__ ____________ ,/X6. /9fls WORLD BANK TECHNICAL PAPER NUMBER 317 Information Technology and National Trade Facilitation Guide to Best Practice Robert Schware and Paul Kimberley ,, I I,- , J A5,IIIII/ _a~~~- --f - '_ RECENT WORLD BANK TECHNICAL PAPERS No. 237 Webster, The Emergence of Private Sector Manufacturing in Poland: A Survey of Firms No. 238 Heath, Land Rights in C6te d'lvoire: Survey and Prospectsfor Project Intervention No. 239 Kirmani and Rangeley, International Inland Waters: Conceptsfor a More Active World Bank Role No. 240 Ahmed, Renewable Energy Technologies: A Review of the Status and Costs of Selected Technologies No. 241 Webster, Newly Privatized Russian Enterprises No. 242 Barnes, Openshaw, Smith, and van der Plas, What Makes People Cook with Improved Biomass Stoves? A Comparative International Review of Stove Programs No. 243 Menke and Fazzari, Improving Electric Power Utility Efficiency: Issues and Recommendations No. 244 Liebenthal, Mathur, and Wade, Solar Energy: Lessonsfrom the Pacific Island Experience No. 245 Klein, External Debt Management: An Introduction No. 246 Plusquellec, Burt, and Wolter, Modern Water Control in Irrigation: Concepts, Issues, and Applications No. 247 Ameur, Agricultural Extension: A Step beyond the Next Step No. 248 Malhotra, Koenig, and Sinsukprasert, A Survey of Asia's Energy Prices No. 249 Le Moigne, Easter, Ochs, and Giltner, Water Policy and Water Markets: Selected Papers and Proceedingsfrom the World Bank's Annual Irrigation and Drainage Seminar, Annapolis, Maryland, December 8-10, 1992 No. 250 Rangeley, Thiam, Andersen, and Lyle, International River Basin Organizations in Sub-Saharan Africa No. 251 Sharma, Rietbergen, Heimo, and Patel, A Strategyfor the Forest Sector in Sub-Saharan Africa No. 252 The World Bank/FAO/UNIDO/Industry Fertilizer Working Group, World and Regional Supply and Demand Balances for Nitrogen, Phosphate, and Potash, 1992/93-1998/99 No. 253 Jensen and Malter, Protected Agriculture: A Global Review No. 254 Frischtak, Governance Capacity and Economic Reform in Developing Countries No. 255 Mohan, editor, Bibliography of Publications: Technical Department, Africa Region, luly 1987 to April 1994 No. 256 Campbell, Design and Operation of Smallholder Irrigation in South Asia No. 258 De Geyndt, Managing the Quality of Healthi Care in Developing Countries No. 259 Chaudry, Reid, and Malik, editors, Civil Service Reform in Latin America and the Caribbean: Proceedings of a Conference No. 260 Humphrey, Payment Systems: Principles, Practice, and Improvements No. 261 Lynch, Provision for Children with Special Educational Needs in the Asia Region No. 262 Lee and Bobadilla, Health Statisticsfor the Americas No. 263 Le Moigne, Subramanian, Xie, and Giltner, editors, A Guide to the Formulation of Water Resources Strategy No. 264 Miller and Jones, Organic and Compost-Based Growing Media for Tree Seedling Nurseries No. 265 Viswanath, Building Partnershipsfor Poverty Reduction: The Participatory Project Planning Approach of the Women's Enterprise Management Training Outreach Program (WEMTOP) No. 266 Hill and Bender, Developing the Regulatory Envi ronmentfor Competitive Agricultural Markets No. 267 Valdes and Schaeffer, Surveillance of Agricultural Prices and Trade: A Handbookfor the Dominican Republic No. 268 Valdes and Schaeffer, Surveillance of Agricultural Prices and Trade: A Handbookfor Colombia No. 269 Scheierling, Overcoming Agricultural Pollution of Water: The Challenge of Integrating Agricultural and Environmental Policies in the European Union No. 270 Banerjee, Rehabilitation of Degraded Forests in Asia No. 271 Ahmed, Technological Development and Pollution Abatement: A Study of How Enterprises Are Finding Alternatives to Chlorofluorocarbons No. 272 Greaney and Kellaghan, Equity Issues in Public Examinations in Developing Countries No. 273 Grimshaw and Helfer, editors, Vetiver Grassfor Soil and Water Conservation, Land Rehabilitation, and Embankment Stabilization: A Collection of Papers and Newsletters Compiled by the Vetiver Network No. 274 Govindaraj, Murray, and Chellaraj, Health Expenditures in Latin America (List continues on the inside back cover) WORLD BANK TECHNICAL PAPER NUMBER 317 Information Technology and National Trade Facilitation Guide to Best Practice Robert Schware Paul Kimberley The World Bank Washington, D.C. Copyright C 1995 The International Bank for Reconstruction and Development/THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing November 1995 Technical Papers are published to communicate the results of the Bank's work to the development com- munity with the least possible delay. The typescript of this paper therefore has not been prepared in accor- dance with the procedures appropriate to formal printed texts, and the World Bank accepts no responsibili- ty for errors. Some sources cited in this paper may be informal documents that are not readily available. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s) and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board of Executive Directors or the countries they represent. The World Bank does not guarantee the accuracy of the data included in this publication and accepts no responsibility whatso- ever for any consequence of their use. The boundaries, colors, denominations, and other information shown on any map in this volume do not imply on the part of the World Bank Group any judgment on the legal status of any territory or the endorsement or acceptance of such boundaries. The material in this publication is copyrighted. Requests for permission to reproduce portions of it should be sent to the Office of the Publisher at the address shown in the copyright notice above. The World Bank encourages dissemination of its work and will normally give permission promptly and, when the reproduction is for noncommercial purposes, without asking a fee. Permission to copy por- tions for classroom use is granted through the Copyright Clearance Center, Inc., Suite 910, 222 Rosewood Drive, Danvers, Massachusetts 01923, U.S.A. The complete backlist of publications from the World Bank is shown in the annual Index of Publications, which contains an alphabetical title list (with full ordering information) and indexes of sub- jects, authors, and countries and regions. The latest edition is available free of charge from the Distribution Unit, Office of the Publisher, The World Bank, 1818 H Street, N.W., Washington, D.C. 20433, U.S.A., or from Publications, The World Bank, 66, avenue d'Iena, 75116 Paris, France. ISSN: 0253-7494 ISBN: 0-8213-3534-0 Robert Schware is Senior Informatics Specialist for The World Bank, Finance and Private Sector Development Vice Presidency, Industry and Energy Department, Telecommunications and Informatics Division. Paul Kimberley is an electronic commerce consultant to the division and principal of Paul Kimberley and Associates (PKA). Comments on this document may be directed to: Robert Schware, Senior Informatics Specialist, Telecommunications and Informatics Division, Industry and Energy Department, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, Tel: (+1 202) 458-0794, Fax: (+1 202) 477-3379, Internet: rschware@worldbank.org Contents Foreword v Costs of Installing EDI 22 Abstract vii The Advocacy and Implementation Model 23 Acknowledgments ix Cost Summary for a 20-Partner Grouping 25 Abbreviations and Acronyms xi Advocay and Implementation: the Last Word 26 A Partner's Internal EDI Costs 26 1 LEGAL AND REGULATORY REQUIREMENTS 1 The Dilemma 26 Alternatives to EDI 27 The Legality of EDI 1 Intermediaries 27 Agreements 1 Low Tech Infrastructure 28 International Law 2 Broader Technology Initiatives 28 Contracts 2 The Hypothesis 28 Evidence 3 The Model 29 Data Disclosure 3 Not So Low Tech EDI 32 Legislation 4 Voice Processing: Telephone Technology 33 Conclusion 4 Scanning and Image Technologies 33 Other Technologies 35 2 SKILLS AND TECHNOLOGY Conclusion 36 INFRASTRUCTURE 5 ANNEX 1: THE CASE STUDIES 37 Technology and Skills Inventory 5 Key Technology and Support Elements 5 Methodology 37 Best Practice 6 Argentina 37 Barriers to Use and Participation 7 Current Status 39 Discussion 40 3 LOCAL BUSINESS CONSIDERATIONS 9 Australia 40 Trade Facilitation 41 Government and Local Business Practices 9 Tradegate 41 Culture and Religion 10 The Organization 41 External Influences 10 The User Community 42 Role of Government 42 4 INVESTMENT COSTS AND BENEFITS 11 VAN Interconnect 43 Vendors 43 Costs and Benefits 11 Industry Associations 44 Cost and Revenue Categories 13 Assessing Tradegate 44 Three Typical Models 14 Lessons from the Australian Experience 44 From Different Perspectives 16 Brazil 45 Summary 18 Current Status 46 Conclusion 19 Discussion 47 Chile 47 5 TECHNOLOGY AND COST OPTIONS 20 Current Status 48 EDI's Brick Wall 20 Discussion 49 Number of EDI users and Hong Kong 49 Message Volumes in Australia 20 Tradelink 50 fii Summary 51 EDI Gateway and Service Functionality 75 Hungary 52 Vendors 75 Trade Facilitation 53 Other Media 76 Current Status 53 A Model for a Gateway-Rollout Service 77 Discussion 54 Summary 77 Malaysia 55 Current Status 55 ANNEX 4: SAMPLE TERMS OF REFERENCE FOR Experience 56 ELECTRONIC TRADE FACILITATION SYSTEM, Discussion 56 IMPLEMENTATION, SUPERVISION, AND Mexico 57 MID-TERM REVIEW 78 Current Status 58 Discussion 58 Background 78 New Zealand 59 Assignment Objectives 79 Singapore 59 Audit and Quality Assistance 79 Current Status 60 Awareness and Training 79 The SNS Business Case 61 Marketing and Business Plan 79 SNS Running Costs 61 Scope of Work 79 Business and Cultural Factors 62 Reporting Relationships 80 Taiwan (China) 63 Deliverables 80 Current Status 63 Mid-term Review 80 Summary 64 Desired Qualifications of Consultant 81 ANNEX 2: IMPLEMENTATION CONSIDERATIONS ANNEX 5: UN-EDIFACT 82 AND TIME FRAMES 65 Whose Standard: Yours or Mine? 83 Technology Project Plan 65 Message Standards 84 Trade Facilitation: EDI Project Plan 65 The Components of EDIFACT 86 Information Gathering 65 The UN-EDIFACT Reference Model 86 Reverse Engineering 67 UN-EDIFACT Syntax 88 Reengineering 67 Data Elements 88 Implementation 68 Data Element Values and Code Lists 90 Technology Issues 70 Composite Data Elements 90 Time Frames 70 Segments 91 Qualifiers 92 ANNEX 3: A SAMPLE TECHNICAL PROPOSAL 72 Messages 93 The Organization 94 Definitions 72 Developing a Message 96 Requirements 73 UN-EDIFACT Deliverables 98 Functions 74 Gateway Services 74 ANNEX 6: GLOSSARY AND ABBREVIATIONS 99 iV Information Technology and National Trade Facilitation: Gutide to Best Practice Foreword Information technology is demolishing territorial boundaries today, and bringing nations together in a single global community-but a community more fiercely competitive than ever before. Change is the order of the day. Trade, banking, and telecommunica- tions are being deregulated. Transport is getting faster, flexible, and available. Reengineered business systems are taking advantage of quick- response and just-in-time strategies; and cargoes, containers, and goods are being tracked around the globe by a variety of automatic identification devices. Electronic data interchange and electronic commerce are replac- ing the slower, more tedious paper trail. Countries now compete in global markets regardless of time zones, national boundaries, and distance, as products and processes are redesigned to adjust to the new business environment. The increasing pressures from the global market are forcing everyone to adopt these new trade practices and standards. Customs, treasuries, and lawmakers are having to reinvent themselves to adapt to the concept of electronic commerce. Nations are adjusting to new methods of finance and tax gathering, opening up their telecommunications systems to private interests, and learning to take full advantage of harmonized procedures, standards, and practices for trade documentation. None of this is easy, but for many countries of the world, it is a matter of survival. The present report, Information Technology and National Trade Facilitation: Making the Most of Global Trade, attempts to make the process of change smoother. It examines costs, benefits, and best practices in applying information technology to trade facilitation. It provides definitions and introduces basic concepts and issues in the substitution of electronics for paper, in the effort to achieve cost-effective international trade. The companion volume, Information Technology and National Trade Facilitation: Guide to Best Practice, is a practical aid for governments to understand the tasks, costs, and time involved in setting up and imple- menting national trade facilitation initiatives. Together the reports offer essential information ,or decision makers promoting better trade practices in concert with international standards, common practice, and most important, specific national goals. JEAN-FRANCOIS RISCHARD VICE PRESIDENT FINANCE AND PRIVATE SECTOR DEVELOPMENT THE WORLD BANK NOVEMBER 10, 1995 V ABSTRACT Designed for use by managers and technical staff involved in the imple- mentation life cycle of an electronic commerce project, this volume begins with a section on legal and regulatory requirements, covering the legality of EDI and electronic commerce and the various agreements which deal with international trade. A second major section considers the skills and technology infrastructure necessary to participate in IT assisted best practice trade facilitation projects. Local business issues are considered next, including government and local business practices, culture and external influences. A discussion of investment costs and benefits at national, industry, and end user levels is followed by a section on technological and service alternatives for smaller, less technologically advanced organizations. Four Annexes contain a series of national case studies; implementation considerations and time frames in order to illustrate a typical project plan; a typical technical proposal covering vendor and end user require- ments; and sample terms of reference for a project review, based on an actual World Bank project-in-progress. A fifth Annex discusses in some detail the need for and the design and application of EDI standards as used in trade facilitation applications. The volume concludes with a glossary of terms and abbreviations used within both reports and in the implementation of best practice, IT-based facilitation applications. vii ACKNOWLEDGMENTS Many organizations have been generous in donating their time and sharing their experiences with the authors of this report. In an introductory volume of the report, Information Technology and National Trade Facilitation: Making the Most of Global Trade, there is a list of all cooperating agencies, organizations, and individuals who contrib- uted information and experiences. They include many international agencies, international industry bodies, government departments, technical and trade associa- tions, vendors, and a wide variety of systems users from banks, corporations and governments. In addition, we received the help of many hardware and software vendois, network services vendors, tele- communications companies and authori- ties, and private individuals. Valuable contributions have been made by profes- sional staff from within the Bank, particu- larly Hans Peters and Francoise Clottes. We would like to thank Shampa Banerjee for a fine job of editing. It is impossible to research global experiences without the help of the pio- neers. We gratefully acknowledge this help. ix ABBREVIATIONS AND ACRONYMS ACS Automated Commercial System APEC Asia Pacific Economic Cooperation ASYCUDA Automated System for Customs Data BT British Telecom CAS Community Access Service, Hong Kong CIM Computer-Integrated Manufacturing CNAB National Council for Banking Automation, Brazil EAN European Article Numbering EANCOM European Article Numbering Communication ECE Economic Commission for Europe EDI Electronic Data Interchange EDIFACT EDI for Administration Commerce and Transport EFT Electronic Funds Transfer EFTA European Free Trade Area EFTPOS Electronic Funds Transfer at Point of Sale ERS Evaluated Receipt Settlement EU European Union FACET Future Automated Commercial Environment Team FDP Finance and Private Sector Development GATT General Agreement on Tariffs and Trade GDP Gross Domestic Product IBM International Business Machines ICC International Chamber of Commerce ID Identity III Institute of Information Industry, Taiwan (China) ISO International Standards Organization IT Information Technology JIT Just in Time (inventory control) LOCODE United Nations Location Code MBK Hungarian Bank for Foreign Trade MOF Ministry of Finance, Taiwan (China) MSTQ Metrology, Standards, Testing, and Quality MTCW Ministry of Transport, Communication, and Water Management, Hungary NAFTA North American Free Trade Agreement NCB National Computer Board, Singapore OECD Organization for Economic Cooperation and Development OFTP Open File Transfer Protocol PAXLST (UN-EDIFACT) Passenger List PSA Port of Singapore Authority QR Quick Response RFID Radio Frequency Identity SITPRO The Simpler Trade Procedures Board xi SME Small and Medium Enterprise SNS Singapore Network Services SPEDI Shared Project for EDI, Hong Kong TDB Trade Development Board, Singapore TP (UNCTAD) Trade Point UN United Nations UNCTAD United Nations Conference on International Trade and Development UN-EDIFACT United Nations EDI for Administration Commerce and Transport U.S. United States U.K. United Kingdom VAB Value Added Banking VAN Value Added Network VANS Value Added Network Service WCO World Customs Organization WTC World Trade Center WTO World Trade Organization WWW World Wide Web Xii Information Technology and National Trade Facilitation: Guide to Best Practice 1 Legal and Regulatory Requirements Over the last 20 years, an intensive debate * Should the law be amended for its has surrounded the legal issues in the application to EDI? substitution of electronics for paper. There are many good books and a plethora of There is a third group of issues which papers on the subject. The development of also requires discussion: privacy, confiden- national models for trading partner agree- tiality, negligence and protection of intellec- ments, efforts of international agencies such tual property. as the United Nations Conference on Trade It is now generally acknowledged that and Development (UNCTAD) and the legal considerations do not present any World Trade Organization (WTO) in rede- major impediments to EDI implementation, fining international trade and maritime but organizations do need to recognize laws, and the emerging national trend when it is prudent to seek advice. There are toward redefining evidence acts and the now well-tested models that would help to rules of evidence under electronic com- develop EDI guidelines and agreements merce, make this a very dynamic topic. In practically and realistically. addition, technological trends of cryptogra- Wherever statutes require documents to phy, security and signature devices both be written or signed to be legally effective, help and confuse the issues further. these were originally enacted to preclude This section is intended as a basic oral communication, not to exclude elec- introduction to some of these legal issues. tronic communication, which could hardly Anyone requiring advice should seek have been foreseen up to 20 years ago. The guidance from a professional. flexibility of commercial law will generally ensure that the writing and signing require- THE LEGALITY OF EDI ments will not be a major inhibitor for EDI. Agreements such as trading partner Any discussion concerning the legality of agreements, designed to cover the transi- EDI raises the question of how EDI systems tion from paper to electronic trading may accomplish the traditional legal between organizations may be informal or purposes of paper communication. These formal. At both national and international are typically concerned with authentication levels there is a trend toward greater use of and permanent storage of information; trading partner agreements, dealing specifi- communication of trade terms and condi- cally with commercial and technical issues. tions; and compliance with laws that Many national EDI associations now require certain legal information to be have appropriate national trading partner "written" and "signed." model contracts for use by their members. On closer examination other issues arise, such as: AGREEMENTS * How should trading partners and net- Many of the national models were based on works divide risks for errors among the early efforts of the International Cham- themselves? ber of Commerce (ICC). As long ago as * How will government record storage and 1987, ICC developed a set of voluntary control requirements be applied to guidelines for international transactions: electronic communications? the Uniform Rules of Conduct for Inter- * What will replace the hard copy audit change of Trade Data by Teletransmission trail? (UN-CID). The rules were useful as general Legal and Regulatory Requirements 1 principles, but they have been substan- * Should there be rules on signature? tially reworked for national models. They sought to define an acceptable level of Rules on applicable law and for dispute professional behavior and to secure a resolution have since been added by other common approach and identified some of authorities. the problem areas as risks in transmission; acknowledgments; return receipts; secu- INTERNATIONAL LAW rity; data logging; and storage. The objectives for many model agree- At the national level, service providers and ments and for much of supporting control others are subject to normal legal liability system design include: principles but the international situation is complicated by the question of the choice * system reliability and security; of law. * ensuring that network(s) have the Three connecting factors have been necessary level of checks and controls; used by various legal systems to determine * maintaining a detailed audit trail which law should obtain in an interna- consisting of a mixture of acknowledg- tional tort: the law of the forum in which ments, track-and-trace numbers, audit the case is brought; the lex loci delici (the logs and network reports; law of the place where the wrong oc- * systematically matching and reconciling curred); and the so-called "proper law" or audit trail information with messages; a variation on this concept (which is * investigating inconsistencies and record defined differently, for instance, in English reconciliations; and American law.) * keeping an original log of data sent and In the current legal situation a claim for received; damages could be differently decided * regularly auditing the system for simply because of the different connecting weaknesses in reliability and control. factors in the forum selected. The court itself may decide which legal system The UN-CID recomendations for should apply. formulating a trading partner agreement emphasized the following concerns: CONTRACTS * There is always a risk that something EDI is changing the way that businesses may go wrong. Who should carry that negotiate and agree contracts. Technology risk? Should each party carry its own or is also changing some of the potential for would it be possible to link risk to conflict. These changes raise issues of insurance or to a network operator? control, ownership and liability. * If damage is caused by a party failing to It is important to determine exactly observe the rules, what should be the when a contract or agreement has been consequences? This is partly a question made. Regulations may be applied in some of limitation of liability. It also has a jurisdictions which distinguish between bearing on the situation of third parties. instantaneous communications and de- * Should the rules on risk and liability be layed communications. covered by rules on insurance? EDI technologies permit multiple, * Should there be rules on timing, that is, simultaneous transmissions and receipts. the duration within which the receivers Any number of offers to tender may be should process the data, and so on? made around the world and the offerer * Should there be rules on secrecy or may demand responsive bids in diminish- other rules regarding the substance of ing time periods. the data exchanged? EDI standards bodies generally dis- * Should there be rules on a professional claim responsibility for deciding if or nature such as the rules applicable to when the interchange of data forms a the banking and securities industries? contract. In the United States the EDI * Should there be rules on encryption or Association's Programming Guide states: other security measures? "The legality of EDI data as binding 2 Information Technology and National Trade Facilitation: Guide to Best Practice contracts is left to the marketplace and * Value Added Networks (VANs) can negotiation between individual buyers and keep records of messages. For instance, sellers, shippers and carriers, payers and the SWIFT system for international fund payees." transfers keeps an inventory of data. The Bolero program currently under Networks can archive audit trail or way in Europe is a rare example of tech- message content data for legal and nologists, lawyers and legislators working auditing purposes, subject to user together to help adapt existing law to the acceptance, disclosure procedures and new technologies of EDI. Bolero has been privacy legislation. awarded a grant of ECU1.83 million * Each user can keep a data log of mes- (US$2.6 million) by the Economic Commis- sage content, electronic signatures and sion for Europe (ECE)to test electronic audit and control information (the UN- alternatives to maritime documents such CID rules require use of a data log). as the bill of lading and the sea way bill. * An audit trail can be maintained as Bolero is based on the United Nations evidence of what happened to the data EDI for Administration Commerce and between receipt and final archiving. Transport (UN-EDIFACT) IFTM (Trans- * Cryptography can be used to inhibit port) messages and central registries for data fabrication. bills of lading and individual subscribers to the system. Many banks today only An important debate at the national require confirmation that their counterpart level concerns evidence law. The role of has electronic access to the bill of lading; documentary evidence in commercial the sea way bill is now recommended for disputes, in trade practices litigation, and use in every case except where there may in revenue proceedings (for example, be a need for a document of title. Bolero is income tax, sales tax and customs and aiming to solve that particular problem. excise matters) is currently crucial. Legisla- tive changes are necessary in order to EVIDENCE overcome this hurdle. Paper documents are traditionally ac- DATA DISCLOSURE cepted as undisputed evidence. Paper is long lasting and normally allows changes Considerations of data disclosure under or additions to be clearly visible. Electronic the existing laws involve data content and messages, being intangible, are fundamen- contractual obligations for confidentiality. tally different. The data may lead to legal issues when Paper and electronic networks are they contain erroneous information, merely media that carry information. As defamatory or slanderous content, confi- such, it is possible to give electronic dential material or contractual liability networks characteristics which may make disclaimers by providers. them equal or superior to paper, not only Many EDI users impose contractual as carriers of information, but also as obligations to keep all private trade data evidence. However, a legal technicality- confidential, although such confidentiality the best evidence rule-may also question clauses rarely apply to data moving in all the admissibility of an EDI record. The directions. At the same time, providers best evidence rule generally requires that maintaining data records of EDI activity in the original of a writing is presented as different jurisdictions may be subject to evidence. At the same time, computer different national laws regarding disclo- output is acceptable as court evidence sure. when supported by technical witnesses. There may be other considerations, There are precedents to the admissibility of such as: such evidence, for example, the Watergate and the Iran Contra trials. * international confidentiality principles; To enhance the credibility of EDI * transborder data flows; records the following external techniques * technology transfer restrictions; can be applied: * allocation of loss; Legal and Regulatory Requirements 3 * intellectual property. with electronic data. * The need to protect the public from LEGISLATION damage which may arise from electronic communications such as in the areas of Although there is nothing yet that can be privacy, consumer protection and infor- described as a body of EDI law, legislation mation confidentiality. on EDI matters may fall within these * The need to protect intellectual property categories: such as computer software and design. * The need to meet international obliga- * The need to facilitate the admission of tions, such as the OECD privacy rules electronic communications as evidence in which cover transborder data flows. court. * The need to formulate rules about how CONCLUSION people can communicate electronically with governments, including require- If correctly planned and implemented, EDI ments to maintain electronic records, an can answer all fundamental legal objections area increasingly important to revenue which may be raised. However, it is indis- authorities such as taxation and customs putable that much of the law needs revision authorities. to be consistent with EDI, even if in most * The need to prohibit unauthorized access cases EDI can be safely used before the to computers and unlawful interference laws are changed. 4 Information Technology and National Trade Facilitation: Guide to Best Practice 2 Skills and Technology Infrastructure This chapter categorizes the skills and local operations of an international VAN. technology infrastructure required in Or it may be necessary to install a purpose establishing EDI for trade facilitation. As it built system. Naturally, if this were the may be even more useful to know why case, further skills and specialist personnel certain skills are needed and how they will would be necessary. be employed, this section has been orga- The most important technological nized in three parts: the introductory skills infrastructure requirement, apart from the and technology resource catalogue; a EDI host, is a modern telecommunications description of best practices in EDI and network. Most end user connections will be trade facilitation applications; and a discus- adequately served by dial up connections; sion of factors which could inhibit imple- probably less than 5 percent of connections mentation. will require leased lines. Connections will need to be made fairly quickly and techni- TECHNOLOGY AND SKILLS cal expertise be readily available if the INVENTORY implementation is not to suffer from exter- nally induced delays. At the national level A major trade facilitation EDI project can be the telecommunications authority will need carried out at the strictly technical level by to offer a wide range of international a small team of people; perhaps as few as connections and circuits to all of the com- two or as many as five specialists. They will mercial world's trade hubs. need to be supported by end user staff and Thereafter implementation depends a range of other professionals, but in purely upon external factors: local application technological terms a small team is ad- systems, knowledge of local systems, equate. willing and competent pilot trading part- The team should have skills and experi- ners and an appropriate business and ence in EDI and trade facilitation imple- administrative climate to encourage partici- mentations, knowledge of the theory and pation. practical application of EDI standards, especially UN-EDIFACT trade and trans- KEY TECHNOLOGY AND SUPPORT port messages and of a range of application ELEMENTS software. The team would also need experi- ence in interfacing or integrating EDI with For EDI and trade facilitation to work to the application systems. Professional skills required levels of effectiveness there is a set required are project management, educa- of technological preconditions: tion and teaching experience (at least as practitioner, if not end user), and manage- * an adequate telecommunications net- ment and senior client liaison skills and work; experience. * wide acceptance and use of IT in the In order to do their job properly, mem- public and private sectors; bers of this team would need access to * a pool of IT people with the right skill adequate and appropriate computer pro- sets, trained to appropriate levels; cessing facilities, complete with EDI host * hardware, software and communications software and network connection capabili- vendor infrastructure; ties (see Annex 2 , 3 and 5 of this report). * educators and trainers; The host EDI server may already be * access to practical skills and experience available from a local VAN or through the in business processing reengineering Skills and Techinology Infrastrutcture 5 technology, and EDI; originally designed. The two conditions * willing and able end users in both the are often significantly different. public and private sectors at national, Reviewv knowoledge Of wzihat is possible. This industry and enterprise levels; involves an awareness of, and education * supportive external agencies, such as in, all of the technologies previously EDI associations, article numbering discussed. It also involves knowledge of associations (UCC-EAN), trade facilita- case studies, understanding of trading tion bodies, Chambers of Commerce, partners' business processes, and aware- standards bodies, and industry bodies; ness of what is required for an interna- * supporting international agencies, such tional alignment of systems, both in as EDI associations, UN-EDIFACT, technology and business practice. UNCTAD, International Chambers of Reengineering. Having documented Commerce, World Customs Organiza- existing systems, and evaluated how they tion, World Trade Centers, trade facilita- work, it is now possible to redesign and tion bodies, international aid and reengineer current systems, based on a lending agencies, and major national knowledge of what is possible. and international VANs offering EDI Standard messages. Having reengineered and electronic commerce services; the information flow, it is now necessary to * banks and financial institutions offering examine the newly-defined data to be electronic funds clearance facilities with transferred between computers. This data financial EDI capabilities and corporate then needs to be compared with existing electronic banking services. approved standard messages. Message design. Message design and BEST PRACTICE approval can be a time consuming process, and should be avoided if possible. If new The term "best practice" is subject to messages are deemed necessary and the considerable abuse, but in the context of design process therefore unavoidable, they EDI-based trade facilitation it may be must be based on guidelines discussed in considered to be the implementation of the detail in the EDIFACT message syntax and goals of trade facilitation. Such implemen- design guidelines. tation involves simplifying processes, Pilot operations. An initial EDI group of removing excessive and obsolete controls, trading partners is then set up and ex- shortening and easing lines of communica- tended to a small number of competent tion, and using coding systems and EDI for trading partners. This involves installing rapid, accurate transfer of data between translation software, integrating that computers. This implies alignment with software with existing application soft- trading partners' systems and adoption of ware, connecting to a VAN's EDI service, world standards for best practice. and then exchanging test messages. As the Any plan to implement best practice pilot develops, it is possible for partners to conditions presupposes a knowledge of begin to trade electronically and to gradu- EDI, trade facilitation goals, the appropri- ally remove the paper systems it was ate technological infrastructures, willing designed to supplant. participants, and a nationally agreed Ramp utp. This is the term given to program. Key steps in the implementation extending the new electronic methods to of the process follow. the widest range of participants. The Feasibility st udw. Documentation of techniques used to enlarge the user com- information flows and identification of key munity and to encourage participation players for the early stages of implementa- may vary from mandating compliance at tion. one extreme to a range of financial and Project plan. This includes broad aware- business inducements at the other. Only ness, concept marketing, and an education full participation will yield the desired program. results. Docutmentation of existing systems. Progress reviewu. The management and Documenting the systems as they are the project review of each pilot and major currently being used, not as they were initiative will involve public and private 6 Information Technology and National Trade Facilitation: Guide to Best Practice sector organizations, as well as many of Funding. Clearly the most common the major trading partners. The purpose of reason why EDI is imperfectly imple- reviews, apart from commercial and mented or adopted at all is lack of ad- operational necessity, is to confirm that the equate funding. This manifests itself at the original design parameters are being met, national and the enterprise level alike. At and to improve them where possible. the national level an inadequate telecom- Legislation. To the extent that legislation munications infrastructure, lack of a needs to be changed, attention should be sufficient base of IT equipment, or the paid to the Customs Act, the Evidence Act, absence of a body of skilled and experi- and any other legislation that concerns the enced people, would clearly make it validity of electronic commerce, taxation, difficult to achieve a critical mass of users. and banking regulations. There are some appropriate aids, such When presented in this fashion the task as the no tech-low tech initiatives, but an may seem overwhelming, but a properly effective telecommunications infrastruc- constructed plan will create a staged ture and an EDI VAN are the minimum approach, phased and balanced for local prerequisites for EDI, for which govern- conditions, with as much expert help as ment funding should be available. necessary. The plan should take into At the enterprise level there are a range account non-automated small- and me- of aids and inducements. "EDI In" and dium-sized enterprises through a local "EDI Out" service bureaus, fax input- variation of no tech-low tech EDI initia- output services, voice input-output ser- tives. Enterprises and industries from vices, and inducements from government nations of all sizes have already embarked and industry bodies can help. Loans for on such a plan. The fundamental prerequi- equipment and training to be repaid from site is determination. savings, the rental of hardware and soft- ware by VANS, fixed monthly billing and a BARRIERS TO USE AND range of shared operation options, all work PARTICIPATION under the right circumstances. Finally, banks and financial institutions On one hand, EDI has been seen as a tool may choose not to cooperate in clearing for the larger enterprise, the wealthy house functions for electronic funds industry, or for government departments. transfer, trade payments, and electronic The well-funded, well-resourced, techno- trade documentation. Whether this is the logically literate organizations are always result of cynicism, of perceived technologi- among the pioneers, and tend to dominate cal superiority, or the desire for market the standards-setting process. These are advantage over domestic competition, it the people who can afford the right soft- happens all too frequently, to the detri- ware and the right VAN service. ment of national interests. The banks and On the other hand, the unautomated financial institutions must become part of and hence technologically disenfranchized the best practice movement; they must be small companies fall even further behind persuaded to adopt EDI and financial EDI the wealthy organizations as the adoption from the outset. of technology widens the gap between the Vision and leadership. An EDI initiative is smallest and the larger organizations. rarely successful when driven from the Funding, however, is not the only bottom up. Success requires a strategic barrier to the use of EDI. Even in wealthy plan and a shared vision of the outcome countries there can be equally effective and benefits. Without this form of leader- barriers, such as competing infrastruc- ship from the highest levels or from a tures, lack of leadership, and inappropriate strong, unified commercial interest, the message standards. result is a fragmented effort with conse- This section briefly looks at the most quent wasteful use of scarce resources. common limiting factors, to sound a note This breeds only partial commitment and of caution about the methods that may be gradual disinterest. It may be necessary to adopted and implemented at the national mandate certain national processes in level. order to achieve the necessary critical mass Skills and Technology Infrastnictire 7 vital to the overall success of the initiative. or unfocused initiatives can breed unneces- Knowledge. Until a sufficient number of sary competition among vendors, each people concerned with trade facilitation seeking competitive advantage over the issues know what is possible, it is very other. An absence of leadership can also difficult to make progress. Unless a cam- encourage an environment of noncoopera- paign of awareness and education is under- tion among vendors, nonstandard ap- taken at the right level and for the right proaches and, consequently, higher costs duration, the pioneering implementers and and noncompliant systems. The choice and users will have an uphill task. An aware- adoption of industry, national, and interna- ness and education program is possibly the tional standards, and their application in a most important key to success, and con- uniform fashion, so as to be aligned with versely, the lack of such a program is the overseas trading partners, are particularly most likely to cause failure, delay, or important. An uncoordinated vendor compromises. infrastructure leads to confusion and lack Too much competition. Lack of leadership of cohesion. 8 Information Technology and National Trade Facilitation: Guiide to Best Practice 3 Local Business Considerations Even if all of the technological precondi- into account, therefore, are government and tions for success are present in a business local business practices, language, cultural community, there are always other factors and religious practices, and external influ- which have to be considered in overall ences such as multinationals. project planning. A good telecommunica- tions and computer infrastructure, com- GOVERNMENT AND LOCAL BUSINESS plete with adequate skills and knowledge PRACTICES base cannot, on its own, guarantee success. It needs the willing support and active In any area selected as a potential location cooperation of the public sector, most for a technology-based initiative, the local particularly of major government sponsors government ideally needs clearly articu- in the early stages of the project. It also lated national business objectives, and requires a similar attitude from the local capable and honest public servants to business community and any multination- delegate their implementation to. To be als who may be present. If we add to these carried out successfully, these objectives needs those of adapting the traditional local must be free of political and ideological way of doing business and the practices biases, and be seen to be fair to everyone. that have evolved over the centuries to the Traditional Confucian practices, post newly evolving methods of international colonial nepotism and influence from trade then we can assess the initial chances traditional power sources may not be for success. helpful. Many of the case studies for this As a benchmark, take the case of several project yielded instances of these types of Asian countries who have built on a less environments stifling the necessary spirit of than optimum technology base in order to enterprise for technology-based initiatives. implement a national plan. Their achieve- So the judgment is whether the govern- ments have come through governments ment is actually able to distance local who have been able to marshal support practices and pockets of perceived self- from all sectors of the local business com- interest from a project whose dimensions munities in committing themselves to are defined and measured by technological international best practice for the national competence, and success in the interna- good. This concept of the national good has tional arena. helped them to overcome sectoral opposi- It must be remembered that the project tion to the alignment of local practices with will require a great deal of help from international systems. organizations who have already had long On the other hand, take the example of years of experience with the government several advanced western style economies concerned. The cynicism that this type of who, for all of their investment in technol- relationship often engenders is hard to ogy, have failed to set a national agenda shake off, as some Latin American countries through a national shared vision, have left demonstrate. the leadership to the private sector, and as a It is also possible that, in some coun- consequence have secured unsatisfactory tries, the government may have to be the natioinal returns on investment and effort. conduit for funding, but may not be the This has happened because of the resulting right organization for the management role, fragmented efforts, duplicated resources nor for the leadership tasks so vital to the and redundant competition. success of the project. Major nontechnical factors to be taken Local Business Considerations 9 CULTURE AND RELIGION other work to do, or may not be able to fill in the forms accurately. EDI and simplified These are factors with a diminishing automated trade facilitation can help a impact on trade issues but cannot be great deal, but it still requires genuine ignored with impunity. Business practices knowledge of the appropriate language. are often very closely related to cultural Variations in government style, in local issues. business culture, in religious impacts on Japanese and Korean business practices work practices, religious holidays and the date back hundreds, perhaps thousands of local working week, language and the years. They were designed to protect local availability of romance language speakers, industries and craftsmen, to lock out all have an effect on the successful imple- competition, and monopolize suppliers. In mentation of a major trade facilitation an electronic world, these concepts are initiative based on technology. diametrically opposed to the concept of open standards and cooperation between EXTERNAL INFLUENCES competitors on technological and stan- dards issues for the larger good of the Po-t-colonial governments tend to have global industry. very strong or antithetic relationships with Other cultures tend to emphasize the the governments of ex-colonizers. Such small family unit business to the detriment relationships can sometimes distort objec- of national organizations. There are some tive judgments, quite often on technologi- of these tendencies at work in Hong Kong cal issues. It would be dangerous to and China, for example. Family, clan, tribal assume that a previous colonial adminis- and cultural allegiances often transcend tration would be in a better position than open business relationships in developing others to understand local conditions, and economies, thereby making open business that its solutions to specific problems networks more difficult to establish than would therefore be more appropriate than they need be. solutions from elsewhere. There are some EDI standards and documentation have classic examples of EDI systems being so far been developed in the official UN exported to countries least suited for them languages. In order to participate in the merely because such an assumption was EDIFACT movement, countries either have made in the planning stages. to have a nationwide fluency in a Euro- The influence of multinationals would pean language or they have to convert the normally appear to transcend many of standards and documentation to local these local variations and special circum- languages. In some instances, especially in stances, operating as they are in interna- ideographic languages, they first have to tional conditions. But in practice, the agree to a common language for standard successful multinational has absorbed codes for information interchange (like the many of these local factors, while appear- American Standard Code for Information ing to be impervious to them from an Interchange, or ASCII). Korea, China and outsider's perspective. The experiences of Taiwan have all completed this work. multinationals, and the lessons they have English is not the primary spoken learned are extremely important to this language in most countries, but it is the type of technological initiative. They have primary written language for international had to find their way around most of the trade. This often creates problems. Even problems facing a trade facilitation initia- when an organization has someone with tive and are generally the keenest to see it sufficient fluency in English to fill in succeed. official forms, that person usually has 10 Information Technology and National Trade Facilitation: Guide to Best Practice 4 Investment Costs and Benefits The simplifying and speeding up of trade partners. Efficiency improvements in vessel information flows offer significant national turnaround have attracted new entrepot benefits. At one level they ensure that and distribution business to Singapore. efficient approvals and information flow Advanced electronic commerce and EDI can be processed with a smaller number of facilities enabled Australia to increase their steps, fewer people, and in less time- lead in tourism revenue growth. The ability offering savings to government depart- to electronocally communicate with their ments and commercial users alike. But the northern hemisphere trading partners has downstream results of these efficiencies are resulted in a four week extension of their even more important. supply season each year for a New Zealand For example, the use of accelerated, produce industry. simplified systems and EDI to preclear At the enterprise level, the adoption of imports and exports means that goods can quick response (QR) and just in time (JIT) be loaded and unloaded in the most effi- strategies, particularly those supported by cient manner, problems can be anticipated EDI, are enabling textile and apparel and solved before they become problems, manufacturers all arross Asia and Latin facilities can be properly scheduled and America to dynamically satisfy variable maximum use made of the road, rail, ports customer demand and thereby gain signifi- and harbor infrastructure and installations. cantly larger proportions of their business. To take a simple example: if a ship can Auto manufacturing operations all over the be processed in half a day or less rather world are reporting savings of US$200 per than the day or more it may currently take, assembled vehicle through JIT and EDI then the infrastructure capacity is effec- practices. Major multinational retailers are tively doubled: twice the cargo, twice the obtaining a greater variety of fresh produce number of ships, twice the number of from all corners of the globe while simulta- containers. The result is increased harbor neously achieving dramatic savings duties, increased excise, increased revenue through the use of QR strategies. In some from income tax, and the company profits cases supermarkets have increased profits without increased investment in infra- threefold over the last ten years, while their structure. inventory float has been reduced from three Singapore claims that properly applied to four months of supply to less than one trade facilitation is already saving the week. country in excess of 1 percent of its gross Best practice trade facilitation and EDI domestic product (GDP) each year. Returns based industry initiatives have not only on the investment in the national trade produced economic advantages, they have facilitation initiative, Singapore Network turned into a marketing tool for their Services (SNS), came during year two of advanced users. New business is being operation. Taiwan (China) and Korea have attracted to EDI-compliant enterprises at similar stories to tell. the expense of those which are not. Hence the big picture shows more So the ultimate question might be not efficient trade, higher government revenue "What does EDI do for me," but "What will and the ability to defer government invest- the lack of EDI do to me?" ment in major infrastructure projects by optimizing use of existing installations. COSTS AND BENEFITS The benefits of these efficiencies trans- late to wider attractions for the trading There is no magic formula which can Investment Costs and Benefits 11 guarantee a safe return on investment in cost of educational materials and external EDI and best practice. The most successful advice and assistance. Private industry case histories are from countries which and far-sighted vendors may be willing to were able to conceptualize solutions to contribute funds for this early part of the their fundamental trade processing prob- project and make it possible to initiate an lems and then committed themselves to ambitious program. the approach, with the conviction that It should be noted that to be truly substantial benefits would follow. Much of effective the awareness campaign needs to the investment made for implementing be focused first on national advantage and best practice and EDI is indirect in nature, then on individual key industry benefits, but must nevertheless be taken into ac- finally concentrating on the individual and count: for example, the commitment of the small to medium enterprises. The activities time of key people to the project. They should embrace all government depart- represent real costs, but they may possibly ments, all commercial enterprises and all be absorbed within normal budgets. quasi governmental authorities and Investments also depend upon the organizations. Ultimately the campaign scope and the size of any project. To totally should become institutionalized through reengineer a nation's trade process from a the education sector. base of clerical and bureaucratic systems, The evolution from the launch of the involves commitments of a significantly program to its adoption by the education higher order than an individual enterprise system may take about five years. The adopting best practice in an environment public sector will almost certainly have to where there is a good technology infra- bear the brunt of these establishment costs structure and EDI is a common practice. if the job is to be done properly, especially Two extremes illustrate the differences in the market start-up and consolidation in more detail: the macro view (a national phases. A nascent technology vendor-VAN perspective) and the micro view (an industry would be unable to fund such an enterprise-level perspective). Take the extensive program in advance of revenue example of a nation or territory where although it should be able to make an there is no existing national program and increasing contribution as the market few, if any, EDI users. Assume also that the expands. This phase of the project is vital country is operating the traditional paper for a successful national implementation and lengthy approval customs export- therefore sources of funding must be import system. Further, there is no natural established at the outset. candidate for a national organization The third cost category is the technology charged with providing the technology, provider. It may be necessary, at one ex- commercial leadership and project man- treme, to establish a new organization and agement. There are several categories of to install new equipment and software in cost needed to build the infrastructure and order to provide the necessary technologi- awareness from the ground up, assuming cal facilities and level of service required that there is an adequate basic IT infra- for a national approach. To prepare for this structure (telecommunications, computer possibility, and in order to make realistic usage and skills pool). commercial decisions, this exercise will The first is discovery, which includes involve feasibility studies, cost benefit executive time and travel, and external assessments and business planning activi- advice, such as some external agency ties. Thereafter the organization will involvement in the debate on objectives require investment and support until the and options, potential scale and sources of enterprise breaks even, or achieves a level funding. of business performance at which it may Next is azwareness, or the creation of a attract private sector funding or may even national promotional campaign, education be completely privatized. courses and conferences and the use of the There are two categories of investment, media, spread over a long period. with many options depending on local Direct costs will include the costs of the conditions. The first involves the planning campaign, media costs, conference costs, and feasibility activities. Much of this 12 Information Technology and National Trade Facilitation: Guide to Best Practice work may need to be carried out by technical people, and the technology, are external consulting bodies or experienced provided by the local authorized organiza- international agencies. They need to be tion-the local VAN. VAN revenue comes supported by local executives and govern- from software sales, education and train- ment officers, for credibility, for technol- ing, consultancy, and network traffic, ogy transfer and for continuation of perhaps from supporting electronic com- commitment. merce services as well as EDI. Plans need to take into account all Network traffic increases with volumes direct costs of technology, staff and other of transactions and the variety of applica- resources. Service pricing can be a conten- tions. But since the tariff for these services tious issue. It may be tactically necessary must be both attractive and competitive, to offer inducements and attractive pricing and since it is based on the most efficient plans for early users; but once the service contemporary computing and telecommu- is established it is important to levy fair nication technologies it requires a sizable and reasonable pricing tariffs. In any case, number of users and level of activity to the attention of competing vendors and generate a break even income. Virtually overseas trading partners will ensure that every case study illustrates a cash flow hidden subsidies cannot be provided for break even of between 48 and 72 months long. (four to six years) on this type of opera- The second cost category-cost of the tion. So the challenge is to bring forward technology service, or the local VAN, if it that break even point or to support activi- is to be provided locally-may be defrayed ties until returns can be made. in a number of ways. It may be possible to This final point is crucial, and explains subcontract the entire task to a third party why commercial vendors are not breaking and avoid direct investment in technology down doors to fund the start-up opera- in exchange for contractual exclusivity for tions of electronic commerce initiatives in a number of years. It may also be possible developing countries. The funding of the to minimize investment by entering into a infrastructure and market development joint venture with a new or existing costs are outside the normal span of venture for the right contractual arrange- commercial viability for existing technol- ments. Or it may be tactically necessary to ogy suppliers. Their business is to provide invest in a brand new organization and access to the specific technologies neces- technology. sary for electronic commerce but as the Tempting though it may be to debate case studies illustrate, technology alone is this issue at length, the technology pro- a relatively small cost component when vider is relatively insignificant in the compared to overall costs. overall financial picture. Case studies have The range of case studies also illustrates shown that the direct costs of technology that there are no simple rules to determine are typically considerably less than 10 costs, benefits and break-even points. But percent of the total project costs, in some there is an enormous amount of evidence cases as little as 3 percent. The major cost concerning the number, sequence and scale elements are the costs of building aware- of activities. Local business practice, ness, of working with potential users to policies and business accounting methods prepare them for electronic commerce, of need to be applied to individual cases. developing and designing messages and What follows is a guide, a set of tools for guidelines and of reengineering systems. decision making, for measuring progress, The technology vendor is often external to and to help set financial criteria for judg- many of these activities. ing success. The major costs are people costs, much of them invested by end users and govern- COST AND REVENUE CATEGORIES ment departments off-budget. But they are real costs nevertheless. There are four main cost categories: Finally, there are the costs of iniplemen- tation. Increasingly, as the project matures * Research and business planning. These and the local VAN grows in experience, the activities may be expensed or budgeted. Investment Costs and Benefits 13 * Investment in IT equipment, software, EDI software packages (US$500 to US$3,000 personnel and telecommunications per copy); assumes consultancy and imple- services necessary for a national EDI mentation charges of up to US$2,500 for a service. There is a fixed minimum entry small user; and network, communications level price plus a variable cost, depen- and service costs in the range of US$50 to dent upon activity and usage. US$250 per month. The variables are * Implementation activities, composed of implementation and integration complexity, the minimum activity necessary for application types, types and numbers of establishing EDI technology, messages messages and transaction volumes. These and reengineering infrastructure, and a figures do not separately identify the more variable component set by activity and expensive IT implementations for a major take-up volumes. reengineered process, nor do they identify * A variety of indirect and associated any no tech-low tech EDI charges for activities such as management, working nonautomated users. For the sake of sim- party and committee work, legal, audit plicity, these have been bundled in with and security tasks and direction on overall user's costs and revenue figures. ownership and funding activities associ- Models B and C vary only in the size of ated with the national initiative. the economic communities that they serve, that is, their potential user base. To some The revenue may come from two extent, the size of the economy influences sources: the speed and rate of take-up, if only because of the scale of the end user's own i end user payments for EDI products and resources. services; Model B assumes 100,000 enterprises, * privatization, and equity participation in Model C 250,000 enterprises. The three the national technology initiative, or models encompass population of up to 25 from charges for contracted access to million people. government information. It must be remembered that the costs quoted are those directly attributable to a The scale is dependent upon levels of modest yet highly focused reengineering activity and the overall potential of the effort of EDI-based trade facilitation. Extra initiative. Three different models are given product development, extra services, a below for general guidance. more ambitious roll out program and failure to meet what are proven to be THREE TYPICAL MODELS achievable user number targets, can easily result in multiples of the costs illustrated. The three different take-up models pro- The models assume 10 areas of business posed in this section illustrate variations in costs covering the first five years of a scale, activity, and cash flows. The first, national EDI initiative: discovery, aware- Model A, is based on a relatively small ness, feasibility, strategic business plan- scale initiative in an Indian Ocean, Carib- ning, technology acquisition and imple- bean or Pacific island community. It as- mentation, project planning, implementa- sumes about 20,000 enterprises in total, of tion planning, technical implementation, which 5 to 10 percent are involved in the operations, review. Thereafter the business transport and distribution aspects of the should be concentrating on consolidation import-export trade. The remaining enter- and the application of market-driven prises are concerned with the normal pricing to recover initial investments, if activities of primary industries, retail and they have not already been recovered by wholesale, transport and distribution, this time. The list does not include end user service, and government sectors. The first technology, which is taken to be outside the five to seven years of the business plan national initiative business planning. A would be aimed at capturing all partici- contingency plan can assume that one user pants in international trade and the most in three will buy new equipment, at an important domestic enterprises. estimated 1995 price of US$3,000 for hard- All models use global average prices for ware. 14 Information Technology and National Trade Facilitation: Guide to Best Practice Model A Costs Type Costs (US$000 per year) Total Years 1 2 3 4 5 Discovery 100 100 Awareness 150 100 50 50 350 Feasibility 100 50 150 Strategic business planning 100 100 Technology acquisition and implementation 400 250 250 300 300 1500 Project planning 50 50 100 Implementation planning 50 50 Technical implementation 100 100 100 100 100 500 Operations 50 150 150 150 150 650 Review 50 50 50 50 200 TOTALS 900 850 650 650 650 3700 Cumulative 900 1750 2400 3050 3700 Model A End User Revenues Number of Users 5 50 250 500 1000 Revenue (US$ 000) 50 250 800 1500 3000 Cumulative 50 300 1100 2600 5600 Model B Costs Type Costs (US$000 per year) Total Years 1 2 3 4 5 Discovery 200 50 250 Awareness 50 250 250 200 100 850 Feasibility 300 200 50 50 50 650 Strategic business planning 200 50 250 Technology acquisition and implementation 750 500 500 600 600 2950 Project planning 100 100 50 50 50 350 Implementation planning 100 100 50 50 300 Technical implementation 200 300 300 300 300 1400 Operations 100 300 400 400 450 1650 Review 50 75 75 100 300 TOTALS 1900 1900 1725 1725 1700 8950 Cumulative 1900 3800 5525 7250 8950 Investment Costs and Benefits 15 Model B End User Revenues Number of Users 25 200 500 1000 2000 Revenue (US$ 000) 250 1000 2000 3500 7000 Cumulative 250 1250 3250 6750 13750 Model C Costs Type Costs (US$000 per year) Total Years 1 2 3 4 5 Discovery 500 250 750 Awareness 250 1000 1000 500 250 3000 Feasibility 500 500 200 100 100 1400 Strategic business planning 100 250 50 400 Technology acquisition and implementation 2500 1000 1000 1250 1250 7000 Project planning 100 250 100 100 100 650 Implementation planning 250 250 300 300 300 1400 Technical implementation 500 750 750 1000 1000 4000 Operations 250 750 1000 1000 1250 4250 Review 100 250 100 100 500 1050 TOTALS 5050 5250 4500 4350 4750 23900 Cumulative 5050 10300 14800 19150 23900 Model C End User Revenues Number of Users 50 350 1000 2000 5000 Revenue (US$ 000) 500 2000 4000 8000 20000 Cumulative 500 2500 6500 14500 34500 All models illustrate a break-even point FROM DIFFERENT PERSPECTIVES after a period of 48 to 54 months of opera- tions, which is consistent with virtually From a wider perspective, electronic every nonsubsidized initiative. The critical commerce is directly analogous to the dependencies are obviously take-up rates telecommunications and broadcasting and end user revenues. Both are dependent industries: the payback may be longer than upon service, hence on initial investment. for many other commercial enterprises but All are dependent upon the awareness the revenue stream is also correspondingly campaign working. longer, and ultimately more profitable. The ratio of technology costs to total Specific costs are a function of scale. costs is around 15 to 20 percent. The pro- Direct costs are a fairly small proportion of portion of costs attributable to end user and the total. It is unlikely that all technology private sector efforts, as opposed to direct and external advice costs exceed 10 percent investment by the public sector, are around of the overall total of a national trade 30 percent of the total. facilitation EDI initiative. The balance of costs is concerned with people, from the technology provider and end user organiza- tions, in both public and private sectors. 16 Information Technology and National Trade Facilitation: Gutide to Best Practice The national benefits need to be assessed inevitably time consuming. Figure 4.1 in the same way as costs. They will involve demonstrates how goods are delivered direct benefits of efficiencies, of cost reduc- within a couple of days, while the complete tions, in better use of resources and in information processing task and payment deferral of capital expenditure. Growth in takes several weeks. One major supermar- trade needs to be considered, as does the ket admitted that, prior to EDI, its seven- value of the new skills and industries day accounts were taking 27 days to pro- fostered by the new technologies and new cess, with each supplier invoice costing techniques introduced to support the trade them an average of US$30 to process. facilitation initiative. One percent of GDP, In information systems time causes a or even a fraction of that may be incentive phenomenon called "float." It is an inbuilt enough to justify the investment. allowance by the system for investment in At the micro level of the individual time, resources and inventory necessary for enterprise, the costs and potential benefits the system to replenish stock. Float is often are much more tangible, and more immedi- measured in "days of stock." A key objec- ate. Figure 1 represents the normal situa- tive in any business system which has to tion, in which companies use computers to control stock is to reduce float. At one time create paper forms, then place them in float could be measured in weeks or envelopes and post them to their suppliers months. Now, when an EDI-facilitated or customers. It provides the end-to-end business practice has been properly imple- example of buying and selling, or the mented, float can be measured in days, and purchase order to receipt and payment in the future perhaps in hours. process, now often referred to as "the Of course, EDI is not the only reason: supply chain," especially when it includes bar coding systems, electronics funds all of a company's suppliers. transfer at point of sale (EFTPOS), and Even in this simplified presentation, modern computer replenishment systems, there are 10 steps in the complete end-to- all play their part. But EDI is the key end process, each one open to errors and facilitating technology; it is the reason why duplication. Data entry into computers for accurate information can be rapidly trans- internal systems is done on several differ- ferred between computer systems, which in ent occasions (It is a well recorded claim turn provides precise information on float, that 70 percent of computer output becomes thus making QR and JIT systems practical. someone else's computer input). All this is Figure 4.2 illustrates the impact of EDI Time: Days or Weeks? > Order placed Invoice received Check sent EFT initiated BUYING FIRM PAYMENT POLICY CREDIT POLICY SELLING FIRM Order Goods Invoice Check Check Funds received shipped sent received deposited available Figure 4.1: Order to Receipt Cycle (before EDI) Investment Costs and Benefits 1 7 Time Line: Hours or Days > Order placed Invoice received EFT initiated BUYING FIRM SELLING FIRM Order Goods Invoice sent Remittance Funds received shipped advice available received Figure 4.2: Order to Payment Cycle (with EDI and FEDI) and financial EDI on the "order to receipt mainframe computer could cost in the and payment cycle" described earlier in region of US$25,000. They would also have Figure 1. It shows that the computer- to connect to a VAN, rent an electronic generated order goes straight into the mailbox, and initiate a program for educa- suppliers' computer in a matter of minutes tion and training. or hours. That information is processed by Staff time and expense to interface the their order entry system and goods sup- translation software package, and then to plied thereafter. reengineer systems would, of course, be The same is true of payment. No paper internal to the individual company. Clearly is used by the EDI facilitated system and the costs would be significant, but compat- therefore no data entry, no mistakes and ible with a normal mainframe project. reworking, no paper storage and retrieval Cases of this size and complexity, although and no picking paper up and putting paper representing a high proportion of transac- down, that is, processing, need take place. tions exchanged within a given community, The time taken to process paper in the are nevertheless the exception rather than system is often called "the information the rule. Well over 90 percent of all EDI float." By reducing the information float it users around the world use a personal is possible to reduce inventory float and computer for the purpose. thereby make the whole process faster, The annual repetitive costs of using the simpler and cheaper. network for EDI are directly proportional to volume in the vast majority of cases. But SUMMARY this is quite inexpensive when compared to any other alternative. At a typical tariff of The costs of installing EDI by individual US$0.30 per 1000 characters sent and enterprises obviously vary with company received, this represents a fraction of size and project complexity. For example, today's postage rates. the organization might have a mainframe At the level of the smaller enterprise, computer handling high volumes of com- staff involvement may be measured in plex transactions. It might be necessary to hours, and one-off costs from a few hun- totally reengineer a function and to dedi- dred dollars to US$5,000. Repetitive costs cate a team to the task. In addition to are typically less than US$100 per month staffing the project team they would need a for an average small user. translation software package which, for a The benefits may take some time to 18 Information Technology and National Trade Facilitation: Guiide to Best Practice accrue, in line with the number of trading for trade clearances and trade facilitation partners who have adopted EDI practices. purposes. In addition there are many more But in some cases, for example, as in the times the number of users using EDI for case of a customs broker making export industry and efficiency purposes. declarations, the benefits begin to flow The cost-benefit case depends largely from the first day. upon local conditions and the local starting Generic benefits include direct cost point. But the more advanced trading savings, increased productivity, improved partners, at national and enterprise level, trading partner relationships, greater are beginning to demand EDI compliance marketing opportunities and reduced as a condition of doing business or con- inventory levels. Faster response times ducting trade in the future. Already some often lead to increased sales volume but in organizations will only accept new suppli- a greater number of smaller batches. EDI is ers if they can demonstrate an EDI capabil- the only reliable way to cope with this ity. There are cases of companies, particu- trend. New business opportunities, par- larly traditional, small, older firms, who ticularly from overseas trading partners, have gone out of business because of can develop simply because of EDI compli- inability, or unwillingness to comply or ance. disbelief in the need to comply. This has been particularly true of some middlemen CONCLUSION occupations. Ultimately there is an even harder fact to There are around 100,000 users of EDI consider. There is no longer any choice operating to national and international about compliance; the market has made the standards in 1995. The number of users decision for everyone. The remaining increases at a rate of around 25 percent choices involve timing, to a diminishing compound each year; and the volume of extent, and the level of participation. It may transactions and new applications at be possible to adopt a cosmetic approach, significantly higher rates. By no means all or minimum level compliance. But that of these users have experienced the sug- represents considerable pain for a limited, gested benefits, but many have and more and transient, gain. Market conditions will, expect to do so. in time, demand maximum participation The same is true for national initiatives. and the adoption of best practice for sur- In one form or another, over 50 countries vival. At the moment there are still oppor- are now actively using EDI. Traders in tunities for competitive advantage. virtually all these countries are using EDI Investment Costs and Benefits 19 5 Technology and Cost Options EDI is often seen solely as a tool for large slow when compared to many early firms and government. They improve their predictions. The reasons for this are rarely efficiency while their smaller trading those we hear at seminars, such as legal, partners are forced to absorb the costs of audit, security and network interconnec- installing EDI. Of course the issue is much tion issues. These may be reasons to defer more complex than that, but the fact is that serious consideration of EDI in the first EDI can be costly for small enterprises. place, but they are not reasons to stall This section looks at the real costs growth of existing users. involved in installing EDI and at how The real reason is the effort involved in these costs may be reduced so that even installing EDI and integrating EDI into the smallest of trading partners can justify business systems. This is the brick wall EDI. To the technical purist, the techniques facing all EDI implementers. involved in this low-cost EDI approach To take a practical example: may not fit the definition of electronic data Woolworths Supermarkets in Australia has interchange. But they can meet the same over 10,000 trading partners, the majority objectives, and at less cost. within Australia. About 80 percent of those trading partners supply goods which are EDI'S BRICK WALL eventually soid within the supermarkets, the remainder being goods and services for Of the estimated 100,000 EDI users in the internal consumption (among them build- world by end 1995, about 50 percent will ing maintenance, electricity, cleaning). be in North America, and the majority of It will take Woolworths three to five the remainder in northern Europe, with years to achieve full EDI capability. The sizable minorities in Japan, Singapore, company has to change internal disciplines Australia, and New Zealand. Also, t' re and processes. For example, it will no are indications that within three years the longer be dealing with vendors' invoices most rapid growth will be in the Asia- but rather with a new message, the SNM Pacific region. or Ship Note Manifest. It has to redesign EDI users go through a number of information flows, rebalance staff levels phases, but none of those phases have so against redesigned work practices, and far included exponential growth in num- redesign computer systems and databases bers of connections to trading partners. to handle EDI input and output. Even today the average EDI user is part of Woolworths' major trading partners, a community of less than 10 users. After all relatively few in number, are often multi- these years of experience, why is it so? nationals with large MIS departments. What are the real inhibitors and barriers to They will go through the same processes the rapid growth and success of EDI? as Woolworths, but to a different time scale Obvious inhibitors include the lack of and often for different motives. Hence adequate infrastructure, unsuitable soft- their priorities are often out of synch with ware or standards, cultural and language Woolworths'. impediments, monopolized telecommuni- A larger number of Woolworths' cations regimes, lack of skilled personnel, trading partners are mid-sized companies, and so on. But all these are rapidly being which use much less sophisticated forms overcome. Even in the advanced western of automation, from minicomputers and countries where these inhibitors no longer PCs to service bureaus. Once again they exist, the growth of EDI is still relatively have their own priorities, which do not 20 Informiation Teclhnology and Natioinal Trade Facilitationz: Guiide to Best Practice always match Woolworths'. Because they scenario by a variety of techniques, tech- are not Woolworths' most important niques that can reduce costs for all parties partners, Woolworths is not yet placing until such time as EDI is fully implemented pressure on them to implement EDI, so this across the whole community. category of trading partner has a little breathing space. NUMBERS OF EDI USERS AND EDI By far the largest category of MESSAGE VOLUMES IN AUSTRALIA Woolworths' trading partners are small to very small firms, many of them one-or two- The following Australian figures may be person businesses, often without as much useful as a model for what happens else- as a PC or any other form of automation where. except for a phone and a fax machine. In 1992 Australia had about 3,500 users Woolworths is placing no pressure on these of EDI, only 20 percent of whom were trading partners yet, but every trading actively using EDI as a business tool for partner knows that ultimately it must do general business benefits. The remainder EDI with Woolworths if it wishes to protect were using EDI because their major trading its business with the supermarket chain. partners demanded it (the auto industry, Table 5.1: Estimnated Business Message Volumes (1992) Countries Letters/Head Population Total Letters Business Mail in millions in billions in millions Australia 211.70 17.00 3.60 720.00 U.S.A. 715.20 250.00 179.00 35,800.00 U.K. 260.40 58.00 15.10 3,020.00 Singapore 119.60 3.00 0.36 72.00 Hong Kong 95.80 6.00 0.51 102.00 Of course, this is not just true of customs, government, taxation department, Woolworths; it is the same for all major and so on). These users were exchanging companies or hubs in EDI jargon. Australia around 3 million EDI messages a year in has around 500 hubs; their trading partners 1992. (and their partners) total about 800,000 to In contrast, the Australian Post Office 1,000,000. carried over 600 million business letters Many of these EDI hubs now realize (equivalent to EDI messages) each year, their predicament. They chose EDI for the many of them containing more than one familiar range of benefits, not the least of message (see Table 5.1). In addition, certain which is cost reduction. Yet most of them industries (for example, retail industries) are incurring not lower but higher costs use the telephone as the main ordering and through EDI because they are now running call-off medium. Fax is a popular medium hybrid systems: paper for the majority of for formal business messages, and telex, trading partners and EDI for a small but telegram, and electronic mail are also used. growing number of EDI partners. From these figures we can estimate that Clearly every hub would like to flick a there is a grand total of one billion business magic switch and turn all trading partners messages exchanged in Australia each year, over to [DI overnight. But to convert them all of which, potentially, could be converted all could take them well into the 21st to EDI messages. Considering EDI gener- century, maybe 30 years or so. ally involves two actions for each message: Experienced EDI implementers now a send-to mailbox and a receive-from realize there is no short cut to full EDI; it mailbox, the total potential for EDI mes- must be done thoroughly and patiently, and sages in 1992 is to the order of two billion. it will take some time. However, it should However, after six years of experience, be possible to overcome the hybrid system EDI has captured only 0.35 percent of Technology and Cost Options 21 potential Australian users and 0.15 percent of potential message traffic (see Figure 5.1). Actual Actual There are two other factors to bear in Messages - Users mind. First, as we use EDI to remove - 0.15% - 0.35% people from the administration of trading processes we will develop more and more \. specialized EDI messages to replace infor- Potential 99.85% Potential 99.65% mal telephone messages. Second, postal volume alone (that is, business activity) is growing at 3 percent to 5 percent per year worldwide. These two factors could within Figure 5.1: Message Volume and EDI 10 years result in doubling the volumes Potential, Australia 1992 quoted earlier. translation software, mapping and COSTS OF INSTALLING EDI testing. * Applications: as necessary, amending There are two distinct phases to acquiring applications to produce, say, a flat file EDI knowledge and experience: the pilot output. phase and the rollout or ramp up phase. * VAN costs: registration, monthly mini- The pilot phase involves the education and mums, volume costs, and special ser- training needed to set up a generally one- vices. on-one trading partner relationship. The * Software costs: translation package and effort for the pilot phase is not routine, maintenance. because it only happens once for each EDI user. The costs are distributed over the These one-time costs, if totally attributed following areas: to the pilot, fall between US$6,000 and US$9,000 per user. Mid-range and main- * Awareness: reading, researching, semi- frame costs can go as high as US$100,000 nars, visits. with averages in the US$25,000-35,000 * Education: management, project manage- range. It must be emphasized, however, ment, technical. that these are one-time costs, and for entry * Hardware: at entry level, PC and mo- into EDI for the first user. dem. The ramp up or rollout phase involves . Systems consultancy: installing the rolling out EDI to a wider community of Marketing / Awareness Advocacy / Selling cycle _ _ _ _ Implementation Systems development First stage /Pilot operation Maintenance / Administration Figure 5.2: Ramp Up Phase 22 Information Technology and National Trade Facilitation: Guide to Best Practice potential users. What follows is a model for development of relevant handout material. EDI advocacy and implementation that will It is reasonable to assume that this phase help to explain this phase of involvement. would require a total of 38 working (people) days and additional miscellaneous THE ADVOCACY AND expenses of about US$3,000 to cover statio- IMPLEMENTATION MODEL nery, audio-visual production, hotels and travel. This model, based on an actual case study, Advocacy, thie sellinig cyjcle. Once prospec- is designed for the EDI manager at a hub tive partners have attended a seminar and that has completed the pilot phase and is read the handouts, they are already on their now ready to add new users to its EDI way to discovering EDI. If they are suffi- community. It is a 20-partner ramp-up ciently interested, they will want to talk to effort, but makes no reference to who the EDI advocate at the seminar or soon actually carries out the work. Instead, the after. This is the first occasion when pro- emphasis is on the scale of the effort. spective electronic trading partners will The ramp up phase can be divided into meet face to face in the context of an EDI six stages (see Figure 5.2): relationship. The questions will be along the lines of "What do I do next?" * Marketing and awareness The post-seminar discussion will allow * Advocacy, the selling cycle the initiating agency to ascertain levels of * Implementation interest and collect data to help determine • System development what priorities (in terms of effort and * First stage, pilot operation for new users timing) should be allocated to each trading * Trading partner maintenance and admin- partner, and whether a trading partner is istration desirable as an EDI partner. The next step is a more formal data Marketing and awvareness-the discovery gathering process, which can be accelerated phase. This stage involves bringing EDI and by sending out a survey. Bearing in mind its benefits to the attention of selected the level of detail needed to progress at this trading partners, including some "reserve" stage (time, travel, and expense permit- or fall-back partners. A rule of thumb says ting), it is usually better to gather data that one out of two trading partners se- during more than one face-to face meeting. lected actually conforms to the program. This also assists the informal education Hence, for a planned community of 20 it is process for the prospective EDI partner. necessary to initially involve up to 50 A typical survey would include organi- potential partners. zation details, contacts (position, contact Recruitment methods will involve phone, fax, and so on), data processing personal contact, mailings, seminars, and so systems (hardware, operating systems, on. To keep costs to a minimum, it is communication devices), EDI capability, desirable to avoid individual face-to-face translation package or network used, meetings until this stage of the campaign document types, EDI standards, their levels has had a discernible effect, such as the and message sets, sample forms, volumes, conclusion of an introductory seminar. traffic analysis, and information flow The mailing effort should take 14 work- diagrams. At this stage it is also helpful to ing days (three targeted people or ad- have some demonstrations and quotations dresses per potential partner-say, 75 from one or more VANs. people), typically involving about a week's All this can take eight meetings, or four work to establish names, titles, addresses, days per trading partner for one person, fax and phone numbers, then set up the and travel and other expenses. For 20 mailing list, and actually complete the trading partners, the total effort would be mailing. around 80 working days, with miscella- Two people should work for 10 days to neous expenses of US$5,000. prepare for two seminars, with two addi- Implementation. At this stage the process tional days spent on the rehearsal and becomes a little more predictable. It in- delivery of each seminer. This includes the volves people with specific technical skills, Teclnlology and Cost Options 23 such as business analysts and program- System development. Effort at this stage mers. Assuming, for the sake of simplicity, varies with the trading partner. If, for that the trading partner already has appro- example, a flat file has to be mapped priate hardware and has opted for a front- between an application and the EDI transla- end PC approach, the options are a stand- tion process, that would take a few days of alone PC, a front-end PC sending and systems development time. Let's assume it receiving flat files to and from another would be a self-funding operation and the computer (or another program on the PC), work would be specific to an individual or a mainframe or mid-range computer. The trading partneris needs. It will still be implementation steps include: necessary to allow about three days per partner. System development for 20 part- * software product training in a classroom ners at three days each adds up to 60 days environment; for one person (funded), for an estimated * software installation and testing; cost of US$3,000 per installation. * document and standards mapping; Pilot operations. In this period the * network-mailbox testing; trading partner becomes self-reliant and * application training in a normal trading sufficiently competent to add new docu- environment; ments and install new versions of software * pilot commencement; without any external help. Since trading * trading partner or network or vendor partners will be joining the network at certification. roughly two- to three-week intervals, a consistent telephone support service needs Typically this would take a day in a to be available. Assuming each pilot calls classroom, with up to eight other trainees, the service once a week and takes one hour plus up to two days for a support techni- for the call and problem resolution, the cian on customer premises. For 20 trading support effort would require for an average partners, assuming two trainees per part- number of 10 partners, 50 weeks times one ner, this would mean six courses. Although hour per partner, and a total of 70 days for these activities would most likely be one support person. organized and run by VANs, the EDI Trading partner maintenance and adminis- advocate would have to spend some time at tration. Other tasks are involved here in each course. addition to technical support, advocacy and With total classroom training of eight education: for example, establishing a user days (this step also requires two days of group, communications, a newsletter, implementation effort per trading partner), perhaps a message development group; total implementation time would be 40 conversion to new document types; and days. introductions between trading partners. Table 5.2: Case Study, a 20 Partner Group Activity Days of effort Associated expenses($) Advocate Technical/ Support 1 Marketing awareness/ Discovery 38 - 3,000 2 Advocacy/ Selling cycle 80 - 5,000 3 Implementation 8 40 - 4 Systems development* - 60 5 Pilot operation 70 - 6 Maintenance/ Administration - 20 2,000 TOTAL 126 190 10,000 * A self-funding exercise 24 Information Technology and National Trade Facilitation: Guide to Best Practice Even for a 20-partner trading group, this can take four weeks for one Implementation Systems person in a year, plus printing, mail- 16% development ing, travel, and accommodation 19% expenses. Total time, therefore, would be 20 days for one person, with total Iii Pi expenses coming to US$2,000 for Advocacy / operation miscellaneous expenses. 25% -. 22% Table 5.2 illustrates the six stages of ' implementation for a 20-partner unit Marketing / Maintenance Awareness Administration when undertaken independent of 12% 6% vendors' efforts, while the proportions of effort involved in advocacy and imple- Figure 5.3: Proportions of Effort Required for mentation are represented in Figure 5.3. EDI Advocacy and Implementation Assuming a fully burdened professional person (that is, overhead plus salary, expenses, and so on) for both advocacy and COST SUMMARY FOR A technical support costs US$100,000 per year 20-PARTNER GROUPING (200 days per year), the people costs for a 20-partner effort can be estimated for 316 Anyone who is already in an EDI commu- days as US$158,000, plus expenses of nity, and expects to recruit more electronic US$10,000, that is, a total of US$168,000. trading partners, needs to recognize the This calculation ignores the resources and true costs of recruitment. Similarly, anyone costs for customer-specific work, such as intending to join the second stage of an EDI systems development. It may be reasonable community needs to know these costs exist, to assume a full cost recovery or revenue- in addition to the more widely known costs neutral approach for systems development. of the technology. Table 5.3, which summa- Accordingly, the people costs associated rizes this section, does not include hard- with developing the relationships with ware, software, or network-VAN costs. Nor trading partners to the point where they does it include the internal costs for an can actually use an EDI system can reason- installing organization, such as staff time ably be estimated at between US$6,000 and and expenses. US$9,000 per trading partner, or an average of US$8,400. These costs are in addition to ADVOCACY AND IMPLEMENTATION: hardware, software, network, and vendor THE LAST WORD training costs. Vendor costs, not considered in this section, may be categorized as the Someone has to foot the bill for these costs. cost of technology. The only costs described In the future, it will not be one of the EDI here are those of developing EDI trading vendors, but rather a different type of partner relationships. industry or EDI-specific cooperative-an Table 5.3: Cost Summaryfor a 20-Partner Grouping Labor Expense Total Average (days) (US$) (US$) (partner US$) Discovery 38 3,000 22,000 1,100 Advocacy 80 5,000 45,000 2,250 Implementation 48 - 24,000 1,200 Systems Development 60 - 30,000 3,000 Pilot Operations 70 - 35,000 1,750 Maintenance 20 2,000 12,000 600 Total 316 10,000 168,000 9,9000 Technology and Cost Options 25 essential move if communities are to grow. A PARTNER'S INTERNAL EDI COSTS Meanwhile, it is evident from the figures above that two people dedicated full-time The costs of rolling out a large community to the project can cope with a 20-partner are not just borne by the group or vendor implementation within a year. Certain who has undertaken the advocacy and activities, such as seminars, will need more implementation task. For every person people, such as those with experience from involved from the external advocacy and the first phase of the founding EDI partner implementation group, there could well be group. In practice, one person can deal with an equal number involved internally. only five or six attendees at any seminar; These internal people will come from the hence a seminar with 50 people attending various EDI-affected departments (purchas- would need an extra five or six volunteers, ing, accounting, and so on), and may well which adds to its costs. include an organization's complete EDI Similarly, two people can handle 20 committee. In addition to these staff and partners if the work comes in a steady flow. their associated costs, whether regarded as But the work never does come in a steady sunk costs or not, there are technology- flow. Holidays, sickness, training, and peak related costs, such as hardware, software, loads, all have to be accommodated, as and network costs. does a succession plan. Therefore, it may A recent survey of EDI implementation take up to three people, virtually full-time, in the Australian auto industry produced to service this effort. This would raise some surprising results. After a five-year actual costs to the range of US$12,000 to implementation, electronic messages now US$14,000 per new trading partner. represent 92 percent of the total value of The costs identified here are real costs purchases by the industry. The average cost which up until recently have been partly of implementing EDI was US$12,000 per subsidized by or paid for entirely by installation (range US$1,100-125,000); this vendors. Increasingly, however, the end includes hardware, software, integration, user will have to meet these costs, or they staff costs, and expense. must be absorbed by a major hub. Annual operational costs, including EDI Two further points need to be empha- service provider costs, network costs, and sized here. First, the costs described are associated staff costs average US$4,500 per concentrated on the costs of marketing to annum (range US$225-22,500). Although and implementing with individual trading these are Australian costs, and much of the partners. But the reality is that to sell and costs in the early days could have been implement EDI you often need to talk to a avoided if it were not for the pioneering trading partner's EDI committee-a group nature of this particular EDI community, of people. Each trading partner is likely to they illustrate a typical EDI community's send several people to meetings, training internal costs. session and the like. This adds to the costs. Second, although all the steps described THE DILEMMA for the large-scale ramp up model are indispensable, they can be streamlined to For a typical medium-sized EDI community some extent. Some of the information that (or a ramp-up increment of a larger EDI must be imparted to trading partners can community) the approximate costs, taking be delivered at seminars, so that multiple the large-to-medium in the ranges, are people hear the message at once. Further, shown in Table 5.4. each of the first 20 or 30 trading partners These are high figures for small- to mid- could be given responsibility for advocat- sized businesses, high enough to be barriers ing and implementing EDI with 5 to 10 new to entry for those businesses. Similarly partners every year. Over a period of three VANs do not have the resources, or the years this could increase the size of the EDI will, to continue subsidising the advocacy community to 1,000 participants, keeping and implementation tasks. At this rate the down the cost to the hub. necessary critical mass of users may never be reached. As it is, in many cases, EDI has added to business costs. 26 Information Technology and National Trade Facilitation: Guide to Best Practice Table 5.4: Internal and External Costs for technology to be usable by existing staff at EDI Ramp Up minimum effort. VANs want to increase data traffic Cost type One Time Per Annum without incurring substantial extra support (US$) (US$) costs. The local business community (specifi- External 5,200 - cally in the Asia Pacific region) would like to ensure that EDI and its alternatives Internal 9,000 3,375 embrace local business practices, culture and languages. It would like to avoid any TOTAL 14,200 3,375 forced adoption of new practices. The general business community, on the other hand, would want to achieve mass It is good to remember at this point that installation of electronic communications, there are other methods to reach the same using the existing infrastructure of technol- goals as EDI. These methods involve ogy and services. conventional technology (like fax), existing With these various objectives as a processes, and the existing business infra- starting point, it is possible to explore how structure. They can be cheap, easy to the existing infrastructure of technology install, and capable of being integrated into and services might be used to achieve the a full EDI scheme, as time, resources, and objectives of EDI today. cost permit. So far it has been shown that there are INTERMEDIARIES substantial external and internal costs associated with advocating and implement- The existing EDI infrastructure consists of ing EDI. As VANs and vendors will pay less mainframe, mid-range and personal com- and less of these costs in the future, it is puters, VANs and network services, trans- necessary to find innovative ways for lation software vendors and the various meeting or reducing these costs. education, training and consultancy organi- zations. This infrastructure is currently ALTERNATIVES TO EDI used by no more than 5 percent (and probably a lot less) of potential users. Given the expense of implementing tradi- For alternatives to EDI, we must look tional EDI today, users have ample incen- elsewhere. The goal is to find any existing tive to seek alternatives. These alternatives procedure or technology that will accom- must fulfill a variety of objectives, depend- plish the objectives of EDI, but with less ing on the perspective of the user. cost than the existing EDI infrastructure. The hub or large EDI user wishes to The types of business service providers that avoid having to support both electronic and even the smallest traders are accustomed to paper output and input. In other words, it dealing with on a regular basis include the wants (for appropriate applications) to treat postal service and couriers; banks, accoun- all output as EDI output and to receive all tants and attorneys; courts and service input, whatever its sources, as EDI input. bureaus; the local telephone company and The medium or small user who already other public utilities such as electricity, gas has some EDI-computer capability wishes and water; government departments; to have a number of different means of customers and suppliers; and so on. receiving and sending data, so as to satisfy Some of these service providers could each of its different trading partners. This perform an intermediary EDI service. They user wants to progressively install EDI could serve as walk-in service bureaus and while reducing costs. collection or distribution points for hard The least sophisticated user wants to use copy input and output, or even be one of existing installed technology to access EDI- the resources for "low tech-no tech" EDI. capable trading partners, without the expense and effort needed to implement a full EDI system. The same user wants the Technology and Cost Options 27 LOW-TECH INFRASTRUCTURE which scans input and translates it into ASCII-formatted EDI data. A low tech-no tech EDI service must cope * Wireless data technology, including radio with small to very small volumes of input frequency (RF) and cellular data technol- and output. This requires batch processing ogy, using cellular voice or cellular data techniques rather than real-time or interac- networks. tive processes. * Wireless personal computer technology. Intermediaries are just part of the * Customer input terminal, a specially existing infrastructure that could support engineered equipment, specifically built low-tech, no-tech EDI. Another part would for a single purpose. For example, it may be the "noncomputer technology" that is be a hand-held device with a screen, likely to exist today on the premises of keyboard and fax-data modem, capable small businesses. Many small businesses, of inputting, receiving and translating even in Asia, already have personal com- one or two document types. puters, but several of them refuse to use computers for EDI, using the machines for THE HYPOTHESIS specific jobs and nothing else. For EDI, the technology should be relatively current, There are some basic premises for the low electronically-based business machinery, tech-no tech model. connected to some form of communications No matter what the partners' circum- network, such as the telephone, telex, fax, stances, provided they have one of the and point-of-sale (POS) device (EPOS, ECR, suitable basic devices (phone, fax, telex, EFTPOS). As time goes by, the list would and so on) installed, they can be connected also include the cellular phone and other to the world of EDI. cellular data devices, such as portable The very large hubs, on the other hand, faxes. will be able to utilize their existing VAN infrastructure to send and receive data. In BROADER TECHNOLOGY INITIATIVES ideal circumstances, they need not connect electronically to any other service vendor. There are more sophisticated machines and A not too price-sensitive intermediary technology available, which would assist infrastructure in combination with the VAN intermediaries to provide an EDI service infrastructure can facilitate low tech-no between both noncomputerized trading tech EDI. For example, if a small enterprise partners and hub EDI users. Among them only sends one invoice a week, then US$5 would be to US$10 EDI charge for that invoice would not be considered too expensive. * Store-and-forward fax services, fax on This is not a hypothetical situation. demand, toll-free fax services and optical Several Asian territories are trying to make character recognition (OCR) faxes. EDI possible for nonautomated traders. * Telex technology, such as soft, or pro- Singapore has already made some progress. grammable, telex machines and telex-to- Singapore Network Services (SNS) and fax services. Singapore Customs helped to establish * Voice processing services, such as digital several EDI data entry-output bureaus for voice-in, data-out services. their import-export approval and quota * Point-of-sale technology, such as scan- control EDI system. Additionally, Singapore ners, electronic cash registers, EFTPOS Post now offers an EDI fax service for such and intelligent (programmable) POS messages as export approvals and acknowl- devices. edgments. Hong Kong, Taiwan (China), * OCR devices (scanners) and magnetically Malaysia and Korea will now follow suit encoded character recognition equip- within the next two or three years. ment. * Imaging technology, such as digital THE MODEL photography for document storage, archiving and editing. Two functions are needed to complete the * Intelligent character recognition (ICR), EDI loop. These are EDI-out and EDI-in. 28 Information Technology and National Trade Facilitation: Gutide to Best Practice EDI-out is where a hub can deliver mes- Because none of the messages are printed, sages (via an EDI service) to any trading the hub's trading partner enabling tables partner. The messages start in the format can be simplified. and medium of the hub's choice. EDI-in is The messages are transmitted as an where any partner can deliver EDI-format- aggregate file to the VAN. Messages with ted messages to a hub's mailbox, using any valid electronic trading partner addresses appropriate technology or intermediary for are delivered to the appropriate mail boxes. that purpose. For both EDI-out and EDI-in, Those without valid trading partner ad- the conversion of data between the hub and dresses, but with a default address, are the trading partner is transparent to the delivered to a dead letter mail box (DLMB). parties at either end of the transaction. The DLMB can be accessed by the VAN The EDI-out process is shown in Figure itself or by any of the intermediaries 5.4. The idea of EDI-out is that any EDI hub previously identified. can treat all of its trading partners as if they For example, a VAN may access this were EDI capable. With cooperation from DLMB, interpret the file of messages to a the VAN, this is not difficult. The VAN may standard print format and then fax them to have to do a little development and may their destination. This would mean that a have to charge a little extra. However, the fax number would need to be picked up VAN has incentive to cooperate because the from the profile tables and inserted into the EDI-out service will increase data traffic message. The VAN might bill the fax through the VAN. In addition, the hub recipients for processing and fax charges. should be interested in paying a little more Alternatively, a toll-free number could because the service will help the hub be used for a fax-on-demand service. eliminate systems that support both EDI Trading partners could dial into the service and paper. to receive their messages, and the costs All messages for any trading partner are would be billed to the hub. translated by the hub's data processing If an intermediary, such as a postal system from the application format into authority were to access a DLMB, it would EDI format and then delivered to the VAN. retrieve the file, sort it into print or