Telephone and Telegraph (AT&T), 78automation, 27, 30

Communications Act (see Communications Act of1934)

Communications Act of 1934, 3, 11, 16, 17, 76, 77,80, 87, 88, 89, 94, 96, 97

competition, 17, 18, 91in data processing and office automation

technologies, 30in EMS market, 16impact of private delivery services on USPS, 73restrictions on USPS, 77between USPS and private industry, 76, 87-88

computer modeling, 3, 18equation for labor requirements, 107limitations of, 19logistic substitution process, 105-106market penetration model, 19, 23-24, 101-104revenue and cost model, 19, 51-52

computers, 4, 16, 27, 30, 31, 32, 42, 44, 79, 87, 88Congress, 3, 4, 75

concern over USPS role in electronic mail andmessage systems, 17-18

House Committee on Energy and Commerce, 17House Committee on Post Office and Civil

Service, 17House Government Operations Subcommittee on

Government Information and IndividualRights, 77

possibilities for action by, 9-11, 87-97proposed legislation, 17, 18, 77, 95Senate Committee on Commerce, Science, and

Transportation, 17Senate Committee on Governmental Affairs, 17Subcommittee on Government Information and

Individual Rights, 77Consultative Committee for International Telephone

and Telegraph, 30cross-subsidization, 18, 75, 88, 95

Datapost, 27data processing, 17, 18, 27, 30, 87Department of Commerce, 17, 75, 88, 95Department of Justice, 9, 17, 62, 74, 75, 88, 95deregulation

of the telecommunication industry, 87Dialcom, Inc., 78Domestic Mail Classification Schedule, 92

E-COM (see electronic computer-originated mail)EFT (see electronic funds transfer)electronic computer-originated mail (E-COM), 3, 4, 9,

10, 11, 16, 27, 53, 74, 76, 77, 79, 81, 83, 88, 89,90, 93, 95, 97

capital cost of, 75establishment of a separate USPS office for, 96initiation of service, 17

interconnections with telecommunicationproviders, 78, 92, 108

legal challenge to USPS role in, 17projected growth of, 93regulation of, 94-95technical standards of, 78volume of, 75

electronic funds transfer (EFT), 4, 5, 6, 8, 16diversion of first-class mail to, 4, 5, 6, 26-27Generation III systems, 4, 5, 11, 81

electronic mail and message systems (EMS)alternatives for involvement of USPS in, 9, 10competition between Generations 11 and III

systems, 77-78congressional interest in, 17-18cost displacement by, 9diversion of first-class mail to, 6effects on USPS finances, 90effects on USPS labor requirements, 7-8, 64-70, 90effects on USPS postal rates, 5-6, 61-63effects on USPS service levels, 5-8, 63-64fairness of USPS role in, 73-75Generation I systems, 77Generation II growth and timing estimates, 29-30,

32-33, 45-48Generation II systems, 4, 5, 6, 7, 8, 9, 10, 27, 29,

30, 31, 32, 33, 37, 38, 39, 40, 41, 42, 44, 45, 46,52, 53, 56, 61, 62, 64, 66, 68, 69, 77, 81, 88,89, 90, 94

Generation 111 systems, 4, 5, 8, 9, 10, 11, 27, 30,31, 32, 37, 38, 42, 44, 45, 46, 77, 78, 81, 83,88, 91, 94

growth of Generation III services, 30-31legality of USPS role in, 76marketing of, 92market penetration analysis of, 37-48penetration potentials of, 27performance standards for, 93policy implications of, 8-11projected growth of, 4-5relationship between Generation 11 and III

estimates, 31-32revenue and cost assumptions, 56revenue and cost model analysis of, 51-57stimulation of demand for new messages by, 42

electronic message market, 16electronic message service system (EMSS), 9, 16, 32,

53, 90, 93, 97EMS (see electronic mail and message systems)EMSS (see electronic message service system)Equitable Life Assurance Co., 83

FCC (see Federal Communications Commission)Federal Communications Commission (FCC), 10, 16,

77, 78, 87, 88, 93, 94, 95Computer II decision of, 76

Federal Express, 73Federal Financing Bank, 74

111


Federal Property and Administrative Services Actof 1980, 87

Government Printing Office, 83Graphnet, 78greeting cards, 27, 32GTE Telenet, 78

INTELPOST, 16international electronic post (see INTELPOST)International Standard Organization, 30ITT World Communications, 78, 83

judicial delay, 10

mail (also see mailstream)advertising, 27, 31content categories, 24covers, 80financial statements, 31, 32first-class, 6, 56greeting cards, 31growth of volume, 33-34sender/receiver pairs, 24volume of, 4-5

Mailgram service, 3, 4, 9, 16, 27, 52, 89, 90, 93, 97mainstream

baseline, 24-25, 101-104EFT penetration of, 4, 26-27EMS penetration of, 4-5, 27-33growth of, 4, 6, 33-34percent as first-class mail, 6vulnerability to penetration by electronic mail

systems, 24-25Merrill Lynch Pierce Fenner & Smith, 83

National Research Council (NRC), 96National Technical Information Service, 83Netword, Inc., 83

Office of Management and Budget, 76, 77Office of Technology Assessment (OTA), 3, 4, 5, 7,

8, 9, 11, 26, 29, 30, 31, 33, 42, 44, 45, 46, 51, 52,53, 56, 57, 62, 63, 64, 65, 66, 68, 69, 75, 81, 88,97

Omnibus Budget and Reconciliation Act of 1981,52, 75

Opinion Research Corporation, 93

Paperwork Reduction Act of 1980, 87Personal Computer Network, 27Postal Act (see Postal Reorganization Act of 1970)Postal Rate Commission (PRC), 9, 10, 17, 45, 52, 62,

76, 78, 87, 88, 90, 91, 92, 93, 94, 95hearings of, 75

postal ratesimplications of Generation II systems growth for,

61-63

Postal Reorganization Act of 1970, 3, 10, 11, 17, 45,62, 74, 75, 80, 87, 88, 89, 93, 96, 97

as authority for USPS involvement in EMS, 76Postal Service (see United States Postal Service)Postmaster General, 16privacy protection, 10, 88, 93, 96

data encryption, 81of electronic communication, 80-82and EMS, 79-81of sealed mail, 79

Private Express Statutes (PES), 10, 42, 73, 87,89, 93

Purolator Corp., 73

RCA, 5, 9, 10, 32, 46, 53, 69, 79, 90, 91, 93regulation, 11, 93-96

Satellite Business Systems, 78security (see privacy protection)Shell Oil Co., 83Supreme Court, 79Survey Research Center, 18, 24

telecommunication, 17, 18, 77, 80, 87, 89, 91, 92, 95telecommunication and computer industries, 3

effects of a USPS role in EMS on, 73-80Telecommunications Deregulation and Competition

Act of 1981, 18teletext (see viewdata/teletext)TRT Communications Corp., 83TRT Telecommunications Corp., 78Tyme-Gram service, 27Tymnet, 78

United Parcel Service, 73United States Postal Service (USPS)

alternatives for involvement of, in EMS, 9, 10annual appropriations for, 74constraints on, 97contribution of first-class mail to fixed costs of,

6, 56cost and ratesetting process, 75disputed role of, in EMS and E-COM, 3, 78-79effects of reducing mail deliveries by, 5-8, 82innovation stimulated by, 79introduction of E-COM by, 3jurisdictional conflicts with other Federal

agencies, 88labor-intensiveness of, 64labor requirements of, 7-8, 64-70legal mandates of, 3, 60, 76, 79, 82, 87long-term viability of, 11, 81-83performance of, 15-16public service subsidy, 75rates and service levels of, 5-6role in EMS, 9, 10, 17service levels of, 63-64tax-exempt status of, 74

Index 113

United States Treasury viewdata/teletext, 4, 16, 27, 31loans to USPS, 74

University of Michigan, 18, 24, 33 Western Union, 3, 16, 76, 78, 89, 90, 91, 94

0