Diffusion of Electronic Public Procurement in Denmark
Ramanathan Somasundaram M.Sc., Aalborg University (2001)
Submitted to the faculty of Engineering and Science in partial fulfillment of the requirements for the degree of
Doctor of Philosophy at
Aalborg University, Denmark Copyright © 2004 Ramanathan Somasundaram. All rights reserved
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
To those who believed in me
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
2
Executive Summary
Electronic public procurement (e-PP) is being implemented in many countries worldwide. Preliminary analysis brings to attention many challenges that are likely to be encountered while implementing e-PP nation wide. For example, government organizations might face difficulty in authoritatively negotiating frame agreements with suppliers, multiple islands of e-PP computing infrastructure could emerge and suppliers might resist participating in e-procurement. There has been little research effort taken to document challenges encountered in implementing e-PP and to theorize their occurrence. As a remedy, this study is designed to answer the research questions; “what challenges are encountered in implementing e-PP? Why do these challenges arise?”
Explorative case study design is adopted for answering the research questions. The Danish nation is regarded an embedded unit of analysis. The implementation of multiple e-PP networks is studied. The term “diffusion” is used in the chapter title to denote that the spreading of e-PP in Denmark is studied. Primary data for the study is collected using semi-structured interviews. The interviews are primarily guided by IOS adoption literature. Thirty three stakeholders belonging to multiple stakeholder types are interviewed about their involvement in the diffusion of e-PP. The interview data is analyzed using explanation building technique.
Rich data collected from the field study is used for discussing challenges encountered in implementing e-PP. Debate format is used for discussing the challenges as it allows for the presentation of multiple perspectives. The debates were found happening in the three following areas; frame agreement negotiation at the organizational and interorganizational levels and computing infrastructure. Debates in the three areas are about centralization versus decentralization. The debates originate due to differences in perception about the extent to which government is single organization.
The role of policy maker with respect to the diffusion of e-PP is analyzed under six specific actions; knowledge building, knowledge deployment, subsidy, mobilization,
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
3
standard setting and innovation directive. A conjecture is derived for each analysis. The six conjectures that resulted are presented as policy recommendations. The analysis shows that as public administration is politically managed, government is mainly able to influence and not regulate both the supply and demand sides. A regulatory action can be misinterpreted as an effort to alter power structures within the public administration.
A new research approach for studying the diffusion of e-market has emerged from this study. In the approach, the viability of using critical mass as the dependent variable is explored. The term critical mass for a type of e-market is operationalized as follows; potential adopters’ expectation regarding e-market’s profitability is proposed as a measure for critical mass.
The thesis is prepared as a summary of five refereed publications. The summarized papers are referred to in the document.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
4
Preface
Waiting for camera print
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
5
Table of Contents 1
INTRODUCTION................................................................................................................... 9
1.1
OVERVIEW............................................................................................................................. 9
1.2
ROLE OF GOVERNMENT IN THE EMERGING WORLD ORDER ............................................ 9
1.3
INTRODUCTION TO ELECTRONIC GOVERNMENT ............................................................. 11
1.4
INTRODUCTION TO ELECTRONIC PUBLIC PROCUREMENT (E-PP) ................................... 15
1.5
PROBLEM EXPLANATION .................................................................................................... 21
1.6
ANSWERING THE RESEARCH QUESTIONS .......................................................................... 25
1.7
DELIMITATIONS OF THE STUDY ......................................................................................... 26
1.8
REPORT PERSPECTIVE ........................................................................................................ 27
2
THEORETICAL BACKGROUND..................................................................................... 31
2.1
OVERVIEW........................................................................................................................... 31
2.2
RELEVANCE OF THE IOS ADOPTION REVIEW PAPER (HICSS 2003) ............................... 32
2.3
REFERENCE AREAS FOR E-PP DIFFUSION RESEARCH ...................................................... 33
RELEVANCE OF E-GOVERNMENT RESEARCH: ................................................................................ 34 RELEVANCE OF STANDARD MAKING RESEARCH: .......................................................................... 35 RELEVANCE OF IT OUTSOURCING RESEARCH: .............................................................................. 37 RELEVANCE OF PUBLIC PROCUREMENT RESEARCH:...................................................................... 40 3
RESEARCH PARADIGM ................................................................................................... 42
3.1
OVERVIEW........................................................................................................................... 42
3.2
RELEVANCE OF THE RESEARCH PHILOSOPHY PAPER (ECIS 2003)................................. 43
3.3
THE NATION AS THE EMBEDDED UNIT OF ANALYSIS ........................................................ 45
3.4
RESEARCH APPROACH ........................................................................................................ 49
RESEARCH DESIGN:........................................................................................................................ 50 DATA COLLECTION: ....................................................................................................................... 50 DATA ANALYSIS: ........................................................................................................................... 51 3.5 4
VALIDITY OF THE STUDY .................................................................................................... 53 STAKEHOLDERS INVOLVED IN THE DIFFUSION OF E-PP ................................... 55
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
6
4.1
OVERVIEW........................................................................................................................... 55
4.2
THE DANISH PUBLIC SECTOR ............................................................................................. 55
4.3
NATIONAL PROCUREMENT AGENCY (SKI A/S) ................................................................ 56
4.4
BUYER (PROFESSIONAL)..................................................................................................... 57
4.5
BUYER (END USER).............................................................................................................. 58
4.6
POLICY MAKER ................................................................................................................... 59
4.7
TECHNOLOGY PROVIDER ................................................................................................... 60
GATETRADE:.................................................................................................................................. 60 KOMMUNE DATA (KMD):............................................................................................................. 61 KUBUS: .......................................................................................................................................... 62 4.8
SELLER (DECISION MAKER) ............................................................................................... 62
4.9
REGIONAL GOVERNMENT ASSOCIATION (ARF) ............................................................... 64
4.10
E-PROCUREMENT COORDINATOR IN THE PRIVATE SECTOR .......................................... 64
5
CHALLENGES IN IMPLEMENTING E-PP .................................................................... 65
5.1
OVERVIEW........................................................................................................................... 65
5.2
CENTRALIZATION VERSUS DECENTRALIZATION PAPER IN PERSPECTIVE (ECIS 2004). 65
THE TALE:...................................................................................................................................... 66 THE TWIST IN THE TALE:................................................................................................................ 67 6
POLICY RECOMMENDATIONS FOR THE DIFFUSION OF E-PP ........................... 69
6.1
OVERVIEW........................................................................................................................... 69
6.2
SUMMARY OF THE POLICY RECOMMENDATION PAPER (EJEG 2005)............................. 69
7
REFLECTIONS ON RESEARCHING THE DIFFUSION OF E-PP.............................. 71
7.1
OVERVIEW........................................................................................................................... 71
7.2
DIFFICULTIES IN RESEARCHING THE DIFFUSION OF E-PP .............................................. 72
DISENTANGLING E-PP: .................................................................................................................. 72 IS RESEARCHER SHERLOCK HOLMES?........................................................................................... 73 RESEARCH ETHICS: ........................................................................................................................ 74 7.3 8
CRITICAL MASS AS THE DEPENDENT VARIABLE! (BLED 2004)....................................... 74 THESIS AS A SUMMARY OF REFEREED PUBLICATIONS ..................................... 77
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
7
8.1
OVERVIEW........................................................................................................................... 77
8.2
SUMMARIZED PAPERS AS A COHERENT UNIT .................................................................... 77
RESEARCH PREPARATION PHASE:.................................................................................................. 78 DATA ANALYSIS PHASE: ................................................................................................................ 78 RESEARCH DIRECTION PHASE:....................................................................................................... 79 9
REFERENCED LITERATURES........................................................................................ 80
10
LIST OF FIGURES ............................................................................................................ 89
11
LIST OF TABLES .............................................................................................................. 89
12
APPENDIX .......................................................................................................................... 90
12.1
RESEARCH PUBLICATION ONE – THE HICSS 2003 ......................................................... 90
12.2
RESEARCH PUBLICATION TWO – THE ECIS 2003........................................................... 90
12.3
RESEARCH PUBLICATION THREE – THE ECIS 2004 ....................................................... 90
12.4
RESEARCH PUBLICATION FOUR – THE ECEG 2004 AND EJEG 2005............................ 90
12.5
RESEARCH PUBLICATION FIVE – THE BLED 2004.......................................................... 90
12.6
IFIP 9.4 NEWSLETTER ...................................................................................................... 90
12.7
INTERVIEW DATA .............................................................................................................. 90
12.8
INTERVIEW QUESTIONNAIRE ........................................................................................... 90
12.9
CRITICAL MASS GAME ...................................................................................................... 90
12.10
A SAMPLE TRANSCRIBED INTERVIEW (ONLY PROVIDED TO EXAMINERS) .................. 90
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
8
1
1.1
INTRODUCTION
Overview
Several nations world wide have embarked on transforming government procurement using internet technologies. A study three year long was carried out to analyze challenges faced by the Danish nation in transforming government procurement using internet technologies. Details about the study and results obtained from the study were documented during the study period. Some of the documentation was peer reviewed and published. Five of the publications are hereby synthesized to provide an overview of the study and the results obtained.
1.2
Role of Government in the Emerging World Order
Government Procurement Agreement (GPA) is a hotly discussed topic in World Trade Organization (WTO) meetings. A nation that is party to GPA is expected to effectively open up its public procurement business to international suppliers. Despite the negotiations having been initiated in 1981 in the Tokyo WTO meet, GPA remains a plurilateral agreement as of 2004. Plurilateral here implies that only 28 of the 147 WTO member countries have so far signed the agreement. 15 European Union member states, US, Canada, Japan, Israel, South Korea, Norway, Switzerland and Singapore are a few countries that are party to the agreement [www.wto.org].
The European Union prepared a green paper on public procurement in the year 1996. Chapter six in the green paper discusses the importance of opening up public procurement world wide. The following is a quote from the chapter, page 45 (EU 1996);
“With a new World Trade Government Procurement Agreement (GPA) in place, globalization of public procurement is certain to grow. Our industry must reply positively and effectively to this challenge; world-wide competition will be intensifying and success will hinge on innovation and international thinking to meet the challenge. Further market
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
9
opening must be actively pursued; a constructive, ongoing dialogue with our industry is critical to identify new market opportunities and to help set objectives for any future negotiations”
Singapore has a perspective similar to that of the European Union on GPA (Jones 2002). One can assume that other GPA signatories share perspectives similar to the above.
India has so far led the resistance towards GPA. Developing nations such as India and under developed nations have in general opposed free trade. The opposition is mainly for the sake of protecting domestic suppliers from international competition. Two more reasons that cause opposition are;
i)
government machinery underdeveloped to facilitate international trade and
ii)
domestic suppliers poorly prepared to take advantage of opportunities world wide
Nations however have reasons to look at GPA negotiations in positive manner. First and the foremost, public procurement when made transparent helps in reducing corruption. Corruption is a menace in most developing and under developed nations. India for instance has the dubious distinction of being one of the 10 most corrupt countries in 2001. Second, GPA negotiations bring into focus the need for overhauling government machinery to facilitate international trade and for preparing domestic suppliers to take advantage of opportunities world wide. The secretary general of the CUTS centre for international trade, economics and environment, India makes the following statement upon returning back from Doha WTO meet (Mehta 2001);
“Such (GPA) problems in ensuring implementation by the purchasing entities of any accord that may be adopted are likely to be all the greater in a country like India, where the systems for co-ordination among ministers and other government agencies do not always work effectively. However, that is no excuse to not engage in any negotiations that would take place at the WTO”
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
10
Please notice that India shows inclination towards becoming a party in GPA. The resistance it currently exhibits is mainly due to the lack of preparedness. Given the inclination, it is reasonable to assume that India when better prepared will become party to GPA. One can expect several other nations that have been resisting GPA under India’s leadership to follow suit. Thus, GPA would sooner or later become multi-lateral.
With bilateral, regional, plurilateral and multilateral agreements negotiated world wide, governments are forced to work on their competitiveness. The public sector does not enjoy the monopoly status any more in this increasingly open and integrated world. Nations world wide are competing against one another. Ireland, Singapore, Canada and Australia, to mention the prominent, are actively reforming public sector to take advantage of the emerging world order. This forms the context in which this doctoral thesis is prepared.
1.3
Introduction to Electronic Government
A clarification is in order regarding the use of terms “government” and “public sector”. Please note that both the terms are used to denote each entity in administrative machinery that governs a nation. If only a part of government such as local, regional or state government is referred to, then such is made explicit. If there is difference between government and public sector, the author is not aware of such, which he hereby acknowledges.
Governments are in search of means to enhance their competitiveness. In this respect, efforts taken by the private sector for enhancing its competitiveness are of relevance. The private sector within the last decade has relied heavily upon internet technologies for enhancing competitiveness. The U.S. federal government under Al Gore’s leadership conceived using information technology (IT) to improve the functioning of government as early as 1993 (Aldrich et al. 2002). The take up of IT in the public sector however has lagged behind that of the private sector. In the last couple of years, governments world
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
11
wide are concentrating their efforts in enhancing their administrative efficiency using IT (Koh and Prybutok 2003).
What the term “electronic government” means is of considerable debate. Moon (2002) mentions that e-government despite being an important development in the field of public administration “has not been clearly defined and understood”. The term “digital government” is used frequently as an alternative for e-government. Several papers on egovernment have a definition section in where one’s understanding of the term is provided (see for example; Tyndale (2002) and Chandler and Emanuels (2002)). Silcock (2001) has written an entire paper explaining the term. There are several definitions of egovernment. It seems there are as many definitions of e-government as the number of people working with the topic. It is for sure a worthy pursuit to synthesize and provide a detailed definition of e-government. Such however is not being done here. This section has as its objective positioning the study carried out to research the diffusion of electronic public procurement within the e-government research field.
It is important to clarify what “electronic” means in the term “electronic government”. The U.S. government used UNIVAC (Universal Automatic Computer) to process data electronically as early as 1952 (Univac). Since then, the use of computers in the public sector has been quite common. Given that e-government has gained prominence in the last couple of years, electronic in e-government could not possibly refer to all forms of electronic data processing in the public sector. Electronic in e-government instead is associated with the use of Internet technologies. Most definitions of e-government acknowledge such association (see for example Layne and Lee (2001), Marche and McNiven (2003) and Reddick (2004)).
Government in the term “electronic government” is a multi-level organization. If we take the case of Denmark, there is local government (kommune), regional government (amt) and the national government (state). At the continental level, Denmark is one of many member states in the European Union. Government is organized in a federal-state pattern, in countries such as Australia, India and the United States. Multi-lateral bodies such as
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
12
the World Bank, American Development Bank and the United Nations and management consulting firms study the growth of e-government internationally. E-government is researched at the local (Ferguson and Baron 2002), national (Li 2003), continental (Strejcek and Theil 2003) and international levels (Ronaghan 2002). In figure one government is presented as a multi-level organization.
There is much hype associated with e-government International
in the last couple of years. Governments world wide have embarked on e-government in one form or the
Continental
other. Multi-lateral bodies such as the World Bank, Asian Development Bank and the United Nations
National
have been funding those countries that cannot afford to invest in developing e-government. Given
Regional
the realm of possibilities that e-government opens up, one can say that the hype is not necessarily
Local
misplaced. Internet, the very infrastructure on which e-government happens, is ubiquitous in
Figure 1 – Government as a MultiLevel Organization
developed nations and is gaining widespreadness world over. According to Danmark Statistik, 64%
of the Danish population used the Internet in 2002 [www.dst.dk]. Governments now have the opportunity to interact with citizens (G2C), businesses (G2B) and among its own entities (G2G) digitally. Governments would benefit enormously when they transform towards performing G2C, G2B and G2G activities electronically (Accenture (2003), Sharma and Gupta (2002) and Chandler and Emanuels (2002)
There is e-governance and there is e-government. The two concepts point to different aspects of the relationship between citizens and their political structures. The following is a quote from Marche and McNiven (2003) “Governance stresses the ways in which decisions are made, while government stresses the way in which these decisions are carried out”. The process undertaken by public authorities in deciding whether or not to offer a service falls under governance. The election of political authority by the public is
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
13
a governance issue. Government on the other hand denotes the efforts taken by public authorities for delivering a service. The impact of internet technologies in deciding whether or not to offer a service and in delivering a service is discussed under egovernment and e-governance respectively.
The adoption of e-government is not a binary yes or no decision. It instead is a multistage process. Layne and Lee (2001) conceptualize the development of e-government falling under four stages. They are i) cataloguing ii) transaction iii) vertical integration and iv) horizontal integration. In the cataloguing stage, government establishes online presence and makes available downloadable forms. Citizens in the transaction stage are able to fill in forms, interact and perform financial transactions with government online. Governmental organization is automated in the last two stages. Vertical integration involves automation of a function such as tax processing across multiple governmental levels; local, region and nation. Government when horizontally integrated is able to share data across functions seamlessly. A few others have conceptualized the development of e-government similar to that of Layne and Lee (2001). Ebrahim and Irani (2003) make a comparative study of e-government adoption stage models.
Cataloguing and transaction, the first two stages of the four stage adoption model are oriented towards citizens (citizen centric). The vertical and horizontal integration stages are focused on reforming the structural characteristics of government organization and the civil servant (organization centric). Marche and McNiven (2003) propose a two dimensional framework for considering the impact of the internet in government. In one axis, they contrast e-government versus e-governance and in the other citizen centric versus organization centric.
E-public procurement (e-PP), the subject analyzed in this document, is hereby positioned within the e-government research area; procurement activity is carried out by governmental organizations across functions at all levels. e-PP implementation requires government to automate procurement activities both vertically and horizontally. e-PP is implemented mainly to enhance efficiency in operations. Hence, it is identified as an
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
14
organization centric activity. The public sector procures a large variety of goods and services from businesses (G2B) to carry out its day to day activities. e-PP is identified as an activity critical for the functioning of (e)-government. e-PP is positioned within the egovernment research area in figure two.
Impact of Internet on Government
E-government
Citizen centric
Government to Citizen (G2C) interactions
E-governance
Example: E-democracy
Organization centric
Electronic Public Procurement
Research focus
Figure 2 – Impact of the Internet on Government (Marche and McNiven 2003 Adapted)
1.4
Introduction to electronic public procurement (e-PP)
Procurement “is the process of obtaining material and services and managing their inflow into organizations” (Gebauer 1998). Typically between 20 and 60% of an organization’s transaction volume is spent on procurement (Neef 2001). The EU is estimated to have procured for 1000 billion Euros in 1998 (EU 2000).
The definition of electronic procurement (e-procurement) varies depending on the interpretation of the terms “electronic” and “procurement”. Davila et al. (2003) define eprocurement as “… any technology designed to facilitate the acquisition of goods and services…”. The term electronic in the definition can be understood in the broadest sense; telephone, fax, electronic data interchange (EDI) and the Internet are all electronic. Fax
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
15
and EDI have been used for procurement activities for a few decades. It is the use of internet technologies that is relatively new. Boer et al. (2002) acknowledge such in their definition, which is “…EP can be defined as using internet technology in the purchasing process”. Boer et al. (2002)’s notion of e-procurement is used in this document.
Macmanus (2002) brings to attention the difference between purchasing and procurement. Purchasing has narrower focus than that of procurement. It is defined in the dictionary of purchasing terms as “the act and the function of responsibility for the acquisition of equipments, materials, supplies and services. [Purchasing] describes determining the need, selecting the supplier, arriving at a fair and responsible price and terms, preparing the contract or purchasing order, and following up to ensure timely delivery” (NIGP 1996). Procurement is defined in the dictionary of purchasing terms as “the combined functions of purchasing, inventory control, traffic and transportation, receiving and inspection, storekeeping and salvage and disposal operations” (NIGP 1996).
The term e-procurement is often used to refer to the automation of activities that fall under the purchasing definition. A few pioneer organizations such as Novo Nordisk [www.novonordisk.com] and Covisint [www.covisint.dk] partners are automating wider range of activities true to the procurement definition. The public sector in general has fallen behind the private sector in the implementation of e-procurement (Krysiak et al. 2003). E-procurement in most public sector organizations automate basic purchasing functions such as supplier identification, catalogue management, order placement, invoice reception and payments. Pioneer public sector organizations such as Singapore [www.gebiz.gov.sg] and Finland [www.hansel.fi] are implementing advanced functions such as post contract management and logistics management (Shin and Park (2004) and the European Union newsletter on e-Procurement (IDA)
To the author’s knowledge, there is not yet a book or a research report that reviews the state of e-PP worldwide. An effort however is underway to document e-PP efforts undertaken worldwide. Multilateral development banks (MDB), the World Bank [www.worldbank.org], the Asian Development Bank [www.adb.org] and the Inter-
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
16
American Development Bank [www.iadb.org] are organizing this effort (for further details [www.mdb-egp.org]). In the last couple of years, governments advertise their efforts in implementing e-PP in the World Wide Web (WWW). These advertisements are a good source of knowledge. A review of such advertisements shows e-PP as a global phenomenon. Developed nations seem to have embarked on e-PP in one form or the other. As well, developing nations such as Brazil, Romania, Argentina and Thailand [www.mdb-egp.org] have gotten on the e-PP bandwagon as well.
What is driving the public sector towards implementing e-PP? First, public sector procures for a significant sum of money each year. For example, the Irish public sector procures goods and services for app. EUR 10 billion each year (Clark 2003). If the Irish state saves a conservative 10% through e-procurement, then the saving in real terms is significant EUR 1 billion each year. The state can redeploy the saved amount towards other development activities. Second, the public sector by implementing e-procurement prepares supplier base for e-commerce. The supplier base is expected to take advantage of its preparation and engage actively in e-commerce. It is thus the public sector can catalyze the diffusion of e-commerce in society (Heneriksen et al. 2004) and (Tonkin 2003). Third, e-procurement when implemented is expected to enhance transparency and better define accountability; two characteristics that are central for controlling corruption (Thai and Grimm 2000).
A central objective underlying the implementation of e-procurement is to minimize maverick buying. The public sector is notoriously known for its excessive maverick buying. An organization engaged in maverick buying looses out on volume discounts and lacks data on procurement expenditures. E-procurement helps organization collect the much required data on their procurement activities. Professional buyers equipped with such data are able to negotiate better frame agreements with suppliers. These frame agreements when utilized help organizations save significantly. The process adopted for attaining such savings is commonly termed “strategic sourcing”. E-procurement helps organizations significantly reduce administrative costs associated with processing a
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
17
purchasing request. Cisco claims to have reduced administrative costs from $130 to $30 per order (Ulstrup 2001). The public sector could arguably achieve similar savings.
Procurement in organization as large as the public sector is a complex activity. It involves supplier identification, contract negotiation and order processing. Procurement practices vary widely among public sector organizations. In one scenario, there is a professional buyer identifying suppliers and negotiating frame agreements/contracts. An end user who has need for a product or a service channels his or her requirement through professional buyer. In a different scenario, an end user identifies supplier, negotiates contract and places an order with supplier by himself or herself (McCue and Pitzer 2000). This scenario is commonly termed as “maverick buying”. The two scenarios described above need not necessarily be an either-or option. It is quite common among public sector organizations to adopt both the scenarios. An end user can choose to procure a set of products and services through professional buyer and a different set of products and services by himself or herself. Anderson et al. (2003) identify the lack of “professional shopper” notion in public sector organizations as an epistemological challenge.
E-procurement is a networked activity. It requires participation from both buyers and sellers. The value of a network increases as more of both buyers and suppliers participate (i.e.) buyer can reach out to a larger number of sellers and the vice versa. When an eprocurement network attains critical mass, buyers and sellers start flocking and so the network self sustains. However, risk enduring effort is required for a network to attain critical mass. Government can afford to take risks. It’s participation in e-procurement would catalyze the diffusion of e-commerce in society. For example, suppliers interested in government orders would make preparation required for e-procurement. Once prepared, they are more willing to electronically trade with private sector companies; thereby exploit their preparation.
The public sector can choose to electronically perform a step or two in the procurement procedure. E-procurement does not require the automation of the full procedure. For example, public sector organizations in Europe are required to advertise contract
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
18
information in the Official Journal of the European Commission [www.ojec.com] when the contract value exceeds a certain sum. The EU collaborates with national institutions to electronically advertise contract information; e.g. Business Information Publications Ltd [www.bipsolutions.com] in the United Kingdom, Journal Officiel [www.journalofficiel.gouv] in France and the public sector procurement portal [www.etenders.gov.ie] in Ireland.
Commodity goods such as computers, stationery materials and air tickets are easier to procure electronically than complex products such as construction contracts. Governments worldwide are therefore focused on electronically procuring commodity goods. A study conducted by the Danish government finds 35% of its purchasing suitable for e-procurement. A list of State govt.
Regional govt.
Local govt.
Travel and Hotels Office supplies Books Office equipment IT Furniture Provisions
Medicine Hospital supplies Office supplies
Office supplies Provisions Books Material for education Ironmongery Fuel
goods identified in the study as
suitable
for
e-
procurement is presented in table one (E-jump 2000). A study conducted to review government procurement in England recommends; “E-
Table 1 – List of Commodities Suitable for e-Procurement; Heneriksen et al. (2004) translation Used
procurement commerce
and
e-
systems
are
particularly suitable to controlling spending on low value and routine commodities. In the short to medium term, greatest gains may come from reducing the transaction costs of buying these items, which a number of sources suggest make up high proportion of the total costs.” Byatt (2001), paragraph 6.2. The Korean and the Singaporean governments are preparing to do most of their procurement electronically Shin and Park (2004) and Zulfiqar et al. (2001).
To the author’s knowledge, a systematic study has not yet been done to analyze the approach adopted by governments for implementing e-procurement. Anecdotal data gathered on the topic suggests that there are different approaches. A 2x2 framework
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
19
whose dimensions are national standard (yes or no) and private sector involvement (yes or no) helps in conceptualizing the differences. The cases of Singapore, Australia, the state of North Carolina and Denmark are briefly explained here to highlight the differences. Singapore is in the process of implementing its GeBIZ initiative as the national standard for e-PP. In Australia, there is not yet a national e-PP standard. State governments in Australia have developed islands of e-PP computing infrastructure. The state of North Carolina in the USA has entered into a Private-Public-Partnership (PPP) with Accenture consulting for developing and hosting the NC E-Procurement @ Your Service initiative [www.ncgov.com]. The agreement is such that the state will pay
No
De jure Australia Germany Invalid state De facto Singapore Korea
Yes
PrivateSectorInvolvement
1.75% of the cost of goods and services procured over the e-PP infrastructure as
De jure Denmark Norway
USA Spain
De facto Yes
No
National Standard
Figure 3 – A Framework for Classifying e-PP Implementation Approaches
marketing costs. Should the PPP generate revenues higher or lower than the agreed level then the state would either contribute half the deficit or earn half the surplus. The Danish government has entered into a PPP for developing and hosting e-PP infrastructure with Gatetrade. In the Danish case, the state only pays transaction fee and does not benefit from Gatetrade's success and neither does it contribute towards its failure. The Singaporean public sector on the other hand has developed e-PP infrastructure rather successfully on its own.
The national standard dimension in the matrix can be divided further into de jure and de facto. Usually, one or more institutions in a government recommend the use of an e-PP
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
20
infrastructure as the national standard. When most or all government organizations heed to the recommendation, then the national standard is de facto. If otherwise, it is de jure. Due to political reasons, de jure standard could fail to become de facto. It is important to realize that national de facto e-PP standard emerges into existence. The time taken for the emergence of de facto national e-PP standard could significantly vary from one nation to another. Figure three graphically depicts the framework just explained.
1.5
Problem explanation
e-PP is a new phenomenon. Governments worldwide are actively implementing e-PP only in the last couple of years or so. Consequently, much of e-PP documentation is quite new. e-PP developments have so far been documented mainly through;
i)
presentations by government officials
ii)
implementation strategy reports
iii)
case studies that describe best practices
iv)
reports in news media
The main source of such documentation is the Internet. A google search for e-PP and various versions of it results in over 100 hits as in July 2004. By reviewing the search results one for sure gets an idea about e-PP developments worldwide. e-PP documentation available in the Internet is usually one of the following types;
i)
advertisement
ii)
prescriptive and
iii)
anecdotal
The problems associated with such documentation are; first, governments tend to advertise their successes and not their failures. While relying upon such documentation, one runs the risk of not knowing the true state of happenings. Second, prescriptions do not get implemented more often than not. An enquiry is warranted to explain challenges
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
21
faced while implementing prescriptions. Third, news media brings to attention both successes and failures. However, explanations provided in news media are anecdotal in nature.
Based on anecdotal data and scientific knowledge on the implementation of IOS, one can hypothesize that the implementation of e-PP would be quite challenging. Here is a sample of challenges likely to be encountered while implementing e-PP; first, traditionally suppliers have resisted joining the Electronic Data Interchange (EDI) network. e-PP just as EDI is an inter-organizational system. To engage in e-procurement, suppliers have to invest resources for preparing and hosting electronic catalogues. The question therefore arises; would suppliers be willing to invest and join in e-PP? Second, government aims to centrally negotiate frame agreements in order to get volume discounts. The negotiated agreements would be posted in e-PP infrastructure. End users are able to utilize the agreements quite easily by logging into the e-PP infrastructure from their desktop computer. The challenge associated with such then begets the next question; would buyers be willing to let go of their autonomy and utilize centrally negotiated agreements? Third, government has traditionally been supportive of small and medium sized enterprises (SME). A move towards centrally negotiating frame agreements denotes a shift in policy because SME’s lack competitiveness as against their large sized counterparts while competing for government business. How then would local governments react to such shift?
To learn about the above mentioned challenges, there is need for rigorous theory based research on e-PP. Having realized such need, a few in academia have embarked on e-PP research. The few build upon scientific knowledge accumulated in research disciplines such as Inter Organizational Systems (IOS) adoption, e-marketplaces, e-procurement, public procurement and public administration. However, only a few empirical studies have been done so far to research e-PP. Findings from those studies have for sure been enlightening. So far, our understanding of the e-PP subject remains limited. Hence, there is need for in-depth study on e-PP.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
22
Some countries have taken a lead over the others in implementing e-PP. Countries such as Australia, Singapore, the United States, Korea and Denmark lead the pack. The research community would do well to document and explain the experiences of those leading the pack. The nations that later adopt in implementing e-PP would benefit from such documentation.
There are several perspectives from which e-PP can be researched. For example, Tonkin (2003) did an international study to “unravel” reasons underlying the uptake of eprocurement in the public sector. Vaidya and Callender (2004) are in the process of defining critical failure factors for e-PP. Heneriksen et al. (2004) bring into focus the importance of taking into consideration both the economic and political rationality while researching e-PP adoption. Maniatopoulous (2004) analyses the effect of socio-technical mechanisms in the adoption of e-PP. The point here is that e-PP is a new research area and the subject at this stage is understood in a limited way. Hence, while researching ePP, it is important to adopt a perspective that allows for exploration.
One researching e-PP has to decide on the unit of analysis (Yin 1989). The decision is very much so dependent on the perspective adopted for researching e-PP. Tonkin (2003) enquired policy makers in Europe and Australia. Maniatopoulous (2004) enquires eprocurement managers and e-procurement system developers at the local governmental level. Vaidya and Callender (2004) are planning to enquire both buyer and supplier sides. An explorative study is positioned well when multiple stakeholders involved in e-PP are enquired. The challenge however lies in defining the sample to be enquired.
e-PP is identified here as a research topic within the information systems (IS) area. Keen at the first International Conference on Information Systems (1980) thought of the IS as an “applied” field. It looks to established reference disciplines such as economics and computer science “…to get an idea of what good MIS research would look like…”. e-PP is an applied research area. Knowledge accumulated in reference disciplines such as public administration and public procurement is relevant for researching e-PP. The IS field has established its own research tradition in the last two decades or so (Baskerville
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
23
and Myers 2002). Several topics relevant to e-PP (e.g. IOS adoption, e-marketplaces and IT outsourcing) were actively researched during this period. It is important that e-PP research builds upon accumulated knowledge.
The terms adoption, implementation and diffusion are used through out this document. Distinct meaning that characterizes each of the three terms is hereby explained; the term “adoption” is used from the perspective of an individual or an organization. An organization is said to have adopted an innovation when it has agreed/decided to use an innovation. Organization could adopt an innovation and refrain from using it. The term “implementation” is used from the perspective of those who advocate the use of an innovation. The process of advocating the use of an innovation is studied under implementation. The term “diffusion” has a geographic focus. Diffusion denotes that the spreading of e-PP is studied. Diffusion of an innovation in a population is measured by the extent of use.
By taking the above mentioned into consideration, the following research questions are formulated;
Research question one: which stakeholders are involved in the diffusion of e-PP? What is the nature of their involvement?
This question is answered descriptively. The answer provides rich insights on stakeholders involved in the diffusion of e-PP and the role that they play. Diffusion is chosen as the research perspective for two reasons; first, diffusion requires not just the development of e-PP systems but also their uptake. So, a broad range of issues can be discussed under diffusion. Diffusion provides the flexibility much required for exploratory research. Second, e-PP is regarded here as an innovation. Diffusion of innovations has been researched for over a hundred years. French sociologist Gabriel Tarde plotted the Sigmoid diffusion curve in 1903 (Rogers 1976). IS researchers have used the diffusion of innovation work for studying the adoption of EDI and other types of Inter Organizational Systems (IOS) (e.g. (Premkumar and Ramamurthy 1994)). IOS
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
24
adoption science has been actively researched for over two decades (Chwelos et al. 2001). e-PP just as EDI is an IOS. e-PP research could well be built upon theoretical knowledge accumulated in the diffusion of innovation and IOS adoption areas. Chapter four in this document is dedicated for answering research question one.
Research question two: what challenges are encountered in implementing e-PP? Why do these challenges arise?
An analytical response is provided to the second question. An element of unknown is presumed in the first of the two sub questions. Field study data gathered for the study is analyzed to identify challenges encountered in implementing e-PP. A challenge arises due to some form of disagreement. The source of the disagreement is reflected upon to answer the second sub question. Chapter five is dedicated for answering research question two.
Research question three: what role should policy maker play with respect to the diffusion of e-PP?
A prescriptive answer is provided to the third question. Insights gotten from answering the second research questions are used to make policy recommendations. Preliminary analysis brings to attention few adverse impacts of e-PP. Hence, policy makers should cautiously support the diffusion of e-PP. As policy makers are part of the government, they may experience difficulties in reforming public procurement. The third research question is answered in chapter six.
1.6
Answering the research questions
e-PP is a multi-level activity. The hierarchy from bottom to top can assume the following form; end user, organization, nation, region and international. The diffusion of e-PP can be researched at any of the five levels. A researcher can limit himself to issues specific to a level or do a multi-level study. The Danish nation is defined as the embedded unit of
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
25
analysis in this study. The term embedded here includes stakeholders across nation, organization and end user levels. Towards the end of his study, the author relied upon secondary data to study e-PP activities at the European Union (regional) and international levels. This research is hereby positioned as a multi-level study.
An explorative case study was done to answer the above mentioned research questions (Yin 1989). As part of the case study work, stakeholders involved in the diffusion of e-PP in Denmark were identified using snow-balling technique (Moriarty and Bateson 1982). Thirty four stakeholders in all were enquired using semi-structured interviews. The stakeholders interviewed were of the following types; e-PP technology provider, e-PP policy maker, public procurement officer (national, organizational and end user levels), seller and e-procurement coordinator in private sector organizations. All the types except for the last were enquired about their role in the diffusion of e-PP. A couple of eprocurement coordinators in private sector organizations were enquired about challenges faced during implementation. The line of questioning was guided mainly by theoretical constructs in the IOS adoption science. The author relied upon secondary data (e.g. newspapers and websites) as preparation for each of the interviews. All those interviewed were encouraged to share their insights on the research subject. When a new line of reasoning emerged during the interviews, the author willfully enquired further and tried to learn about it. The author’s understanding of e-PP evolved as he did the interviews and associated secondary research. It is this understanding he relies upon to answer the above mentioned research questions.
1.7
Delimitations of the study
The study is delimited along two dimensions; i) research perspective and ii) unit of analysis. In the first, this study is guided primarily by IOS adoption literature. Many other research areas such as public administration, public procurement and IT outsourcing could have been relied upon to guide the study. In the second, nation is regarded an embedded unit of analysis in this study. Diffusion of e-PP could have been studied for example at an organizational level (e.g. in local government as Maniatopoulous (2004)
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
26
does) or at a dyadic level (like Vaidya and Callender (2004) propose). When the unit of analysis differs, the nature of insights generated would likely differ.
1.8
Report perspective
Reporting is one of the most complex phases in doing doctoral research. The final thesis report has to include varied information such as problem formulation, literature review, philosophical foundation, research methodology, data (case) description and analysis, discussion and future research directions. It takes significant effort to prepare a report including all of the above. Smart investigators Yin (1989) explains “…will begin to compose the case-study report even before data collection and analysis have been completed.” page 127.
Traditionally, doctoral researchers have presented their thesis work in monograph style thesis document. The document includes all details relevant to the thesis work. A different system for preparing the thesis report has been practiced in Aalborg University for just over a decade. As per this system, doctoral researchers publish their findings during the course of their research in double-blind reviewed conferences and journals. The researcher can choose to prepare the final thesis report as summary, provided he or she has a handful of publications in recognized outlets.
The author has chosen to prepare summary style thesis for the following reasons; first, one working in research profession has to regularly publish in conferences and journals. Doctoral study prepares one for a career in research. Summary style thesis requires one to publish in blind reviewed venues. Trying to publish in such venues is good preparation for a career in research. Second, each phase (e.g. literature review, research design and data analysis) in doctoral study requires extensive research. Smart investigators, as Yin (1989) explains, get their findings written down at the end of each phase. Summary style thesis encourages researchers not just to document their work at each phase but also to publish their documentation. Third, a monograph is usually about 250 pages long. A summary style thesis is much shorter. The author from a review of six summary style
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
27
theses learned that the size of a summary varies from 25 to 57 pages (Arent 2000; Karlsbjerg 2004; Kjeldskov 2003; Scheepers 1999; Sørensen 1993; Tuunainen 1999).The summary has two main purposes;
i)
to explain research objectives of the doctoral study and
ii)
to put in perspective contributions made by the summarized papers for addressing the research objectives
The four or more papers summarized contain most details about the doctoral study. The summary provides an overview of the thesis work. A reader can choose to read through one or more of the reviewed papers should he or she desire. Hence, a summary style thesis allows for flexible reading than the lengthier monograph.
This report summarizes four refereed publications. Table two has further information about the papers; title, publication venue, published period and contribution. The author documented his research efforts regularly and published them in blind reviewed venues. This is apparent from the order in which publications are made.
Quite early in his thesis work, the author identified IOS adoption work as the scientific basis for his research. He reviewed 73 important papers in the IOS adoption field. He wrote a paper along with associate professor Jeremy Rose explaining the results of his review work. This paper is published in Hawaii International Conference on Systems Sciences (HICSS) 2003. During the next phase, the author reviewed research philosophy literature to prepare the research design. At this time, he analyzed philosophical foundations underlying IOS adoption research. He co-authored a paper along with Jan Karlsbjerg, a fellow Ph.D. student explaining the results. The paper is published in the European Conference on Information Systems (ECIS) 2003. Then, the author embarked on data collection. He interviewed 34 stakeholders in Denmark within a five month period. He documented his data collection and data analysis efforts in two papers. The first paper is published at the ECIS 2004. In the paper, challenges underlying the diffusion of e-PP in Denmark are identified and explained. The second paper is published
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
28
at the European Conference on Electronic Government (ECEG), 2004. It is selected for publication at the Electronic Journal of Electronic Government (EJEG), 2004. This paper, co-authored by Professor Jan Damsgaard, analyzes the role of institutions in the diffusion of e-PP. Based on the analysis, six prescriptions are given to those coordinating the implementation of e-PP in a nation. In the last paper, the author recommends the use of “critical mass” as the dependent variable for researching diffusion. This paper builds upon sociological concepts such as critical mass theory (Oliver et al. 1985), collective action (Olsen 1965) and threshold behavior (Granovetter 1978). The paper is published in BLED e-commerce conference, 2004.
No.
Title and publication venue
Published on
Rationalizing, Probing and Understanding: The Jan. 2003 Evolution of the Inter Organizational Systems Adoption Field – the HICSS Research Philosophies in the IOS Adoption Field – June 2003 the ECIS Diffusion of E-procurement in the Public Sector – June 2004 Revisiting Centralization Versus Decentralization Debates as a Twist in the Tale – the ECIS Government’s Role in the Diffusion of E- June 2004/ Procurement in the Public Sector – the ECEG and Jan. 2005 the EJEG Operationalizing Critical Mass as the Dependent May 2004 Variable for Researching the Diffusion of Emarketplaces – its Implications – the BLED Table 2 – Refereed Publications Summarized for this Thesis
1
2 3
4
5
Contribution Literature review Research philosophy Research description + findings Research description + findings Future directions
A case study report can be organized in one of the following ways;
i)
linear analytic
ii)
comparative
iii)
chronological
iv)
theory-building
v)
suspense and
vi)
un-sequenced
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
29
Linear analytic is the standard format for composing case study reports. A report written in linear analytic format is organized in the following sequence; problem formulation, literature review, research methodology, case description, case analysis, findings and future research directions. Please look at Yin (1989), page 137 for further details on the five other formatting styles.
This summary is written in linear analytic format. The summarized papers are referred to in the document. The thesis has seven more chapters which are;
i)
theoretical background
ii)
research paradigm
iii)
stakeholders involved in the diffusion of e-PP
iv)
challenges involved in implementing e-PP
v)
policy recommendations for the diffusion of e-PP
vi)
reflections on researching the diffusion of e-PP
vii)
thesis as a summary of referred publications
Each of the seven chapters begins with an overview of its content.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
30
2
2.1
THEORETICAL BACKGROUND
Overview
e-PP is one of many new topics researched within the Information Systems (IS) discipline. Research within the IS discipline has traditionally been built upon methods and theories from established research disciplines such as management science and economics. The IS has emerged as a distinct discipline on its own within the last 25 years. At this juncture, those researching IS topics have the opportunity to apply methods and theories developed both within the IS discipline and in other established disciplines.
This research is primarily built upon theoretical work in the Inter Organizational System (IOS) adoption area. IOS adoption is an actively researched topic in the IS discipline. Theoretical knowledge in the IOS adoption area has cumulatively developed for over 20 years. The author realized the relevance of IOS adoption work for researching e-PP diffusion in the early stages of this study. He reviewed IOS adoption literature in detail while preparing for field work. The results of the review work are published in HICSS 2003. Jeremy Rose, associate professor at the department of computer science, Aalborg University has co-authored the paper. The HICSS paper, which is one of the five summarized papers, is included in the thesis as an appendix document. The paper is titled “Rationalizing, Probing and Understanding: The Evolution of the Inter Organizational Systems Adoption Field”. There is no reference to e-PP in the paper. The review paper was prepared as a stand alone document to satisfy conference requirements. The relevance of the HICSS paper for researching e-PP diffusion is explained in this chapter.
The author learned from field work and secondary research that IOS adoption is one of many reference areas for e-PP diffusion research. The relevance of few reference areas for e-PP diffusion research is briefly introduced in this chapter.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
31
2.2
Relevance of the IOS adoption review paper (HICSS 2003)
“Inter organizational information systems (IOS) are systems based on information technology that cross organizational boundaries; a typical IOS is an information system which links one or more firms to their customers and/or their suppliers and allows the exchange of information-based products and services.” Bakos (1991), page 296. e-PP is a subset of IOS due to the following;
i)
e-PP is based on internet technologies
ii)
e-PP system interconnects public sector buying units and supplier organizations
iii)
Information regarding one or more stages in the procurement activity is electronically exchanged
The review paper studies the historical development of the IOS adoption field. The paper is written based on a detailed review of 73 important literatures. The earliest of the 73 literature is from the year 1966 (Kaufman) and the latest is from 2002 (Kurnia and Johnston). The author finds the field to have developed in three stages; rationalizing, probing and understanding. In the rationalizing stage researchers tried to discover and communicate reasons for some apparently successful IOS. However it soon became clear that IOS adoption was relatively limited and at the probing stage the focus changed to investigating the failure of the promised seamless integration.
No unified coherent
theoretical account emerged from these studies, so in the understanding stage researchers widened the focus of study to try to develop a richer account of the many complex influences on IOS adoption. The three stages are associated with different objectives, assumptions, theoretical backgrounds and research approaches. Please refer to the appendix section for a full version of the paper.
The literature review helped the author a great deal in preparing the research design. The author learnt about choices he had with respect to research approaches, reference theories and philosophical assumptions. As well, he realized that he could contribute towards the accumulation of scientific knowledge in different ways; rationalize, probe and understand.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
32
e-PP diffusion research is in the early stages of growth. Most existing e-PP documentation is of the rationalization type; advertisements, prescriptive, anecdotal and conceptual analysis (Coulthard and Castleman (2001) and Heneriksen et al. (2004)). At this juncture, e-PP diffusion research requires probing or understanding type study. A question to be answered in this regard is whether e-PP diffusion research and IOS adoption research should follow a similar stage wise accumulation of knowledge? If the answer is yes, then probing type studies would be the way to go. The author alternatively recommends understanding type studies for the following reason;
e-PP is a complex subject. Our understanding of the subject is preliminary at this juncture. There is need to learn about multiple stakeholders involved in the diffusion of e-PP and dimensions that define the subject. Once our understanding of the subject is improved, cause-effect relationships can be hypothesized and probed. This line of thinking is similar to the one proposed by Lee (1991) for integrating positivistic and interpretive research approaches. In the next chapter, research methodology issues are discussed in detail.
The key objective of this study is to develop a rich account of complex influences on ePP diffusion. The author would classify this study as an effort to understand e-PP diffusion.
2.3
Reference areas for e-PP diffusion research
This study is driven primarily by theoretical explanations in the IOS adoption field. From field work and secondary research, the author learned that more areas of research are relevant for studying e-PP diffusion. The relevance of the following reference areas is explained in this section; e-government, standard making, IT outsourcing and public procurement. Each of the reference areas is just as expansive as IOS adoption. Hence, a detailed analysis of literature is beyond the scope of this document. This section, by explaining the relevance of a few research areas for studying e-PP diffusion, highlights the scope for further research.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
33
Relevance of e-government research: e-government is an actively researched subject at this time. There are as many as four journals on the subject; E-Government Quarterly, Journal of E-Government, Electronic Journal of E-Government and International Journal of E-Government. In Europe, there are at least two annual conferences on e-government; European Conference on e-government and DEXA conference on e-government. Moreover, there usually is a track on e-government in most information systems conferences.
“E-Government refers to the use by government agencies of information technologies (such as Wide Area Networks, the Internet, and mobile computing) that have the ability to transform relations with citizens (G2C), businesses (G2B), and other arms of government (G2G)” [www.worldbank.org/publicsector/egov/]. The successful implementation of ePP requires transformation of relations with businesses and among government organizations. There are many other activities that require G2B and G2G transformation (e.g.) Internet based export assistance, tax reporting and international trade process streamlining. So far, there have not been many empirical studies on the implementation of e-government systems. However, given that e-government is actively researched, it is quite reasonable to assume that there will be in the near future. A research tradition on the implementation of e-government systems would hopefully develop. It is recommended here that those researching e-PP diffusion associate with such tradition.
There have so far been a few conceptual analysis of e-government. Insights obtained from such analysis are useful for researching e-PP diffusion. An example is provided here to make the case;
Layne and Lee (2001) conceptualize the development of e-government in four stages; i) cataloguing ii) transaction iii) vertical integration and iv) horizontal integration. A government when vertically and horizontally integrated is fully “joined up” (EURIM 2000). Bannister (2003) brings to notice “Turf wars: The hidden menace” that have to be overcome to implement “joined-up” government. Effective integration of public sector organizations requires more than simple integration of heterogeneous technical and
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
34
administrative systems. It means not just process redesign but also shifts in responsibility and power. The latter and especially the power aspect causes problems in the public sector where politics “not only involves the internal politics of organizations, but also inter organizational, local and national politics” Bannister (2003), page 33, paragraph 2.
Procurement in most governments is a decentralized activity. Governments get better value for money when they centrally negotiate frame agreements – for goods and services - and effectively make use of the negotiated agreements. The negotiated agreements are published in e-PP infrastructure. End users in public sector entities can log into the infrastructure and make use of the negotiated agreements. The author learnt from field work that public sector employees, department level and organizational level decision makers resist using centrally negotiated frame agreements. They prefer to negotiate their own frame agreements. It is power and responsibility that they do not want to let go.
e-PP’s success is dependent not just on the development of technical infrastructure but also on the extent to which the infrastructure is used. e-PP infrastructure is not used when the frame agreements posted in it is not used. Bannister (2003)’s conceptual analysis is useful for explaining the resistance against using centrally negotiated frame agreements, which makes his work relevant for researching e-PP diffusion.
Relevance of standard making research: There are several definitions of the term standard. According to one definition, standard is “something established by authority, custom or general consent as a model or example” [www.m-w.com/dictionary]. A standard established by authority/standards body is termed as “de jure”. The Latin term “de jure” when translated to English means “by law”. The following are examples of de jure standards; ISO 9000 quality management systems, the French devised metric system for measuring distances, UN/CEFACT defined EDI standards and the Railway Gauge length enacted by the British parliament. A de facto standard is something that “…is so dominant that everybody seems to follow it like an authorized standard” [http://en.wikipedia.org]. De facto just as de jure is a Latin term. It means “in reality or fact”.
Microsoft Windows for operating system and Microsoft Word for desktop
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
35
publishing are examples of de facto standard. Please refer to Karlsbjerg (2004), page 7 for detailed discussion about de jure and de facto status of standards.
Standard making has been researched for over two decades. Kindleberger (1983)’s explanation of standards as public, collective and private goods is an influential piece of work. He relies upon anecdotal data to explain different types of standards there are and different ways in which they get developed. Katz and Shapiro (1986) analyze one’s decision to adopt a technology standard when he or she is faced with two competing options. The options are differentiated along two dimensions; i) sponsorship and ii) strategic advantage. They explain the implications of one’s decision to adopt on results of the competition. Damsgaard and Truex (2000) critically examine the application of EDI standards in different types of trading relationships. Their examination is based on a few exemplary case studies. They question the suitability of “institutionalized” standards such as EDIFACT and ANSI at the “local” organizational level.
Technology standards are vital for the fast growing information and communication technologies (ICT) area. There is ever growing need to integrate heterogeneous systems across two or more entities. It is not only IT systems that are integrated but also business processes. Often there are two or more initiatives competing to become the de facto standard. Organizations face dilemma in selecting the initiative that would emerge as the de facto standard. Firms that create successful standards on the other hand are set to reap long term benefits. Despite the significance of standard making, there have only been a few studies on the development, adoption, diffusion, deployment and management of standards in the ICT field. In the last 10 years, slightly over 2% of published journal articles in the IS area dealt with standard related issues King and Lyytien (2003). However, the importance of standard making is increasingly recognized. MIS Quarterly, the premier information systems journal, is expected to publish a special issue on standard making in summer 2005 [www.si.umich.edu/misq-stds].
e-PP as mentioned earlier is a networked activity. It requires the buying and selling organizations to interact and trade over a technology standard (e.g.) electronic market, e-
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
36
procurement solution and sales portal. National standard for e-PP is being developed in Singapore (GeBiz) and Korea [www.g2b.go.kr]. In Australia and the US, state governments have self embarked on e-PP. Hence, islands of e-PP networks/standards have emerged. The Danish central government has recommended the use of an e-market (Gatetrade) as the national e-PP standard. It has been three years since the recommendation was made. Gatetrade has so far attracted participation only from some segments (especially ministries) of the government. One or two more e-PP standards appear to have emerged in Denmark. European Union level/regional standard on e-PP seems far away.
The lack of de facto e-PP standard has an adverse effect on the diffusion of e-PP. Some Danish governmental organizations just as supplier organizations have not joined any ePP network. They have adopted “wait and see” approach. Governmental buying organizations hesitate to integrate their financial systems with an e-PP network that is not de facto standard. Supplier organizations get disappointed when they have to participate in several e-PP networks for trading with the government. They ask “is not government a single organization? Why then can’t we trade with government over single e-PP network?” How best to overcome such hesitation and disappointment? What is the most efficient and effective strategy to establish national e-PP standard? A critical researcher might look at the same issue and enquire “should there be national e-PP standard? What are the adverse effects of national e-PP standard?” Standard making research is relevant for answering such questions.
Relevance of IT outsourcing research: IT outsourcing is defined as “…the handing over of some or all of an organization’s information technology (IT) systems (IS) and related services to a third party”(Willcocks et al. 1997). Organizations outsourcing IT;
i)
gain access to specific IT skills and services,
ii)
are able to focus on their core competencies and
iii)
in some cases are able to reduce cost of IT service provision (Seddon 2004)
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
37
The economic value of IT outsourcing is large. EDS, the world’s most experienced IT outsourcing firm, reported $21.5 in billion dollars as revenues in 2003 [www.eds.com]. In the last few years, there has been increasing interest in outsourcing to off shore locations (Khan et al. 2004). India, Ireland, Philippines and Eastern Europe are the commonly used offshore locations (Rajkumar and Mani 2001). The availability of skilled employees for low salaries drives offshore IT outsourcing.
IT outsourcing has been actively researched for over a decade. Alborz, Straub and Tiwana maintain an endnote library on IT outsourcing and systems integration at the ISWorld website [www.isworld.org/endnote]. The library is updated to the year 2003. It has 897 references. A detailed review of all the references is beyond the scope of this report. To give an idea, IT outsourcing research sheds light on the following topics;
i)
determining need for outsourcing (i.e. whether or not to outsource and what to outsource)
ii)
choosing suppliers
iii)
negotiating contracts and
iv)
maintaining outsourced relationships ((Ketler and Walstrom 1993; Lacity 1995; Powell 1993).
Application service provision (ASP) is a recent development in the IT outsourcing area. “Application service provision is a form of selective outsourcing where a third party organization rents generally available packaged software applications and related services” Bennett and Timbrell (2000), page 197. Selective outsourcing is one of many outsourcing variants in where less than 80% of an organization’s IS activities are outsourced Willcocks et al. (1997). ASP is delivered using the Internet mechanism.
Government institutions face unique set of challenges in outsourcing IT. First, they must consider the good of citizenry as a whole while deciding on IT outsourcing. Job creation, societal development and social equality are examples of criteria government must consider. Second, the decision to award an outsourcing contract has to have transparency
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
38
made. Citizens and political parties in a nation have the right to scrutinize the process adopted for awarding a government contract; a right that is often exercised. Third, someone in the government has to be held accountable for consequences resulting from IT outsourcing. Fourth, government institutions may have to relocate or lay off employees in order to obtain full benefits from IT outsourcing. Such however might be hard to achieve as government employees are usually protected by stringent legislative safeguards (Gordon and Walsh 1997).
“Private-public partnerships” (PPP) is gaining prominence as a study area in the recent years. PPP refers to “…forms of cooperation between public authorities and the world of business which aim to ensure the funding, construction, renovation, management or maintenance of an infrastructure or the provision of a service.” EU PPP green paper (2004), section 1.1. Many governments are exploring the suitability of PPP for the development and hosting of e-government systems. The Indian government for example is
actively
seeking
private
sector
involvement
in
its
e-governance
efforts
[www.mit.gov.in/actionplan]. Private sector involvement in e-government is widely being solicited mainly due to two reasons; first, governments can expedite the development of e-government infrastructure using private sector involvement. Second, driven by new public management ideology, governments have been moving towards leaner, efficient and output oriented administrative structure (Schedler and Scharf 2001).
Government has to decide on private sector involvement in implementing e-PP. As in Singapore, government can choose to rely upon their internal IT expertise for implementing e-PP instead of involving the private sector (Devadoss et al. 2002). Or in contrast, government can choose to involve the private sector for implementing e-PP as it is the case of Denmark and the state of North Carolina. The following three questions require answering in this regard; should e-PP implementation be outsourced? What are the advantages and disadvantages of involving the private sector in implementing e-PP? Which of the two alternatives is suitable for expediting the diffusion of e-PP in a nation? When the private sector is involved, government has to decide on the nature of privatepublic partnership (PPP). To what extent should government invest in PPP? To what
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
39
extent should government share risk in the partnership? When engaged in PPP, government runs the risk of getting inextricably connected. With e-PP, due to its networked nature, the risk is higher. As size and activity in an e-PP network grows, switching costs for government increases (Bakos 1991). What exit strategies are there for e-PP PPP? IT outsourcing literature and its variants PPP and ASP literature are relevant for answering the above mentioned questions. Relevance of public procurement research: Procurement accounts for about 1/3rd of public spending (please refer to Jones 2002 and Anderson et al. 2003 for data on Singapore and Denmark respectively). A saving of about 10% in procurement spend implies an overall reduction in public expenditure by about 3-4%. The Irish public sector expects to save about 11% of its spending for goods and services using e-procurement (Clark 2003). Many governments worldwide are increasingly under pressure to reduce their fiscal deficit. Reforming public procurement is one way to minimize fiscal deficit.
Public procurement has existed as an institutionalized activity for a long period of time. The state of Oklahoma established a central purchasing board in the year 1810 (Thai and Grimm 2000). Due to such long history, there must be archives of information on public procurement. Research in public procurement seems to have gotten activated within the last five years. At least, one journal is dedicated to the study of public procurement; “Journal of Public Procurement” [www.fau.edu/pprc/journal]. The journal was established in the year 2001. The usefulness of public procurement research for studying e-PP diffusion is explained here using an example.
Government has to decide on the extent of control/flexibility to be exerted on its procurement activities. Control on procurement is exerted through many means. One way to exert control is by altering “acquisition threshold”; contract value set by government. A government institution is mandated by legislation to advertise a contract opportunity consequently follow formal procedures - when the value of a contract exceeds acquisition threshold. The European Union has set £99,695 as the threshold for supplies and services from the 1st of 2004 [www.ntu.ac.uk/purchasing/JPages/EU_Thresholds_2004]. Control
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
40
is enhanced by reducing acquisition threshold and the vice versa. Another way to exert control is by altering the authority of employees to procure. Control is enhanced by reducing number of employees authorized to procure and the vice versa. The following are other ways in which government can exert control; by altering buying practices and bid protest rights. Pegnato (2003) finds procurement practices in the US federal government to have oscillated since the Second World War times like a pendulum between high control/low flexibility and low control/high flexibility. He predicts the pendulum would continue to oscillate in the years to come.
The pendulum oscillates because there are pros and cons for both enhancing and minimizing control. Take for example the debate on the authority of employees to procure. If number of employees authorized to buy is large, then government institution is not able to get volume discounts. If the number is small (procurement is done by purchasing specialists), then employees (end users) have to go through lengthier buying process. Argumentation underlying the debate changes when e-PP is implemented. Purchasing specialists can negotiate frame agreements on behalf of an institution and make negotiated frame agreements available at e-PP system. Employees working at the institution can log into e-PP system from their desktop computer and directly order supplies using the negotiated agreements. One must not think that the implementation of e-PP has solved the debate on the authority of employees to procure. e-PP instead redefines the debate. Employees continue to seek higher authority claiming dissatisfaction with frame agreements negotiated by purchasing specialists. Their dissatisfaction often results in resistance towards e-PP. The pendulum continues to oscillate.
e-PP diffusion research studies the process of transforming from traditional ways of public procurement to e-PP. Public procurement research sheds light on issues that are specific to the domain. Insights generated by public procurement research are relevant for understanding resistance and support exhibited towards e-PP.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
41
3
3.1
RESEARCH PARADIGM
Overview
The nature of contribution this study makes to scientific knowledge is explained in this chapter. Research philosophy and research methodology issues are taken into consideration to determine the study’s scientific contribution. The author analyzed in detail the accumulation of knowledge in the IOS adoption field from both research philosophy and research methodology perspectives. Please refer to chapter two to learn about the relevance of the IOS adoption field for researching e-PP diffusion. The analysis contributed significantly towards the preparation of research design. The research design has been implemented since the publications were made. In this chapter, the author reflects upon his experiences in preparing and implementing the research design.
The use of the term “paradigm” in the title of this chapter requires elaboration. Kuhn (1962) popularized the use of paradigm in the context of scientific discipline. Lewis and Grimes (1999) define paradigm as “the assumptions, practices and agreements among a scholarly community”. Terms such as method, methodology and technique are often used as replacements for paradigm (Mingers 2001). This chapter could well have been titled “research method” or “research methodology” or “research technique”. The use of paradigm is preferred because broader range of topics can be accommodated under the term than any of the above mentioned replacements. The research paradigm title represents well the topics discussed in this chapter; research philosophy and research approach.
This chapter has three sub sections. The author analyzed the evolution of the IOS adoption field from research philosophy perspective while preparing the research design. Results of the analysis are published in the European Conference on Information Systems (ECIS), 2003. The relevance of the ECIS paper for this study is explained in the first section. e-PP is a multi-level, multi-stakeholder activity. The rationale for selecting nation the unit of analysis is explained in the second section. Research approach adopted for studying the diffusion of e-PP in Denmark is explained in the third section. Diffusion of Electronic Public Procurement in Denmark - Exam Copy
42
3.2
Relevance of the research philosophy paper (ECIS 2003)
The ECIS 2003 paper is the second of five papers summarized in this document. It relies on the same literature sample (73 relevant papers) used for writing the HICSS 2003 paper. However, the ECIS paper is written from a different perspective. It analyzes the application of research philosophies in the IOS adoption field. Orlikowski and Baroudi (1991)’s framework is used to categorize the reviewed literature sample as under positivist, interpretive and critical philosophies. Then, accumulation of scientific knowledge within each of the three philosophies is critically analyzed. In the discussion section, the prospect of inter-relating research philosophies for studying IOS adoption is evaluated.
Research philosophy Positivist “descriptive” Theoretically grounded Interpretive Critical
Frequency 56 (26) (30) 17 0
Percent 76.7 (35.6) (41.1) 23.3 0
73
100%
Table 3 Categorization of Reviewed IOS Adoption Articles by research philosophy
Key findings of the paper are as follows; first, about 3/4th of the reviewed literature sample is rooted in positivist research tradition. The rest 1/4th is interpretive in nature. None of the reviewed papers study IOS adoption from critical philosophy perspective. Table three specifies further details on the categorization of reviewed papers. Second, a number of positivist studies constantly grew from the early 80’s up until the late 90’s. However, the number has dropped drastically in the last couple of years. Interpretive research tradition is getting established since mid 90’s. Figure four depicts growth in the number of positivist and interpretive studies. Third, a few research models such as the following have gained prominence within the positivist research tradition; an empirically derived adoption model (Grover 1993), a small business adoption model (Iacovou et al. 1995) and an EDI adoption and use model (Hart and Saunders 1997). However, a Diffusion of Electronic Public Procurement in Denmark - Exam Copy
43
research model has not emerged as the standard. Fourth, the adoption dependent variable has been operationalized differently such as i) intent/decision to adopt ii) rate of adoption and iii) extent of adoption. Fifth, interpretive studies extend knowledge developed in the positivist research tradition. They provide in-depth understanding of dynamic and complex interactions that affect one’s decision to adopt. Consequently, interpretive studies are limited by their reduced ability to make general statements. Sixth, integration of research philosophies is recommended by one school of thought ((Falconer and Mackay 1999; Kurnia and Johnston 2000; Lee 1991). However, according to another school of thought paradigmatically different research philosophies are incommensurable (Falconer and Mackay 1999).
20 15
Interpretive studies
10
Positivist studies
5 0 19651969
19701974
1975- 19801979 1984
1985- 19901989 1994
19951999
20002002
Figure 4 – Graphical Depiction of Growth in Positivist and Interpretive Studies in the IOS Adoption Field
Jan Karlsbjerg, who was then a doctoral researcher at the department of computer science, Aalborg University, has co-authored the paper. The ECIS paper titled “Research Philosophies in the IOS Adoption Field” is enclosed in the appendix section.
The author learned the following from writing the ECIS paper;
i)
defining characteristics of positivist, interpretive and critical research philosophies
ii)
debate about incommensurability of research paradigms
iii)
application of research philosophies in the IOS adoption field
iv)
scientific progression of the IOS adoption field
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
44
The knowledge acquired from writing the ECIS paper helped the author in defining the philosophical basis for this study. Due to the following reason, this study is rooted in the interpretive research tradition; two questions are to be answered while planning a study rooted in the positivist research tradition; whom to enquire? What relationship to test? ePP is identified as an emerging phenomena. There is lack of knowledge on stakeholders involved and their nature of involvement in the diffusion of e-PP. One for sure could identify stakeholders and define hypotheses based upon IOS adoption and public procurement literature. However, such studies fail to capture uniqueness surrounding the diffusion of e-PP. An explorative study is required at this juncture to learn about complex relationships underlying the diffusion of e-PP. A researcher is well equipped for exploration when he adopts the principles of interpretive field research (Klein and Myers 1999) for understanding the complexity. The understanding that results could be developed further through positivist studies. This study supports the research framework developed by Lee (Lee 1991) for integrating positivist and interpretive research traditions.
3.3
The nation as the embedded unit of analysis
Government represents the buy side in e-PP. While government is commonly referred to as single entity, in reality it is comprised of several institutions. e-PP refers to the use of e-procurement in any governmental institution. Just as it is with the implementation of organizational technologies (e.g. intranets), few governmental institutions initiate the use of e-PP in bottom-up manner. Then, recognizing the advantages of e-PP, a national or a regional policy maker develops strategy for the diffusion of e-PP across governmental institutions in top-down manner. Sellers and technology providers are the two other stakeholders involved in e-PP. Their involvement is just as important as that of government. e-PP happens in a network that has one or more buyers and multiple sellers. The following question requires answering while planning a study on e-PP diffusion; from whose perspective should the diffusion of e-PP be researched; governmental institution, policy maker, seller, technology provider or e-PP network?
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
45
The diffusion of e-PP is measured by the extent to which governmental institutions in a nation or a region use e-procurement. e-PP in India is hardly diffused because only the State of Andhra Pradesh is in the process of implementing e-procurement. An effort to examine the adoption of e-PP in governmental institutions (e.g. enquiry by Maniatopolous (2004) in local government) is for sure useful but inadequate for explaining the diffusion of e-PP. Vaidya and Callender (2004) have developed a research design as per which they query both buyer and seller organizations. e-PP is usually a triadic relationship in where a technology provider intermediates buyer and seller organizations. Hence, the role of technology provider in the diffusion of e-PP is to be enquired as well. Tonkin (2003) studies the role of policy maker in the diffusion of e-PP. Policy maker develops strategy to enhance the diffusion of e-PP in a nation or a region. The implementation of the strategy is mediated by politics within the public administration. Hence, there is need to learn about the involvement of all stakeholders in the implementation of the strategy and not just that of the policy maker.
The unit of analysis selected for a study has to be well bounded. The term “bounded” is explained in Webster’s dictionary as “having the limits or boundaries established”. As ecommerce technologies interconnect multiple organizations, it is a challenge to limit the unit of analysis. A few contributions are made in the recent years to address the challenge1; (Gregor and Johnston 2000; Kurnia and Johnston 2000; Kurnia and Johnston 2002; Reimers 2002; Reimers et al. 2004). The author as well has made a contribution, which is one of the five papers summarized in this document. The paper is explained in detail in the future directions chapter. The most comprehensive study on the unit of analysis is found in Reimers (2002). The application of fourteen units of analyses for researching e-commerce technologies is studied in the paper; (e.g.) transaction, buyersupplier relationship, supply chain, industry, economy, market as networks and industry clusters. He finds neither of the fourteen entirely bounded! 1
The challenge of defining the unit of analysis is not specific to the e-commerce field. Industrial economics have long tried to define business network/cluster/industry (e.g. Jarillo (1988) and Porter 1998). In the economics area, price was for long the undisputed unit of analysis until Williamson (1981) popularized the use of transaction as the unit of analysis.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
46
Earlier studies show that the diffusion of IOS is affected by a broad range of factors (e.g. (Damsgaard and Lyytinen 1998; Gregor and Johnston 2000; Kumar et al. 1998). One can expect the same with e-PP as it is a subset of IOS. The unit of analysis defined for researching the diffusion of e-PP has to account for broader influences (e.g.) organizational culture, international influences and path dependence. An enquiry focused on a narrowly defined unit of analysis runs the risk of making invalid explanation.
Arguments for regarding nation the unit of analysis for researching the diffusion of e-PP are hereby provided;
i)
Government is a non-profit oriented organization that is driven towards serving citizen’ needs. Citizen demands but does not purchases service from government. Supply chain and its variations used as the unit of analysis in the private sector context (Gregor and Johnston 2000) does not suit government well.
ii)
e-PP is more complicated than just good business as Coulthard and Castleman (2001) explain. The diffusion of e-PP has broader implications. For example, e-PP is yet another opportunity for government to reform its administrative infrastructure. As well, governments often embark on e-PP to catalyze the diffusion of e-commerce in society. A national perspective is required to learn about the societal impacts of e-PP.
iii)
e-PP network could be a viable unit of analysis only if government procures through de facto standard. In often cases, two or more e-PP networks compete in a nation. Nation as the unit of analysis provides the overview required for studying the emergence of de facto e-PP standard.
iv)
Countries worldwide are engaged in e-PP. A country implementing e-PP has the opportunity to learn from other’s experiences. Nation is a broad yet well defined unit of analysis. Theorizing the diffusion of e-PP at the national level is a good way to share experiences. See for example, a paper by the author in IFIP 9.4, 2004 titled “Diffusion of e-Procurement in the Public Sector – What
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
47
can we Learn from the Danish Experience?” The IFIP paper is enclosed as an appendix document. v)
Reimers (2002) finds nation (economy as he terms it) the “most easily bounded as decisions regarding the appropriate affiliation of an organized unit (such as a firm) with a political or a regional entity only occasionally raises problems”. He however criticizes the use of nation as the unit of analysis due to the following; statistical analysis, a research technique that “supports” studying nation, is likely to discover only weak associations. To remedy the criticism, nation in this study is regarded an embedded unit of analysis. The use of the term “embedded” is explained later in this section.
Denmark as the Embedded Unit of Analysis
Buy Side National Procurement Agency
Technology Provider
Gatetrade
Kommune Data Policy Maker
Other Stakeholders
Technology Developers (e.g.) Ariba
Standardization bodies (e.g. UN/SPSC)
Kubus Government Association Buyer Professional Buyer End User
Consultants (e.g.) Accenture Seller Continental Government (e.g.) the EU
Figure 5 – The Danish Nation as the Embedded Unit of Analysis
There is unit of analysis and there is level of analysis. Diffusion of e-PP in a nation can be studied at single or multiple levels; employee, organizational, inter-organizational and national. To learn about studies that regard nation as a single holistic unit of analysis Yin (1989), page 49, please refer to the special issue on the Information Society journal
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
48
(Kling et al. 2003). The need for multi-level analysis for studying the impact of IT on organizational change is explained in Markus and Robey (1988). The need for multi-level analysis in the IOS adoption field is acknowledged by a few researchers in recent years (e.g.) Damsgaard and Lyytinen (1998), Johnston and Gregor (2000), Kurnia and Johnston (2001) and Reimers et al. (2004). Damsgaard and Lyytinen (1998) analyze the diffusion of EDI along micro (organizational), meso (industry) and macro (national) levels of analyses. Johnston and Gregor (2000) propose a variant of the micro-meso-macro classification; industry unit-industry group-remote environment.
Traditionally, organization’s decision to adopt IOS has been the object of enquiry. The use of multi-level analysis for researching the diffusion of IOS is becoming common only in the recent years. In a multi-level study, nation is regarded the unit of analysis in the work of Damsgaard and Lyytinen (1998). Alternatively, e-PP network and supply chain are regarded the unit of analysis in the work of Kumar et al. (1998) and Kurnia and Johnston (2001) respectively.
In embedded study design, attention is given to subunit/subunits within a case Yin (1989), page 49. Nation in this study is regarded the embedded unit of analysis due to the following; the diffusion of e-PP in a nation is dependent on the coordinated participation of multiple stakeholders. For example, an e-PP standard recommended by policy makers becomes de facto only when organizational level decision makers and end users adopt in both buying and selling institutions. The embedded unit of analysis notion adopted in this study is similar to that of Damsgaard and Lyytinen (1998). The Danish nation is depicted as the embedded unit of analysis in figure five.
3.4
Research approach
The following criteria are taken into consideration for determining the research approach;
i)
the approach has to allow for exploration
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
49
ii)
the approach has to allow for studying complex relationships that underlie the diffusion of e-GP
Case study research approach (Yin 1989) is selected for studying the diffusion of e-PP as it satisfies both the criteria. The research approach as implemented in this study is hereby explained under research design, data collection and data analysis.
Research design: Most aspects of the study design have already been explained in this document. To summarize, explorative case study design is adopted for answering the research questions. The study is guided primarily by IOS adoption literature. Please refer to chapter two for details on the IOS adoption field. The Danish nation is chosen as the embedded unit of analysis. Stakeholders/subunits involved in the diffusion of e-PP are identified using snow balling technique (Moriarty and Bateson 1982). The technique has its roots in institutional sociology where effort is made to define “organizational field” (DiMaggio and Powell 1983). The author learned about network of stakeholders involved in the diffusion of e-PP as he did his research. Thus, the research sample is known only after the study is completed and not a priori. Such technique is commonly used in multilevel IOS diffusion studies (e.g. (Damsgaard and Lyytinen 1998) and (Reimers et al. 2004)).
Data collection: Primary data for the study was collected using semi-structured interviews. Thirty three stakeholders were enquired in five month period. The stakeholders are of the following types; buyer (professional), buyer (end user), technology provider, seller (decision maker), e-procurement coordinator (in the private sector), national procurement agency, policy maker and regional government association. A table titled “interview data” is enclosed in the appendix section. The table has information on the sequence in which the interviews are made. A page long interview guide was e-mailed to the interviewees before the enquiry. The guide is enclosed as an appendix document. The author learned as much as possible about interviewed organizations through secondary research (web pages primarily) in preparation. Interviews were audio recorded and transcribed upon obtaining interviewee’s permission. All stakeholder types were in principle enquired about the same issues; their involvement
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
50
and their perception of others involvement in e-PP. Questions posed during interviews are open ended with few exceptions. When some interviewees were simply too long winded, a few well specified questions are posed. During interviews, the author sought information on other stakeholders involved in the diffusion of e-PP. An interview lasted on an averaged for about 75 minutes. A transcribed interview enclosed in the appendix section is typical. Secondary data was collected from the following sources; consultant reports, national statistics, newspaper reports and Internet home pages.
Data analysis: The author analyzed the data as he collected it. The analysis is focused on explanation building Yin (1989), page 113. In the beginning, the interview was informed primarily by IOS adoption literature. As the data collection effort progressed, the author’s understanding of the subject evolved (i.e.) he learned who were the stakeholders involved and the nature of their involvement. The improved understanding however did by no means alter the focus of the study. Interview guide e-mailed to the interviewees remained the same through out, except for few representational alternations. Here is one example of how the data is analyzed;
Purchasing officers at ABC, XYZ and DEF Universities were interviewed as part of the study in the same time sequence. All the three were enquired about their decision to join Gatetrade, an e-PP technology standard recommended by the ministry of science, technology and innovation (policy maker). The purchasing officer at ABC University (Mr. X) was skeptical about joining Gatetrade. Due to several reasons, he did not expect Gatetrade to succeed in the long run. Mr. X preferred to adopt “wait and see” approach. During the course of the interview, the author learned that XYZ University was considering joining Gatetrade and that DEF University was one of the first to adopt Gatetrade. Purchasing officers in Danish universities apparently knew one another; this find fits well the assumption DiMaggio and Powell (1983) make for defining “organizational field”, the theoretical basis for designing the study. Mr. X kindly helped in setting up an interview with DEF University.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
51
The purchasing officer (Mr. Y) at XYZ University was interviewed next. Mr. Y’s perception about joining Gatetrade was enquired during the course of the interview. Mr. Y was inclined towards joining Gatetrade but had not yet joined. He explained that eprocurement through initiatives such as Gatetrade is the way forward. Moreover, he was aware that DEF University had joined Gatetrade. He reasoned that should difficulties arise, he could rely upon DEF University’s experience in implementing e-PP.
Then, Ms. Z at DEF University was enquired about her decision to join Gatetrade. DEF University has invested heavily in integrating its financial system with the Gatetrade infrastructure. Ms. Z reasoned as follows for joining Gatetrade; DEF University is one of the largest and oldest governmental bodies in Denmark. It can afford to risk investing in Gatetrade. The Danish society will benefit if Gatetrade succeeds. Ms. Z was optimistic about Gatetrade’s future.
The three interviews were analyzed as follows; organizations have to make asset specific investments to join Gatetrade. If Gatetrade fails to break-even then the asset specific investments are lost. Those who join/invest in Gatetrade expect the initiative to succeed and those who refrain from joining are skeptical about its future. Thus, one’s expectation regarding Gatetrade’s future affects one’s decision to join. This conclusion forms the very basis for a research approach, which is explained in one of the five summarized papers titled “Operationalizing Critical Mass as the Dependent Variable for Researching the Diffusion of E-marketplaces – its Implications”. A summary of the paper is provided in the future directions chapter.
The author arrived at the above mentioned conclusion just after having interviewed Ms. Z. This evolved understanding was put forth to Gatetrade’s marketing manager. The manager’s response helped in building a better explanation. Many more issues were researched in the manner explained above. For the full study period, the author was focused on building a better explanation of e-PP diffusion in Denmark.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
52
A multi-level study results in vast amounts of data. Some form of logic has to be applied for analyzing and making sense of the data. To the author’s knowledge, three approaches are used in the IOS adoption field for processing data in multi-level study. First, Damsgaard and Lyytinen (1998) define and use a concept termed “diffusion pattern”. As per the concept, a study is done to find diffusion patterns. A pattern could be found in and across any of the three levels of analyses; micro, meso and macro. Rich data gotten from multi-level study is used to explain the discovered pattern. Second, Structuration theory is used for understanding the adoption of IOS at the industry level (Johnston and Gregor 2000). A key assumption underlying the approach is as follows; an organization’s (individual unit) decision to adopt (action) is enabled and constrained by industry and remote environment (structure). The Structuration type approach recommends;
i)
tracing relationships that define structure and
ii)
studying constraining and enabling effects of structure on one’s decision to adopt.
An approach similar to that of Damsgaard and Lyytinen (1998) is adopted for processing rich data collected from the study. Instead of many patterns, the author sought after one coherent explanation of e-PP diffusion in Denmark. Rich data collected from this study is presented as debates on centralization versus decentralization. The debates happen in the areas of organizational and inter-organizational decision making and regarding computing infrastructure. The discussion is focused on understanding why these debates originate. The result of the discussion is presented as single line answer! Please refer to chapter five for details on the debate.
3.5
Validity of the study
The final explanation resulted from the study evolved as the author interviewed many stakeholders over five month period. As different types of stakeholders were enquired about the same issues, internal validity for the study is ensured. The study’s result is theorized using analytical generalization technique (Eisenhardt 1989; Lee 1991; Markus
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
53
1983; Yin 1989). Six policy recommendations are made for facilitating the diffusion of ePP in nation. The recommendations are applicable for any nation engaged in implementing e-PP and not just for Denmark. Please refer to chapter six for further details on the recommendations.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
54
4
STAKEHOLDERS INVOLVED IN THE DIFFUSION OF E-PP
4.1
Overview
The first of the three research questions is answered in this chapter. The involvement of the following stakeholder types in the diffusion of e-PP in Denmark is explained in this chapter;
i)
national procurement agency
ii)
buyer (professional)
iii)
buyer (end user)
iv)
policy maker
v)
technology provider
vi)
seller (decision maker)
vii)
regional government association and
viii)
e-procurement coordinator (in the private sector)
The diffusion of e-PP is dependent on coordination among the above mentioned stakeholders. Please note that some parts of the chapter are taken as it is explained in ECIS 2004 and EJEG 2005 papers. This chapter provides the most comprehensive overview of the case. The reader after having read this chapter can choose to ignore case description in ECIS 2004 and EJEG 2005 papers. An overview of the Danish public sector is provided first. Then, the involvement of the above mentioned stakeholder types are explained.
4.2
The Danish public sector
In terms of population Denmark accounts for 1.4 percent of Europe. The public sector plays an important role in the Danish economy. One third of the Danish workforce is employed in the public sector [www.denmark.dk]. The government’s expenditure amounts to around 25% of the GDP. Danish per Capita GDP is one of the highest at $29,700 USD [www.usadk.org] In the 2002 year, Global Corruption Report (Lamsdorff
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
55
2002) ranks Denmark as the second least corrupt country. It is a welfare state with the notion of collectiveness prevalent. For instance, Denmark was one of the few founding members in developing the world’s first mobile system [www.sintef.no] with roaming, which then evolved to GSM.
The Danish public sector administration as already mentioned is tri layered. The three layers are i) municipality (Kommune) ii) county (Amt) and iii) the state (ministry). There are 275 local bodies, 14 regions and 18 ministries. The local bodies account for about 50% of the nation’s expenditure. Local bodies are to a large extent autonomous as local government taxes account for 33% of the nation’s tax yield. Local bodies are selected by the people and do not as such serve the state administration [www.denmark.dk].
4.3
National procurement agency (SKI A/S)
National procurement agency (SKI) was established as a joint venture between the ministry of finance (55%) and the national association of local authorities in Denmark (45%) a decade or so back [www.ski.dk]. SKI negotiates frame agreements on behalf of all Danish governmental organizations. The agreements are available to all Danish governmental organizations as choice. Many buying organizations prefer negotiating their own agreements instead of using SKI agreements. Due to this lack of utilization, SKI has not been to negotiate good agreements with suppliers. The lack of good agreements reinforces buyer’s decision against adhering to SKI agreements. SKI is facing a typical catch-22 situation.
SKI negotiated agreements are available in Gatetrade. End users can log into Gatetrade and utilize SKI negotiated agreements using a simple procedure. SKI works closely with Gatetrade to the extent that SKI executives are in Gatetrade’s steering committee. A SKI executive explains collaboration with Gatetrade as follows;
“…actually because it is such a big and important project, our owners in the Danish state wanted us to be a part of it and be involved in it, and that means that we are part of
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
56
the steering committee and part of some of the project teams that involve DOIP, and therefore we have tight relations with Gatetrade.net”.
4.4
Buyer (professional)
A government organization’s purchasing activities are usually managed by one or more professional buyers. Professional buyers maintain overview of an organization’s procurement activities. They are also responsible for purchasing goods and services required by an organization through the best possible means (measured by price, quality and efficiency). In some cases, professional buyers adhere to SKI negotiated frame agreements. They could as well choose to negotiate organization specific frame agreements. The extent to which frame agreements selected by professional buyer is adhered to by end users in an organization is dependent on culture and politics.
Work culture among governmental organizations appears to vary significantly. In organizations with hierarchical work culture, end users are mandated to use frame agreements selected by professional buyer (e.g.) as in tax authority and military. An end user in a hierarchical organization explains the point as follows;
“We are not allowed to go outside….I cannot go to a store and buy some pens because it is cheaper. I have to buy by the (frame) agreement”
Organizations in where authority is decentralized, it is difficult to mandate end users to adhere frame agreements (e.g.) as in Universities. A professional buyer in a Danish university explains the difficulty as follows;
“Nobody wants to get into conflict with the departments. The idea they have more and more buying for their own department.
Previously, teknisk forvaltning (purchasing
department) was the place for their purchasing ends. That is how it (purchasing) should be done. The whole idea their (purchasing department) part of purchasing is declining. The departments own buying is rising. There is not much left of the teknisk purchasing.”
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
57
Professional buyer representing an organization looses credibility for negotiating frame agreements when end users choose to purchase goods and services on their own. A Catch-22 situation results just as explained in the previous section. In general, professional buyer decides on implementing e-PP in an organization. The implementation of e-procurement reinforces hierarchy in the former case and provides opportunity to centralize decision making in the later.
4.5
Buyer (end user)
The need for goods and services purchased originates with end user. End user can represent his or her need (e.g. researcher buying a flight ticket), a department’s need (e.g. secretary purchasing stationary for a department) or an organization’s need (e.g. professional buyer representing subsidiary). Professional buyer in subsidiary organization becomes an end user when he or she is mandated to use frame agreements negotiated by parent organization. Parent-subsidiary concept does exist in governmental institutions such as the following; the fishery ministry, the tax office and police. All the three organizations have offices/subsidiaries through out Denmark.
End user (individual, department or organization) is allocated/budgeted a sum of money, which he or she uses for carrying out everyday activities. End users are inclined towards using resources allocated to them optimally. Typically, end users explain that they could negotiate frame agreements better than the centrally negotiated agreements. Here are two such explanations;
Explanation one by department secretary in a university; “We (end user) thought so (centrally negotiated agreements would be cheaper). But in a couple of years, we started checking on the prices ourselves. If we buy furniture directly from furniture stores, we can get a lot cheaper than if we buy from the university because her (professional buyer) salary needs to be paid. If they (professional buyer) get 500 DKK table and sell it for 550 DKK. Then the 50 is for her salary. We would not allow that. So we buy it ourselves.”
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
58
Explanation two by a purchasing professional in a subsidiary organization; “The (centrally) frame agreements are very expensive. They tell us that they are cheap because we are a lot of people who is buying by the agreement. That’s not the reality. We can see that. We can see that that is much more expensive. Colors for the printers where I got examples, if I buy outside the agreement, it is about 100 Kroner cheaper than in the agreement. That’s a lot of money. If I have 50 of these colors, that’s a lot of money.”
There is an element of paradox in the explanations provided above. One would think that due to volume discounts, centrally negotiated frame agreements would be better (e.g. cheaper). But, typically end users explained that they could self negotiate better deals. Their statements appear to be correct. Why are not centrally negotiated agreements better than individually negotiated agreements? The paradox became apparent quite early in the study. The author enquired many stakeholders about the paradox. He however has not been able to find a seemingly valid explanation. A study for resolving the paradox is a worthy pursuit.
The successful diffusion of e-PP depends on the extent to which end users utilize centrally negotiated frame agreements. Professional buyers acknowledge that educating end users is the most effective way for enhancing the use of centrally negotiated agreements. But then, if centrally negotiated agreements are not that effective, should they be used at all?
4.6
Policy maker
Around the year 2000, the ministry of science, technology and innovation [www.videnskabsministeriet.dk] developed a strategy for expediting the diffusion of e-PP. An officer explains the rationale behind the strategy as follows;
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
59
“Basically it was three, four years ago 2 we started seeing a marketplace (fails to recollect the name) but we looked at that and got the idea that may be we could make a marketplace for the public sector as well. And then we started; we presented it for the sminister and we presented for the government and they agreed that we should try something like it. Because, it was the opinion that public sector could save some money by doing e-commerce and hardly any in the public sector had started using e-commerce. So, we thought that making the infrastructure ready with a portal, we could push the development of e-commerce… That is the whole idea. Because when we ask the ministries and the communities why they were not buying things electronically, they would say well it is too expensive, it is too difficult, what systems should we choose? The public sector cannot use this and a lot of excuses. If we make sure that they have the infrastructure and the right conditions at least they don’t have that excuse.”
The government called for proposals for developing and hosting electronic trading infrastructure over which public bodies can procure. Out of five submissions, Gatetrade, a consortium formed by four large Danish organizations (Maersk data, Tele Denmark, Post Denmark and Danske Bank), was selected the winner on the 21st of March 2001. The ministry recommends Gatetrade as the national e-PP technology standard.
4.7
Technology provider
In addition to Gatetrade, the two following organizations are competing to provide e-PP technology in Denmark; Kommune Data and Kubus. The involvement of all the three organizations in the diffusion of e-PP is explained in this section.
Gatetrade: It is an e-market open for both private and the public sectors [www.gatetrade.com]. Some in the public sector support Gatetrade for the following reasons; first, governmental organizations are able to access and utilize SKI negotiated agreements available in Gatetrade with much ease. Second, the Danish state gets a better overview of its purchasing activities by channeling its trade through Gatetrade. The 2
The interview was conducted during March 2003.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
60
enhanced overview helps while negotiating frame agreements. Third, procurement related administrative costs are reduced with the enhanced use of e-commerce. Fourth, the public sector by channeling its trade through Gatetrade supports the development of ecommerce standard. The availability of standard helps in the diffusion of e-commerce in society. Fifth, interoperability is enhanced when governmental organizations and sellers regard Gatetrade the e-procurement standard.
Gatetrade has been operational since Jan. 2002. Gatetrade earns revenue primarily by charging transaction fee for documents exchanged over its infrastructure. In recent times, it has added a portfolio of value added services such as IT and business consulting. The organization however has had difficulties in breaking even. As of spring 2004, the Danish public sector had not used Gatetrade as much as it was envisioned. Local bodies that account for about half the nation’s expenditure have resisted joining Gatetrade. Problems for Gatetrade are aggregated in that its four large owners are yet to fully utilize the emarket. Gatetrade declared a loss of 32 million DKK (app. 5 million USD) in 2003 as against 44.2 million DKK (app. 7 million USD) in 2002. Gatetrade’s owners however declared their support by infusing 160 million DKK (app. 27 million USD) in the initiative, recently. They invested an additional 60 million DKK (app. 10 million USD) in July 2003 (Anonymous 2003a). Gatetrade declared quadruple increase in trade for the 2003 year (Anonymous 2004). The director Steen Gade expects Gatetrade to break even around the 2007-2008 period (Anonymous 2003b). The Danish state, Gatetrade’s owners and the Danish business community account for 40, 30 and 30 percentage of trade conducted through Gatetrade (Anonymous 2004).
Kommune Data (KMD): KMD was formed by the national association of local authorities and the association of county councils in 1972 to serve local governmental organizations’ IT needs [www.kmd.dk]. Most local governmental organizations have implemented KMD’s financial system (KMD Ø). KMD has recently developed an e-procurement/webpurchasing module. A chief officer in KMD reasons the development of e-procurement module as follows;
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
61
“One of the key demands from the municipalities at that time and still today is; KMD you must develop a procurement system which is closely integrated to our financial system. Because the whole process is about ordering, about the invoice; all related back to our financial controlling system…This is our situation. We are building iprocurement system that is at a minimum (here meaning very) closely related to the financial. Because we know that is what the municipalities want and that is what they will pay for”
Kubus: It is a small sized software development firm focused on the contract management aspects of e-procurement [www.kubus.dk]. Trade Builder, its e-procurement solution, won the e-business prize (e-handelsprisen) for the 2002 and 2003 years (Anonymous 2003c). The trade builder solution is primarily used by a few local governmental organizations. The technical director of Kubus explains the value added by the Trade Builder solution;
“In about the last quarter of year 2000, we got involved in two Kommune (municipalities). They were doing e-procurement or procurement together where the institutions and schools were buying at the same partners. They wanted to make the catalogue they had in paper; they wanted to make it electronic… and then we focused more on the (frame) agreements you have…. All agreements, all those you cannot procure on the Internet, also them you have to call him on the phone, those you have to print out some reports and fill out with a pen; 100%, one place you find it all. So, you don’t have the portal of services you can buy on the Internet. But you also have all the other agreements and the things you have to do at one place.”
4.8
Seller (decision maker)
e-PP requires participation not just from buyers but also from sellers. To engage in e-PP, sellers have to prepare and regularly maintain electronic product catalogues. The process of preparing electronic catalogues can be quite resource consuming as number of products sold ranges in tens of thousands. As well, supplier has to select the infrastructure for hosting electronic catalogues. Technology providers usually charge a recurring fee for
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
62
catalogue hosting. Large suppliers prefer to host catalogues in their own web shops than in third party infrastructure as it is less expensive. Most e-markets have the facility to seamlessly “punch-out” buyers from their infrastructure to supplier web shop. Small supplies (e.g. furniture shops) serving local needs tend to find e-procurement unattractive.
Suppliers are attracted by the size of government business. If government exhibits inclination towards e-procurement, then capable suppliers follow suit as explained below;
“We (a supplier) had…about half of our business-to-business (company also has business-to-consumer sale) with the government or local public organization …because the government is sort of sanctioning these sales channels (market X)… sort of …this is what you should use…so we though this might work, so we said we cannot afford not to be there. If it kicks off then we must be there, so we decided to join early, well, that’s about it.”
However, suppliers loose interest in a third party infrastructure when adequate business does not result from participation. The supplier who made the previous quote states the following;
“We do sell on market X (name removed for practical reasons) and our products are available there, we have a limited selection of products available there. In the beginning we had our whole products listed there, generally available on market X but then we decided to cut down a bit. We have not sold, we sold practically nothing at that. I think its been tough because we do not described our products very well on market X but I think it is also a part that market X has not sold anything much at all. I do not know this but that is the way I feel.”
As hosting electronic catalogues is resource demanding, supplier prefers to trade with government over de facto e-PP standard.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
63
4.9
Regional government association (ARF)
The role of regional governments in Denmark was being scrutinized by a structure commission when the regional government association (Amts Råd Foreningen) was interviewed [www.arf.dk]. As per one scenario, regional governments would be dissolved. The Danish government would then move from three-tier to two-tier administrative structure. ARF maintains overview of procurement in the Danish regions. They facilitate the diffusion of e-PP by preparing reports and coordinating meetings. ARF’s role as explained by one of its employees is as follows;
“We are trying to make reports, we and Danish region. We are trying to take all the group of products, for example, pens, bottles, and healthcare products. And try to divide them into three categories… We have this coordinating function. We are not buying anything. Our job is mainly to facilitate the common initiatives that the purchase chiefs decide they want to do.”
4.10 E-procurement coordinator in the private sector E-procurement coordinators in the private sector were interviewed to learn about their experiences in implementing e-procurement. Private sector and the public sector face similar challenges in the following;
i)
Convincing end user to use centrally negotiated frame agreements!
ii)
Deciding on technology standard
iii)
Enticing supplier to participate in e-procurement
E-procurement coordinator’s involvement in the diffusion of e-PP is minimal.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
64
5
5.1
CHALLENGES IN IMPLEMENTING E-PP
Overview
The second of the three research questions is answered in this chapter. In one of the five summarized papers, challenges encountered in implementing e-PP in Denmark are analyzed. The summarized paper, which is published at the European Conference of Information Systems, 2004 is put in perspective in this chapter. Please note that “implementing e-PP” in the chapter title is used in generic sense. A wide range of challenges encountered in implementing e-PP can be analyzed from multiple perspectives under the title. Challenges encountered in implementing e-PP in Denmark are theorized in the summarized paper and in the following section. The theory is applicable not just for Denmark but also for other nations. Hence, Denmark is referred neither in the chapter title nor in the second research question.
5.2
Centralization versus decentralization paper in perspective (ECIS 2004)
In the ECIS paper, rich data collected from the field study is used for discussing challenges encountered in implementing e-PP. Debate format is used for discussing the challenges because it allows for presenting multiple perspectives. The debates were found happening in the three following areas; frame agreement negotiation at the organizational, inter-organizational levels and computing infrastructure. Debates in the three areas are about centralization versus decentralization. The nature of debate is revealed using quotations from interviews. Theoretical knowledge especially from the information systems area is relied upon for explaining the debates scientifically. A Meta level explanation on the challenges is made by reflecting upon the debates’ genesis; why do these debates originate? The question is answered with single line response. The response is presented as a “twist in the tale” in where the debates are the tale. It is hoped that the reader would understand the tale much clearly after learning the twist. The paper is titled “Diffusion of e-Procurement in the Public Sector – Revisiting Centralization versus Decentralization as a Twist in the Tale”. A summary of the tale and the twist in the tale are hereby presented. Please refer to the appendix for the full paper.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
65
The tale: Centralization versus decentralization debate arises in three areas; organizational, inter-organization and computing infrastructure. In the organizational area, the debate is between end user and professional buyer stakeholder types. Professional buyer recommends the end user to use centrally negotiated frame agreements posted in the e-PP infrastructure. The end user typically finds centrally negotiated agreements dissatisfactory especially on price. In order to optimally make use of the budget allocated to him or her, end user prefers to self negotiate deals with suppliers (i.e.) decentralized decision making. e-PP is successfully implemented only when end users utilize centrally negotiated frame agreements. Hence, the debate has come to the forefront. The work of Gurbaxani and Whang (1991) is mainly used for scientifically analyze the debate. Please refer to the full paper for scientific analysis of the debate.
The debate on centralization versus decentralization at the inter-organizational level is similar to the one at the organizational level. Stakeholders involved and arguments presented for and against centralization however differ. The debate at the interorganizational level is between professional buyer and the national procurement agency (SKI) stakeholder types. Citing efficiency reasons, SKI recommends the use of frame agreements centrally negotiated on behalf of the Danish nation. Professional buyer representing governmental organizations prefer to self negotiate (i.e. de-centrally decide) frame agreements as they find SKI negotiated agreements dissatisfactory. Argumentation made in this debate brings to light the paradoxical role of government. Those arguing for decentralization explain that government should be driven not just by efficiency but by wider range of variables such as encouraging small and medium sized enterprises, catalyzing economy and encouraging innovation. Please refer to the full paper for a detailed analysis of the debate.
The third debate is about the centralization of computing infrastructure. The ministry of science, technology and innovation, as explained in the previous chapter, recommends Gatetrade as the national e-PP standard. Those arguing for centralized computing infrastructure explain that de facto e-PP standard would enhance interoperability among
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
66
governmental organizations and catalyze the diffusion of e-commerce in society. Local governmental organizations however mainly due to path dependency issues are inclined towards adopting Kommune Data’s web procurement module. A few governmental organizations have chosen Kubus Trade Builder System to encourage the development of an innovative firm. Research done on the late 70’s and early 80’s on centralized versus decentralized computing (King 1983) has relevance for analyzing the emergence of de facto national e-PP standard. The full paper has further details on the debate.
The twist in the tale: To find a resolution for the three debates, it is important to know the genesis of the debates. Precisely, wherein lays the difference? Do all the three debates originate due to a particular reason? Yes, they do. The author contends that all three debates originate due to differences in perception about the extent to which government is single organization. The differences in perception are manifested in all three debates. Here is an example to show the manifestation; those supporting the uptake of SKI negotiated agreements regard government as one organization that administers the Danish nation. To them, government as a whole has to become efficient and the national economy as a whole has to be catalyzed and so on. If one perceives government as single organization, then by default he or she supports SKI negotiated agreements. However, usually only those working for the state government tend to perceive government as single organization. Those working in universities and in local bodies perceive the organization (in a limited sense) that they work for as government. Most governmental organizations are autonomous to an extent. The local bodies for example are dependent on local taxes for about 2/3rd’s of their expenses. Hence, to a large extent they can act as they wish. Local bodies would prefer to award contracts to domestic companies as such would help them enhance their autonomy. One level further down, end user tends to perceive department that they work for as the government. Department is allocated a certain sum of money for its operations. End user is aligned towards using the allocated sum in the most efficient manner. Given free choice, end user would use centrally negotiated frame agreement posted in e-PP infrastructure only if it suits his or her purposes. Values such as catalyzing national economy, e-government leadership and
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
67
societal development tend not to make as much impression to an end user as it would to a national policy maker.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
68
6
6.1
POLICY RECOMMENDATIONS FOR THE DIFFUSION OF E-PP
Overview
The last of the three research questions is answered in this chapter. In one of the five summarized papers, policy recommendations with respect to the diffusion of e-PP are made. The paper, which is published in the European Conference on Electronic Government (ECEG), 2004, is summarized in this chapter for answering the third research question. The paper is co-authored by my supervisor Professor Jan Damsgaard. It is selected for publication at the Electronic Journal of Electronic Government (EJEG), Jan. 2005; provided minimum of 20% text is revised by the 22nd of October 2004. The paper tentatively titled “Government’s Role in the Diffusion of E-Procurement in the Public Sector” is added to the appendix.
6.2
Summary of the policy recommendation paper (EJEG 2005)
Government often intervenes to facilitate the diffusion of an innovation. King et al. (1994) s institutional intervention framework is used for analyzing the diffusion of e-PP in Denmark. The framework has as its dimensions supply push and demand pull in one axis and regulation and influence in the other. Institutional intervention in Denmark with respect to the diffusion of e-PP is analyzed under six specific actions; knowledge building, knowledge deployment, subsidy, mobilization, standard setting and innovation directive. A conjecture is derived for each analysis. The six conjectures that resulted are presented below as policy recommendations for the diffusion of e-PP. Please refer to the full paper for analyses that resulted in the following recommendations.
Recommendation one:
A part of the government has to take proactive efforts for
encouraging the diffusion of e-procurement in the public sector. In the process however, it has to adequately involve all segments of public administration while deciding upon the trading infrastructure.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
69
Recommendation two: A government’s involvement in educating employees is minimized when the development and the hosting of e-procurement solutions are outsourced.
Recommendation three: Government via outsourcing the development and hosting of eprocurement solutions can share subsidizing responsibilities with private investors. Given the networked nature of e-procurement, it is recommended that subsidies be provided to both the supply side and the demand side for attracting a critical mass of users and participation.
Recommendation four: Governmental associations can via active mobilizations efforts have a positive influence in the diffusion of e-procurement. It should however be noted that e-procurement is successfully diffused only when both organizational level decision makers and end users are convinced to adopt.
Recommendation five: A solution selected as the standard should be provided as a recommended choice and not imposed upon.
Recommendation six: Innovation directives such as that of reducing budget allocations to governmental organizations can backfire when organizations post implementation realize that savings obtained do not match the budget cuts.
As the recommendations are generic in nature, they are valid not just for Denmark but for any other nation engaged in the implementation of e-PP.
The analysis shows that as public administration is politically managed, government is mainly able to influence and not regulate both the supply and demand sides. A regulatory action can be misinterpreted as an effort to alter power structures within the public administration.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
70
7
7.1
REFLECTIONS ON RESEARCHING THE DIFFUSION OF E-PP
Overview
As explained in chapter one, only minimal efforts have so far been taken for studying ePP as it is a new subject. The objective of this study is to “explore and learn” about the diffusion of e-PP. Challenges encountered in researching e-PP and in implementing e-PP are discussed in prior chapters. In chapter two for example, the relevance of five research strands (IOS adoption, e-government, standard making, IT outsourcing and public procurement) for studying the diffusion of e-PP is discussed. The discussion highlights potential for further research. Challenges encountered with respect to implementing e-PP in Denmark are discussed in chapter five.
Given the e-PP is at its early stages, research tradition for studying the diffusion of e-PP is just emerging. For the rationale explained in chapter three, explorative case study design is adopted for this study. The research design adopted in this study was found useful for learning about e-PP. However, the author while implementing the design encountered a few difficulties. The difficulties realized are explained in this chapter. It is hoped that those researching e-PP diffusion further on would find ways to effectively counter such difficulties.
During the course of field work, the author developed a research approach for studying the diffusion of networked technologies. In chapter three, the role of field work in conceiving the research approach is explained. An initial version of the approach was presented at the Research Symposium on Emerging Electronic Markets (RSEEM), 2003 (Ramanathan 2003). The approach went through many revisions before it was published at the Bled e-commerce conference, 2004. A summary of the paper is provided in this chapter.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
71
7.2
Difficulties in Researching the Diffusion of e-PP
A few difficulties arose while implementing the study design. In the section, difficulties encountered while implementing the study design are reflected upon. The reflections are meant to pave way for further research.
Disentangling e-PP: e-PP, a networked technology, requires participation from both buyers and sellers. On the buy side, procurement related decision making is centralized differently. The decision maker can be placed anywhere in governmental hierarchy. To learn about the diffusion of e-PP in a nation, who on the buy side should be queried? It could not be that all governmental employees are queried about participating in e-PP. How about progressing in steps? First, take efforts to classify governmental organizations based on their purchasing practices (e.g.) decision making at national, organizational and end user levels. Then, enquire decision makers in a targeted manner. But then, governmental organizations are hard to classify based on purchasing practices for the following reasons;
i)
an organization could purchase some types of goods using national frame agreements and other types through maverick buying. In which case, decision making is both highly centralized and highly decentralized.
ii)
procurement decision making within an organization could be constantly shifting like pendulum swinging. In one of the interviewed organization, the author learned that purchasing department has been gradually loosing its authority. However, an effort was underway to recentralize purchasing.
Usually, tens of thousands organizations sell to government. It is unrealistic and even ineffective to enquire all sellers about their decision to participate in e-PP. Who then in seller population should be queried? It is important that a sample selected for research represents well the seller population. Upon what dimensions should a representative sample be selected (e.g. adopter, non adopter, network specific, and nation specific)? How about classifying sellers based on characteristics of products (e.g. frequency of purchase, commodity and value) they sell? Then, relationship between a type of seller
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
72
and decision to participate in e-PP can be studied. A practical challenge lays in identifying suppliers to fit the classification.
There is adoption of e-PP and there is use of e-PP. The diffusion of e-PP requires organization to both adopt and use e-PP. An organization that has adopted e-PP does not by default use e-PP. The time taken between adoption and use of an innovation (e.g. ePP) is termed as “assimilation gap” (Fichman and Kemerer 1999). Assimilation gap is researched at the organizational level. In recent years, there is an increase in interest to learn about assimilation gap. When diffusion of e-PP is researched at the national level, how should researcher take into account assimilation gap at the organizational level? Is there a mechanism to relate adoption and use challenges at the organizational level to diffusion challenges at the national level? Gidden’s Structuration theory is suggested as a mechanism for relating multi-level data. Please refer to chapter three for further details on the suggestion. Later on in this chapter, the author recommends a mechanism for processing multi-level data.
Is researcher Sherlock Holmes? Yes. It appears so. The author had to do detective work for explaining the diffusion of e-PP in Denmark. By and large, all stakeholder types were very cooperative. They were honest in their response. However, there is a technique involved in setting up interviews with stakeholders. In the beginning, the author mailed out detailed questionnaires to potential interviewees and sought an interview. Almost all the solicitation resulted in rejections. Then, topic based questionnaire enclosed in the appendix was mailed. The result of such process was spectacularly good. The process of identifying stakeholders, setting up interviews and interviewing to an extent is quite similar to detective work. Hence, the quality of explorative study is affected by researcher’s detective abilities.
Until the very end, factual data on e-PP in Denmark was hard to get. Stakeholders were simply not interested in sharing hard transaction data. Had the author gotten access to hard data, the validity of this study would have enhanced; stakeholders could have been identified for precise reasons. However, seeking hard data is utopian. The stakeholders
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
73
explained that they are able to participate only to limited extent for practical reasons. The author is grateful for their participation. A researcher could get access to hard data if he or she establishes cooperation with a critical stakeholder (ministry or e-PP network provider). A drawback of such approach is that the other stakeholders involved would be suspicious of researcher’s intention. Hence, they might not be honest in their participation.
Research ethics: Is it ethical for researcher to cooperate with a stakeholder type? Results of this study show that the diffusion of e-PP is mediated by debates among stakeholder types. Each stakeholder type involved in the debates provides valid explanation for supporting or resisting e-PP. Researcher by explaining support or resistance towards e-PP risks altering the nature of the debates. This alteration could adversely affect one or more stakeholder types. For example, research that explains resistance exhibited by end users against centralized procurement could be used by organization level decision makes to devise a strategy for centralizing procurement. Should one or more stakeholder types benefit at the expense of other stakeholder types from an academic study? Those working higher in the hierarchy are better positioned to fund academic research. Should academic research support those higher in the hierarchy enhance their power? A decision maker in local governmental organization brought up the previous question when interviewed. She cautioned the author against unethical behavior.
Sincere efforts were made in this study to maintain neutrality. Debate format is used to report the study’s results as it allowed for presenting multiple stakeholders’ perspectives on the diffusion of e-PP. Please refer to chapter five for further details on the debates. Even in reporting this paragraph, the author has consciously avoided using stakeholder names in negative sense. It is better the research community prepares to gradually learn about the diffusion of e-PP.
7.3
Critical mass as the dependent variable! (BLED 2004)
The use of multi-level analysis in studying the diffusion of IOS is recent. Researchers have used Structuration theory (Johnston and Gregor 2000) and pattern analysis
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
74
(Damsgaard and Lyytinen 1998) to guide multi-level study. An approach quite similar to pattern analysis is used in this study. The rationale for selecting pattern analysis type approach is discussed in chapter three. The author while doing field work thought of a new research approach for studying the diffusion of e-market; a networked technology just as e-PP. The research approach is explained in details in one of the five summarized papers (BLED 2004). The BLED 2004 paper is put in perspective in the rest of this section.
As explained in chapter three, the author realized the following from his field study work; one decides to join and be part of an e-PP network, when he or she perceives the network to do well. If one is uncertain about the future prospect of a network, then he or she adopts wait and see position. The author derived from this realization that one’s decision to join an e-PP network is dependent on his or her perception of the network’s future prospect. Network externalities research explains that once a networked technology attains critical mass its future prospect is excellent (Katz and Shapiro 1986; Shapiro and Varian 1998). The notion of “critical mass” however is vaguely understood so far. Markus (1987), page 507 brings to attention the need for “operationalizing and measuring critical mass”. If critical mass can be operationalized, then potential adopters’ can be enquired about their perception of a network’s future prospect; future prospect for networked technology that has attained critical mass is excellent. Findings from this study show that when potential adopter perceives a network to do well, then he or she would join to benefit from being a part of the network… The diffusion of a network can be studied by querying potential adopter population about their perception of the network’s future prospect. However, this logic would work only if critical mass for a network can be operationalized.
Traditionally, researchers have studied one’s decision to adopt a networked innovation; adoption is the dependent variable. A set of factors (i.e. independent variables) are hypothesized to affect one’s decision to adopt. Based on the logic explained in the previous paragraph, the author quips; “what if we regard critical mass the dependent variable?” The unit of analysis is networked innovation. The network in focus is an open
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
75
e-market owned by private consortia. For the network in focus, critical mass is operationalized as follows; potential adopters’ expectation regarding e-market’s profitability is proposed as a measure for critical mass.
Nature of technology
External environment
Capability of organization
E-market /IOS host
Influences via marketing
Unit of analysis
Action
Perceived cost variables
Perceived benefit variables
Outcome A. A typical adoption model with the factor approach* Nature of technology
Action guided by decision rule “act when perceived benefit > perceived cost”
Capability of organization
Number of adopters External environment
Interorganizational environment
Action
Unit of analysis Unit of analysis
Outcome B. The processual approach*
Population characteristics – risk, resource and interest distribution
Critical mass /outcome C. The critical mass approach
* A and B are obtained from Kurnia and Johnston (2000)
Figure 6 – The Critical Mass Research Approach
The proposition is based on the critical mass theory propounded by Oliver et al. (1985). Critical mass theory has its origins in the logic of collective action (Olsen 1965). Markus (1987) has applied the theory to explain the diffusion of interactive media. In the summarized paper, the author extends knowledge by operationalizing “critical mass”. Please refer to the full paper to learn about logic underlying the operationalization. A research approach developed based on such operationalization is depicted in figure six. Please refer to the BLED paper to learn about the research approach. The BLED paper titled “Operationalizing Critical Mass as the Dependent Variable for Researching the Diffusion of e-Marketplaces – its Implications” is added to the appendix. The logic underlying the critical mass research approach is used to develop a game. The game explains the dynamics underlying the emergence of an e-market. Please refer to the appendix for details about the game.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
76
8
THESIS AS A SUMMARY OF REFEREED PUBLICATIONS
8.1
Overview
This document is prepared as a summary of five referred publications. The contribution made by each of the five papers towards this study is explained in earlier chapters. In this final chapter, coherency among the five papers is brought to attention. An overview of contributions made by the papers for answering the research questions is provided while highlighting the coherency.
8.2
Summarized papers as a coherent unit
It is common practice in summary style thesis to explain the coherence of summarized papers. Typically, four or more papers summarized in a thesis are positioned within a 2x2 matrix. An example of dimensions that define the matrix is normative-descriptive and theoretical approach-empirical approach (Karlsbjerg 2004). Coherence among the five papers summarized in this thesis is hereby explained.
The author documented his research efforts as refereed publications during the course of the study. For example, result of his literature review work is published in HICSS 2003 and his data analysis work in ECIS 2004. The author sought after opportunities to publish the results of his work as they happened. The five papers summarized in this document resulted at the end of the study. As the reader might have noticed, the papers were referred to in the document and not just summarized.
The coherence of summarized papers is explained using a research phase based classification. In the process, an overview of contributions made by the papers for answering the research questions is provided. The five summarized papers are classified under three phases;
i)
research preparation (sub divisions: literature review and research philosophy)
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
77
ii)
data analysis (sub divisions: analytical and prescriptive)
iii)
research direction
In figure seven, the refereed publications are explained as a coherent unit.
Research preparation phase: The HICSS 2003 and ECIS 2003 papers resulted during the research design phase. The author reviewed IOS adoption literatures while preparing research design. The result of the literature review work is published as co-authored paper in HICSS 2003. Please refer to chapter two for further details on the HICCS paper. To determine the study’s philosophical orientation, the author (co) analyzed the application of research philosophy literatures in the IOS adoption field. The result of the analysis is published at ECIS 2003. Please refer to chapter three for further details on the ECIS paper. The phase lasted between July 2001 and Dec. 2002.
Referred publications as a coherent unit
Data preparation
Data Analysis
Literature review
Research philosophy
Analytical
Prescriptive
HICSS 2003
ECIS 2003
ECIS 2004
ECEG 2004/ EJEG 2005
Research direction
BLED 2004
Figure 7 – Refereed Publications as a Coherent Unit
Data analysis phase: The research design was implemented between Jan. and July 2003. The data resulted from the study was analyzed during autumn 2003. Two papers resulted
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
78
from the analysis are published in ECIS 2004 and ECEG 2004 (now selected for publication in EJEG 2005). Different elements of the case are described in both ECIS and ECEG papers. The description provided in the papers contributes answers to the first of the three research questions which is “which stakeholders are involved in the diffusion of e-PP? What is the nature of their involvement?” The first research question is answered in full in chapter four of this document.
Challenges underlying the diffusion of e-PP in Denmark are analyzed in the ECIS 2004 paper. The analysis is presented as debates on centralization vs. decentralization. The author contends that the debates originate due to differences in perception about the extent to which government is single organization. The ECIS 2004 paper answers the second of the three questions; “what challenges are encountered in implementing e-PP? Why do these challenges arise?” Please refer to chapter five in this document for further details on the ECIS paper.
The author reflects upon the role of policy maker in the diffusion of e-PP in the ECEG 2004 paper. The reflections result in six policy recommendations (prescriptions). The recommendations are valid to all nations implementing e-PP. The ECEG paper answers the third research question which is “what role should policy maker play with respect to the diffusion of e-PP?” The six recommendations are listed in chapter six of this document.
Research direction phase: The author reflects upon his experiences in implementing this study to suggest research directions. A research approach for studying the diffusion of emarket is one such suggestion resulted from such reflection. The approach is developed based on critical mass theory propounded by Oliver et al. (1985). The term critical mass for a type of e-market is operationalized as follows; potential adopters’ expectation regarding e-market’s profitability is proposed as a measure for critical mass. The logic underlying the operationalization is explained in the BLED 2004 paper. Further details about the BLED paper are provided in the previous chapter.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
79
9
REFERENCED LITERATURES
1
Accenture (2003) "E-Government Leadership - Engaging the Customer".
2
Aldrich, D., Bertot, J.C., and McClure, C.R. (2002) "E-Government - Initiatives, Developments and Issues", Government Information Quarterly, 19, pp 349-355.
3
Anderson, K.V., Juul, N.C., and Pedersen, J.K. (2003) "Fractional Institutional Endeavors and E-Procurement in Local Government", 16th Bled Electronic Commerce Conference, Bled, Slovenia.
4
Anonymous (2003a) "60 Million Kr Til E-Handel Plads," in: Fyens Stifstidende.
5
Anonymous (2003b) "Elektronisk Indkøb Portal Giver Overskud I 2005," in: Erhvervs bladet.
6
Anonymous (2003c) "Odder Rykker Front På Nettet," in: Computer world.
7
Anonymous (2004) "E-Handel På Gatetrade Firedoblet På Et År," in: Erhvervs Bladet.
8
Arent, J. (2000) "Normative Software Process Improvement," in: Department of Computer Science, Aalborg University, Aalborg.
9
Bakos, J.Y. (1991) "A Strategic Analysis of Electronic Marketplaces", MIS Quarterly, 15 3, pp 295-310.
10
Bannister, F. (2003) "Turf Wars - the Hidden Menace", European Conference on Electronic Government, Trinity College, Dublin, Ireland.
11
Baskerville, R.L., and Myers, M.D. (2002) "Information Systems as a Reference Discipline", MIS Quarterly, 26 1, pp 1-14.
12
Bennett, C., and Timbrell, G.T. (2000) "Application Service Providers - Will They Succeed?" Information Systems Frontiers, 2 2, pp 195-211.
13
Boer, L.D., Harink, J., and Heijboer, G. (2002) "A Conceptual Model for Assessing the Impact of Electronic Procurement", European Journal of Purchase & Supply Management, 8.
14
Byatt, I. (2001) "A Review of Local Government Procurement in England - Main Report".
15
Chandler, S., and Emanuels, S. (2002) "Transformation Not Automation", European Conference on Electronic Government, St Catherine's College, Oxford, United Kingdom.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
80
16
Chwelos, P., Benbasat, I., and Dexter, A.S. (2001) "Empirical Test of an Edi Adoption Model", Information Systems Research, 12 3, pp 304-321.
17
Clark, M. (2003) "E-Procurement Could Save Eur 1 Billion," in: Electronicnews.net.
18
Coulthard, D., and Castleman, T. (2001) "Electronic Procurement in Government - More Complicated Than Just Good Business", European Conference on Information Systems, Bled, Slovenia.
19
Damsgaard, J., and Lyytinen, K. (1998) "Contours of Diffusion of Electronic Data Interchange in Finland; Overcoming Technological Barriers and Collaborating to Make It Happen", Journal of Strategic Information Systems, 7 4, pp 275-297.
20
Damsgaard, J., and Truex, D. (2000) "Binary Trading Relations and the Limits of Edi Standards: The Procrustean Bed of Standards", European Journal of Information Systems, 9, pp 173-188.
21
Davila, A., Gupta, M., and Palmer, R. (2003) "Moving Procurement Systems to the Internet - the Adoption and Use of E-Procurement Models", European Management Journal, 21 1, pp 11-23.
22
Devadoss, P.R., Pan, S.L., and Huang, K.C. (2002) "Structurational Analysis of E-Government Initiative: A Case Study of Sco", Decision Support Systems, 34.
23
DiMaggio, P.J., and Powell, W.W. (1983) "The Iron Cage Revisited - Institutional Isomorphism and Collective Rationality in Organizational Fields", American Sociological Review, 48 2, pp 147-160.
24
Ebrahim, Z., and Irani, Z. (2003) "E-Government Adoption - Analysis of Adoption Staged Models", European Conference on Electronic Government, Trinity College, Dublin, Ireland.
25
Eisenhardt, K.M. (1989) "Building Theories from Case Study Research", Academy of Management Review, 14, pp 532-550.
26
E-jump (2000) "E-Springet: Et Debatoplæg Om Elektronisk Indkøb I De Offentlige Virksomheder [the E-Jump: A Discussion Paper on Electronic Procurement in Public Institutions]", The Ministry of Science, Technology and Innovation, Denmark.
27
EU (1996) "Public Procurement in the European Union - Exploring the Way Forward", Commission of the European Communities, Brussels.
28
EU (2000) "Public Procurement - Meps Start to Debate Consolidated Legislative Package".
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
81
29
EU-Commission (2004) "On Public-Private Partnerships and Community Law on Public Contracts and Concessions", The European Union, Brussels.
30
EURIM (2000) "A Shock to the System - Joined up Electronic Government", EURIM.
31
Falconer, D.J., and Mackay, D.R. (1999) "Ontological Problems of Pluralist Research Methodologies", Proceedings of AIS AMCIS 99: 1999 Americas Conference on Information Systems, Association of Information Systems, pp. 624-626.
32
Ferguson, M., and Baron, F. (2002) "Local E-Government - Perspectives on Strategy and Implementation Worldwide", European Conference on Electronic Government, St Catherine's College, Oxford, United Kingdom.
33
Fichman, R.G., and Kemerer, C.F. (1999) "The Illusory Diffusion of Innovation: An Examination of Assimilation Gaps", Information Systems Research, 10 3, pp 255-275.
34
Gebauer, J. (1998) "The Use of Emerging Technologies in Procurement - State of the Art and a Look into the Future", 9th International Conference of the Information Resources Management Association (IRMA).
35
Gordon, M.L., and Walsh, T.P. (1997) "Outsourcing Technology in Government: Owned, Controlled, or Regulated Institutions", Journal of Government Information, 24 4, pp 267 - 283.
36
Granovetter, M. (1978) "Threshold Models of Collective Behavior", American Journal of Sociology, 83 (May), pp 1420-1443.
37
Gregor, S., and Johnston, R.B. (2000) "Developing an Understanding of InterOrganizational Systems: Arguments for Multi Level Analysis and Structuration Theory", Proceedings of the 8th European Conference on Information Systems.
38
Grover, V. (1993) "An Empirically Derived Model for the Adoption of CustomerBased Inter-Organizational Systems", Decision Sciences, 24 3.
39
Gurbaxani, V., and Whang, S. (1991) "The Impact of Information Systems on Organizations and Markets", Communications of the ACM, 34 1.
40
Hart, P., and Saunders, C. (1997) "Power and Trust: Critical Factors in the Adoption and Use of Electronic Data Interchange", Organization Science, 8 1.
41
Heneriksen, H.Z., Mahnke, V., and Hansen, J.M. (2004) "Public E-Procurement Adoption: Economic and Political Rationality", Proceedings of the 37th Hawaii International Conference on System Sciences.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
82
42
Iacovou, C.L., Benbasat, I., and Dexter, A.S. (1995) "Electronic Data Interchange and Small Organizations: Adoption and Impact If Technology", MIS Quarterly, pp 465-485.
43
IDA.E-Procurement News, from http://europa.eu.int/ida/en/document/2084/17
44
Jarillo, J.C. (1988) "On Strategic Networks", Strategic Management Journal, 9, pp 31-41.
45
Johnston, R.B., and Gregor, S. (2000) "A Theory of Industry-Level Activity for Understanding the Adoption of Inter-Organizational Systems", European Journal of Information Systems, 9, pp 243-251.
46
Jones, D.S. (2002) "Procurement Practices in the Singapore Civil Service: Balancing Control and Delegation", Journal of Public Procurement, 2 1, pp 29-53.
47
Karlsbjerg, J. (2004) "Standard Information Systems in Organizational Information Infrastructures," in: Department of Computer Science, Aalborg University, Aalborg.
48
Katz, M.L., and Shapiro, C. (1986) "Technology Adoption in the Presence of Network Externalities", The Journal of Political Economy, 94 4, pp 822-841.
49
Kaufman, F. (1966) "Data Systems That Cross Company Boundaries", Harvard Business Review, pp 141-155.
50
Keen, P.G.W. (1980) "Mis Research: Reference Disciplines and a Cumulative Tradition", Proceedings of the First International Conference on Information Systems, Philadelphia, PA, pp. 9-18.
51
Ketler, K., and Walstrom, J. (1993) "The Outsourcing Decision", International Journal of Information Management, 13 6 (December), pp 449-459.
52
Khan, N., Currie, W.L., Weerakoddy, V., and Desai, B. (2004) "Evaluating Offshore Outsourcing in India - Supplier and Customer Scenarios", Hawaii International Conference on Systems Sciences (HICSS), Hawaii.
53
Kindleberger, C.P. (1983) "Standards as Public, Collective and Private Goods", KYKLOS, 36.
54
King, J., and Lyytinen, K. (2003) "Standard Making - a Critical Research Frontier for Information Systems; Cfp, Misq Special Issue Workshop".
55
King, J.L. (1983) "Centralized Versus Decentralized Computing: Organizational Considerations and Management Options", Computing Surveys, 15 4.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
83
56
King, J.L., Gurbaxani, V., Kraemer, K., McFarlan, W., Raman, and Yap (1994) "Institutional Factors in Information Technology Innovation", Information Systems Research, 5 2.
57
Kjeldskov, J. (2003) "Human-Computer Interaction Design for Emerging Technologies," in: Department of Computer Science, Aalborg University, Aalborg.
58
Klein, H.K., and Myers, M.D. (1999) "A Set of Principles for Conducting and Evaluating Interpretive Field Studies in Information Systems", MIS Quarterly, 23 1, pp 67-93.
59
Kling, R., Kraemer, K.L., and Dedrick, J. (2003) "Introduction to Special Issue on Globalization and E-Commerce", The Information Society, 19 1-3.
60
Koh, C.E., and Prybutok, V.R. (2003) "The Three Ring Model and the Development of an Instrument for Measuring Dimensions of E-Government Functions", Journal of Information Systems, Spring, pp 34-39.
61
Krysiak, M., Tucker, C., Spitzer, D., and Holland, K. (2003) "E-Procurement; State Government Learns from the Private Sector", in: Digital Government; Principles and Best Practices, A. Pavlichev (ed.), Idea Group Inc., Hershey PA, USA.
62
Kuhn, T. (1962) "The Structure of Scientific Revolutions", University of Chicago Press.
63
Kumar, K., Dissel, V.H.G., and Bielli, P. (1998) "The Merchant of Prato Revisited: Toward a Third Rationality of Information Systems", MIS Quarterly, 22 2, pp 199-226.
64
Kurnia, S., and Johnston, R.B. (2000) "The Need for a Processual View of InterOrganizational Systems Adoption", Journal of Strategic Information Systems, 9, pp 295-319.
65
Kurnia, S., and Johnston, R.B. (2001) "Adoption of Efficient Consumer Response: The Issue of Mutuality", Supply Chain Management: An International Journal, 6 5, pp 230-241.
66
Kurnia, S., and Johnston, R.B. (2002) "A Review of Approaches to Ec-Enabled Ios Adoption Studies", Proceedings of the 35th Annual Hawaii International Conference on System Sciences, Hawaii.
67
Lacity, M. (1995) "It Outsourcing: Maximize Flexibility and Control", Harvard Business Review, May-June, pp 84-93.
68
Lamsdorff, J.G. (2002). Corruption Perceptions Index. Retrieved April 4, 2004, from
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
84
http://www.globalcorruptionreport.org/download/gcr2003/24_Data_and_research. pdf 69
Layne, K., and Lee, J. (2001) "Developing a Fully Functional E-Government Model - a Four Stage Model", Government Information Quarterly, 18, pp 122136.
70
Lee, A.S. (1991) "Integrating Positivist and Interpretive Approaches to Organizational Research", Organization Science, 2 4, pp 342-365.
71
Lewis, W.L., and Grimes, A.J. (1999) "Metatriangulation: Building Theory from Multiple Paradigms", Academy of Management Review, 24 4.
72
Li, F. (2003) "Implementing E-Government Strategy in Scotland - Current Situation and Emerging Issues", Journal of Electronic Commerce in Organizations, 1 2, pp 44-65.
73
MacManus, S. (2002) "Understanding the Incremental Nature of E-Procurement Implementation at the State and Local Levels", Journal of Public Procurement, 2 1, pp 5-28.
74
Maniatopoulos, G. (2004) "E-Procurement Technologies and Institutional Change", European Conference on Electronic Government, Dublin.
75
Marche, S., and McNiven, J.D. (2003) "E-Government and E-Governance: The Future Isn't What It Used to Be", Canadian Journal of Administrative SciencesRevue Canadienne Des Sciences De L Administration, 20 1, pp 74-86.
76
Markus, L.M. (1983) "Power, Politics and Mis Implementation", Communications of the ACM, 26 6, pp 430-444.
77
Markus, L.M. (1987) "Toward a "Critical Mass" Theory of Interactive Media", Communication Research, 14 5, pp 491-511.
78
Markus, M.L., and Robey, D. (1988) "Information Technology and Organizational Change: Causal Structure in Theory and Research", Management Science, 34 5, pp 583-598.
79
McCue, C.P., and Pitzer, J.T. (2000) "Centralized Vs. Decentralized Purchasing: Current Trends in Governmental Procurement Practices", Journal of Public Budgeting, Accounting & Financial Management, 12 3.
80
Mehta, P.S. (2001) "Govt Procurement Agreement: A Chance to Reduce Corruption," in: Financial express.
81
Mingers, J. (2001) "Combining Is Research Methods: Towards a Pluralist Methodology", Information Systems Research, 12 3, pp 240-259.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
85
82
Moon, M.J. (2002) "The Evolution of E-Government among Municipalities Rhetoric or Reality?" Public Administration Review, 62 4, pp 424 - 433.
83
Moriarty, R., and Bateson, J. (1982) "Exploring Complex Decision Units: A New Approach", Journal of Marketing Research, 19 May.
84
Neef, D. (2001) "E-Procurement - from Strategy to Implementation", Prentice Hall.
85
NIGP (1996) "National Association of State Purchasing Officials/National Association of State Information Resource Executives Joint Force on Information Technology and Procurement Reform".
86
Oliver, P., Marwell, G., and Teixeira, R. (1985) "A Theory of the Critical Mass. I. Interdependence, Group Heterogeneity and the Production of Collective Action", American Journal of Sociology, 91 3, pp 522-556.
87
Olsen, M. (1965) "The Logic of Collective Action: Public Goods and the Theory of Groups", Harvard University Press, Cambridge.
88
Orlikowski, W.J., and Baroudi, J. (1991) "Studying Information Technology in Organizations: Research Approaches and Assumptions", Information Systems Research, 2 1, pp 1-28.
89
Pegnato, J.A. (2003) "Assessing Federal Procurement Reform: Has the Procurement Pendulum Stopped Swinging", Journal of Public Procurement, 3 2.
90
Porter, M.E. (1998) "Clusters and the New Economics of Competition", Harvard Business Review, November-December, pp 77-90.
91
Powell, D. (1993) "To Outsource or Not to Outsource?" Networking Management, 11 2 (February), pp 56-61.
92
Premkumar, G., and Ramamurthy, K. (1994) "Implementation of Electronic Data Interchange: An Innovation Diffusion Perspective", Journal of Management Information Systems, 11 2.
93
Rajkumar, T.M., and Mani, R.V.S. (2001) "Offshore Software Development - the View from Indian Suppliers", Information Systems Management, Spring.
94
Ramanathan, S. (2003) "Electronic Market as a Collective Action: A Research Framework for Studying Critical Mass", Research Symposium on Emerging Electronic Markets, Bremen.
95
Reddick, C.G. (2004) "A Two-Stage Model of E-Government Growth: Theories and Empirical Evidence for U.S. Cities", Government Information Quarterly, 21, pp 51-64.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
86
96
Reimers, K. (2002) "The Unit of Analysis in E-Commerce Studies", Tsinghua University, Beijing.
97
Reimers, K., Johnston, R.B., and Klein, S. (2004) "The Shaping of InterOrganizational Information Systems - Main Design Consideration of an International Comparative Project", Bled E-commerce Conference, Bled, Slovenia.
98
Rogers, E.M. (1976) "New Product Adoption and Diffusion", The Journal of Consumer Research, 2 4, pp 290-301.
99
Ronaghan, S.A. (2002) "Benchmarking E-Government - a Global Perspective", The United Nations and the ASPA.
100
Schedler, K., and Scharf, C. (2001) "Exploring the Interrelations between Electronic Government and the New Public Management", 13E Conference, Zurich.
101
Scheepers, R. (1999) "Intranet Implementation: Influences, Challenges and Role Players," in: Department of Computer Science, Aalborg University, Aalborg.
102
Seddon, P. (2004). Research Topic: It Outsourcing. Retrieved 20th August, 2004, from http://www.dis.unimelb.edu.au/staff/peter/research/IToutsourcing/DIS_ITOResea rchInterestPage.htm
103
Shapiro, C., and Varian, H. (1998) "Information Rules: A Strategic Guide to the Network Economy", Harvard Business School Press.
104
Sharma, S., and Gupta, J. (2002) "Transforming to E-Government - a Framework", European Conference on Electronic Government, St Catherine's College, Oxford, United Kingdom.
105
Shin, G.-C., and Park, M.S. (2004) "Learning from the Award-Winning Korean Public Procurement Service", 89th Annual International Supply Management Conference.
106
Silcock, R. (2001) "What Is E-Government?" Hansard Society of Parliamentary Government, 54, pp 88 - 101.
107
Sørensen, C. (1993) "Introducing Case Tools into Software Organizations," in: Institute for Electronic Systems, Aalborg Universoty, Aalborg.
108
Strejcek, G., and Theil, M. (2003) "Technology Push, Legislation Pull? EGovernment in the European Union", Decision Support Systems, 34 3, pp 305-313.
109
Thai, K.V., and Grimm, R. (2000) "Government Procurement: Past and Current Developments", Journal of Public Budgeting, Accounting & Financial Management, 12 2, pp 231 - 247.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
87
110
Tonkin, C. (2003) "E-Procurement in the Public Sector; Story, Myth and Legend", The Policy Institute, Trinity College, Dublin.
111
Tuunainen, V.K. (1999) "Different Models of Electronic Commerce," in: Helsinki School of Economics and Business Administration, Helsinki.
112
Tyndale, P. (2002) "Will E-Government Succeed?" European Conference on Electronic Government, St Catherine's College, Oxford, United Kingdom.
113
Ulstrup, L. (2001) "The Case for Federal E-Procurement Translating Private Sector Benefits to the Federal Government", American Management Systems.
114
Univac. Retrieved 11th October, 2004, from http://wwwcsif.cs.ucdavis.edu/~csclub/museum/items/univac.html
115
Vaidya, K., and Callender, G. (2004) "Implementing E-Procurement Initiatives Critical Failure Factors and Lessons Learnt across the Public Sector", IAWTIC.
116
Willcocks, L.P., Currie, W., and Jackson, S. (1997) "In Pursuit of Re-Engineering in Public Administration", Public Administration, 75 Winter, pp 671-649.
117
Williamson, O.E. (1981) "The Economics of Organization - the Transaction Cost Approach", American Journal of Sociology, 87 3, pp 548-577.
118
Yin, R.K. (1989) "Case Study Research: Design and Methods", 2nd ed Sage Publications, Thousand Oaks, California.
119
Zulfiqar, K.A., Pan, S.L., Lee, J.N., and Huang, J.C. (2001) "E-Government: An Exploratory Study of on-Line Electronic Procurement Systems [Case Study]", The 9th European Conference on Information Systems, Bled, Slovenia.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
88
10 LIST OF FIGURES Figure 1 – Government as a Multi-Level Organization Figure 2 – Impact of the Internet on Government (Marche and McNiven 2003 Adapted) Figure 3 – A Framework for Classifying e-PP Implementation Approaches Figure 4 – Graphical Depiction of Growth in Positivist and Interpretive Studies in the IOS Adoption Field Figure 5 – The Danish Nation as the Embedded Unit of Analysis Figure 6 – The Critical Mass Research Approach Figure 7 – Refereed Publications as a Coherent Unit
11 LIST OF TABLES
Table 1 – List of Commodities for e-Procurement; Heneriksen et al. (2004) translation Used Table 2 – Refereed Publications Summarized for this Thesis Table 3 – Categorization of Reviewed IOS Adoption Articles by Research Philosophy
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
89
12 APPENDIX
12.1 Research publication one – the HICSS 2003 12.2 Research publication two – the ECIS 2003 12.3 Research publication three – the ECIS 2004 12.4 Research publication four – the ECEG 2004 and EJEG 2005 12.5 Research publication five – the BLED 2004 12.6 IFIP 9.4 newsletter 12.7 Interview data 12.8 Interview questionnaire 12.9 Critical mass game 12.10
A sample transcribed interview (only provided to examiners)
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
90
Appendix 12.1 Research Publication One – the HICSS 2003 Rationalizing, Probing, Understanding: the Evolution of the Inter Organizational Systems (IOS) Adoption Field By Ramanathan Somasundaram and Jeremy Rose
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Rationalizing, Probing, Understanding: the Evolution of the Inter-Organizational Systems Adoption Field Somasundaram Ramanathan and Jeremy Rose Department of Computer Science, Aalborg University Frederik Bajers Vej 7, 9220 Aalborg, Denmark.
[email protected] and
[email protected] principle1: 20% of business partners (accounting for 80% of transactions) adopted EDI, while the rest did not.
Abstract Previous surveys of the Inter-Organizational Systems (IOS) adoption field have developed taxonomic schemes to improve general understanding of the field. This paper instead studies the historical development of the field and describes its evolution in a stage model. In the rationalizing stage researchers tried to discover and communicate reasons for some apparently successful adoptions of IOS. However it soon became clear that IOS adoption was relatively limited and in the probing stage the focus changed to investigating the failure of the promised seamless integration. No unified coherent theoretical account emerged from these studies, so in the understanding stage researchers widened the focus of study to try to develop a richer account of the many complex influences on IOS adoption. The three stages are associated with different objectives, assumptions, theoretical backgrounds and research approaches. Key words: Inter-Organizational Systems adoption, review, research approach and application.
1
(IOS), theory
Introduction
Interest in the adoption of IOS was initially fuelled by a number of remarkable success stories. Merrill Lynch linked to its customer firms by introducing its Cash Management Accounts (CMA) system and was able to increase its customer base substantially, retaining its competitive advantage for some time [7] [14]. Other examples included American Hospital Supplies and the SABRE reservation system. However, both recipes for commercial success and the dream of seamless information integration proved elusive. Many electronic data interchange (EDI) systems failed to fully automate inter-organizational transactions as anticipated. They were expensive to install and often forced on small supplier firms by large conglomerates. Johnston and Mak [38] showed that EDI adoption followed the Pareto
Faced with variable IOS adoption success, researchers have adopted a wide range of approaches, methods and theories to understand and explain phenomena, and to advocate courses of action. Theories appropriated for these purposes include: industrial economic theories [17], transaction cost theory [33], critical mass theory [9], resource dependency theory [60], diffusion of innovation theory [55] and structuration theory [37]. As Astley and Van de Van [1] point out, the use of different theories may result in different explanations for the same phenomena. Various research methods have also been employed. For instance, Grover and Goslar [30] use a factor approach while Damsgaard and Lyytinen [22] use a processual approach. The objectives of the researchers have also been different. When is IOS adopted, in what circumstances does it provide competitive advantage and why [18]? How can we explain the adoption of IOS in certain industries and not in the others? [37]? There are a number of existing surveys of the IOS adoption field. Kurnia and Johnston [41] reviewed the research methods used, Damsgaard and Lyttinen [22] investigated the field’s use of micro, meso and macro levels of analysis, Kumar et al. [40] classified research studies as technical/economical or socio-political and Prescott and Conger [62] analyzed the uses of diffusion of innovation (DOI) theory based on their IT locus of impact. These analyses focus on the epistemological (process of knowledge creation) aspects and provide taxonomic schemes which increase our general understanding of the field. However there are two reasons why a historical review of the field is justified at this juncture. The first is that the IOS field has been studied intensively for almost 20 years, and so is old enough to have a history. The second is that, with the emergence of the Internet and internet standards such as 1
The term”Pareto Principle” is often used to describe this occurrence. Pareto, an Italian Economist observed that 20% of the Italian people owned 80% of the country’s assets [52].
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
2
Sample selection and classification
Webster and Watson [71] suggest that literature reviews are an important part of the development of the IS field. They offer the opportunity to synthesize and reflect on previous theoretical work, thus providing secure grounding for the advancement of knowledge. They suggest that the elements of a good literature review include a structured approach to identifying the source material and the use of a concept matrix or other analytical framework leading to ‘a coherent conceptual structuring of the topic’. Fig. 1.0: Stage wise classification of reviewed papers Understanding, 17, 24%
Rationalizing, 25, 35%
The analysis framework used is reflected in the structure of the paper. For each stage we analyzed the basic assumptions held, the use of theory, the choice of research methods and the limitations of the approach. The results from the three stage analyses are used for discussing the evolution of IOS adoption field.
Fig. 2.0: Stage wise evolution No. of papers
XML, net-based IOS are being adopted even by those organizations who traditionally resisted the EDI based system [50]. Researchers studying the adoption of such net-based IOS should have the opportunity to learn from the experience of the earlier related studies. By undertaking a historical survey of the field we aim to help researchers understand i) predominant research perspectives ii) the pattern of development iii) current trends and iv) contradictions and problems in the IOS adoption field.
20 15
15
10 5 0
8 2 0 upto 1980
0 81-85
10
9 5 0 86-90
16
5 1 91-95
1 96 - 02
Years Rationalizing
Probing
Understanding
The literature was structurally conceptualized as a stage model with three stages: rationalizing, probing and understanding. The titles are intended to reflect the character of the analysis carried out. Of the 72 contributions, 25 were classified under the rationalizing stage, 30 under the probing stage and 17 under the understanding stage (as depicted in figure 1.0). Though the three stages are basically sequential there is some overlap. This is graphically depicted in figure 2.0
Probing, 30, 41%
3 Rationalizing Stage 3.1 Overview of Stage It is common for researchers to select their literature sample for conducting a field review from a few well regarded journals (e.g. [57], [67] and [12]). Though this approach clearly defines the boundary of the study, it may leave out critical contributions published in other journals, conferences and book chapters. Moreover this kind of study is better suited to provide a snapshot of the reviewed field than to depict developments in an evolutionary manner. We used the three stage search strategy proposed by Webster and Watson (leading journals search, backwards tracing of citations, forward citation search) to identify 72 important contributions to the IOS adoption field. Our objective in selecting contributions was to follow the emergence and development of the IOS adoption concept.
The IOS field began to emerge as part of the better established IT and corporate strategy field. Adoption was at first taken for granted, and IOS were viewed as a means for achieving competitive advantage. There were well-reported successes at companies such as Merrill Lynch, American Hospital Supply, Digital Equipment Corporation and USA Today [7]. Researchers began to seek explanations for these successes, and on the basis of their explanations, provide advice to managers. Although these early explanations were informative, their role in the cumulative development of knowledge is limited. Bakos and Treacy [3] and Clemons and Kimbrough [18] attempt to remedy this limitation by using theories for generating causal relationships. These analyses form the basis for much of the formal modeling work [16], [70], and [5] that follows [17].
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
3.2
Assumptions held
The contexts in which companies are operating are assumed to be competitive. A company’s trading partners are assumed to be part of the external environment. The use of information technology to connect a company with its trading partners is expected to bring not just competitive, but also strategic benefits. Causal agency is attributed to a “technology imperative” [47]. To obtain competitive advantage, opportunities for implementing IOS should be identified. Theoretical assumptions governing these analyses are rooted in economics, from industrial organization and institutional economics studies [2].
3.3
Research Approach
Clemons and Kleindorfer [16] summarize the evolution in research approaches in this stage. In the beginning the literature was ‘largely anecdotal, recounting interesting examples of strategic systems and offering them as exemplars of corporate competition’. According to Clemons and Kimbrough [18] and Clemons and McFarlan [19] the analysis was based on a few overworked successes and hardly included failure. McFarlan [48] uses Porter’s five forces to discuss the opportunities that IT provides and relates the discussion to his newlydeveloped framework. Later ‘verbal models based on relevant theory or reasoning from detailed case studies’ are developed. An example of this is the work of Clemons and Kimbrough [18]. They hypothesized three IS characteristics that are necessary for providing sustainable strategic advantage for a firm. Towards the end of this stage the analysis of anecdotal information is used ‘to construct analytic models of strategic systems and reason formally from these models’. This formal modeling identifies the assumptions behind a defined scenario, defines notations for attributes, develops theoretical frameworks and provides proof using economic theories. Here adoption is taken less for granted, and studied directly. For instance, Clemons and Kleindorfer [16] modeled the circumstances under which an ‘efficient investment in IOS would be undertaken’ and Barua and Lee [5] investigate strategies for facilitating adoption of EDI. The effectiveness of such analysis is questionable because it is difficult to model the ‘dauntingly’ complex inter-organizational context. Those supporting such modeling techniques would, however, argue for more rigorous models.
3.4
Porter’s generic strategy classification [58], [48], [10] and [14]. The five forces framework is used to analyze the potential of IOS in increasing a firm’s power over its trading partners (customers and suppliers), in increasing the entry barriers and in decreasing the threat from substitutes, thereby changing its competitive position in the industry to its advantage. The generic strategy classification is used to study the role of IOS in a firm’s positioning strategy. The frequent use of Porter’s framework may have been due to its simple conceptualization of the external environment and the proposal of a limited number of ways to deal with it. However, these analyses tend to gloss over the many complexities of the adoption process.
Analyzing the application of theories
Porter’s thinking about ‘industrial organization’ predominantly influences this stage. The most commonly used framework for analysis in this stage is Porter’s five forces [58], [48], [36], [10], [3] followed by
Reviewed papers in the rationalizing stage
1. Kaufmann 2. Porter 3. Barrett and Konsynski 4. Parsons 5. McFarlan 6. Notowidigdo 7. Benjamin et al. 8. Ives and Learmonth 9. Cash and Konsynski 10. Porter and Millar 11. Clemons and McFarlan 12. Bakos and Treacy 13. Clemons and Kimbrough 14. Clemons 15. Clemons and Row 16. Malone et al. 17. Johnston and Lawrence 18. Suomi 19. Johnston and Vitale 20. Clemons and Row 21. Clemons and Kleindorfer 22. Bakos et al. 23. Clemons at al. 24. Wang and Seidmann 25. Barua and Lee
- Jan./ Feb. 1966 - March/April 1979 - Dec. 1982 - Fall 1983 - May/June 1984 - Feb. 1984 - Spring 1984 - Dec. 1984 - Mar./Apr. 1985 - Dec. 1985 - July/Aug. 1986 - June 1986 - Dec. 1986 - Issue 11, 1986 - Dec. 1987 - vol. 6, 1987 - July/Aug. 1988 - Issue 15, 1988 - June 1988 - Sep. 1991 - vol. 8, 1992 - vol. 2, 1993 - vol. 2, 1993 - March 1995 - Dec. 1997
- Harvard Business Review - Harvard Business Review - MIS Quarterly - Sloan Management Review - Harvard Business Review - Financial Executive - Sloan Management Review - Communications of the ACM - Harvard Business Review - Harvard Business Review - Harvard Business Review - MIS Quarterly - ICIS - Information and Management - ICIS. - C. of the ACM - Harvard Business Review - Information and Management - MIS Quarterly - MIS Quarterly - Decision Support Systems - J. of MIS - J. of MIS - Management Science - IS Research
Under the ‘institutional economics’ category, transaction cost analysis (TCA) is applied. Firms operate with ‘bounded rationality’ and ‘opportunistic behavior’ [2]. TCA takes transaction rather than commodity as the unit of analysis for assessing governance structures (of which markets and hierarchies are the leading alternatives). In conducting a transaction, a firm searches the market to the extent that the cost of additional searching is less than the expected benefit from the search. A firm should decide whether to buy or produce by comparing the co-ordination costs to the production costs. The adoption of IOS is expected to increase transaction efficiency by reducing co-ordination costs and increasing monitoring capabilities, resulting in an overall shift towards the market governance mechanism [51]. Highlighting different factors - the role of incentives and the benefits of long term collaboration – led other commentators to a different conclusion
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
however: the ‘move to the middle’ hypothesis [15] and [4]. Starting with TCA, Bakos and Treacy [3] use bounded rationality and improved production efficiency as factors that form the basis for theories explaining how IT can improve the competitive strategy of a firm.
3.5
efficient manner. The contexts in which the technology is implemented are assumed to be similar. A few researchers such as Bouchard [9] contradict the central assumption by arguing that organizations are acting strategically when deciding not to adopt.
Critique of the stage
Reviewed papers in the Probing stage
Transaction cost analysis is often criticized for assuming competitive and opportunistic behavior, and for not taking co-operative behavior into account. Neither the organizational context in which the adoption actually takes place or the complex and situated nature of the technology is recognized in this stage. Interaction between firms and the environment is assumed to be unidirectional. Rigorous research methods are not always used for studying adoption, making it difficult to assess the validity of findings or derive cumulative conclusions from related research studies.
1. Emmelhainz 2. Monczka and Carter 3. Reich and Benbasat 4. Cooper and Zmud 5. Benjamin et al. 6. Moore and Benbasat 7. Saunders and Clark 8. O’Callaghan et al. 9. Bergeron and Raymond 10. Grover 11. Grover 12. Bouchard 13. Premkumar et al. 14. Boon et al. 15. Cavaye et al. 16. Premkumar et al. 17. Prescott and Conger 18. Iacovou et al. 19. Teo et al. 20. Daugherty et al. 21. Drury and Farhoomand 22. Crum et al. 23. Hart and Saunders 24. Bergeron and Raymond 25. Raymond and Blili 26. Premkumar et al. 27. Mahler and Rogers 28. Heck and Ribbers 29. Chong 30. Chwelos et al.
4 Probing Stage 4.1 Overview of Stage Many of the optimistic predictions for IOS failed to materialize and the rate of adoption was smaller than foreseen. Researchers probed this occurrence with the objective of explaining the circumstances that lead to adoption or non-adoption. Most of the work studied the adoption of Electronic Data Interchange (EDI). The factor approach was the predominant research method. Theories such as diffusion of innovation (DOI) [65], critical mass theory and socio-political theories are used either alone or in combination for identifying the factors. The use of multiple theoretical perspectives for studying the adoption phenomena in different contexts resulted in fragmented and differing explanations. The same factors when tested in different contexts sometimes proved to have contradictory relations [41]. The comprehensiveness of the research frameworks increases as the strand develops. Recently, Ling [45] and Chwelos et al. [13] have suggested that knowledge obtained from studying IOS adoption could be extended for studying the adoption of Internet based systems that interconnect businesses.
4.2
Assumptions held
The buyer-supplier context in which IOS are implemented is recognized. The adoption of technology/innovation is predominantly assumed to be beneficial, with non-adoption implied to be failure. The studies aim to identify a finite number of factors (independent variables) and explain their relations with adoption (the dependent variable). When this relation is well understood and validated, appropriate actions can be taken to achieve the adoption objectives in the most
4.3
- 1988 - Spring 1989 - vol. 1, 1990 - Feb. 1990 - vol. 23 1990 - 2 (3), 1991 - vol. 9, 1992 - Apr. 1992 - Fall 1992 - May/June 1993 - Summer 1993 - Dec. 1993 - Fall 1994 - Dec. 1994 - vol. 1, 1995 - vol. 26 1995 - May/Aug. 1995 - Dec. 1995 - Dec. 1995 - Dec. 1995 - 1996 - Summer 1996 - Jan/Feb. 1997 - vol. 31, 1997 - vol. 3, 1997 - vol. 6, 1997 - 1999 - 1999 - 2000 - vol. 12, 2001
- J. of Business Logistics - Journal of Purchasing and Mgmt. - Information Systems Research - Management Science - Long Range Planning - Information Systems Research - Information R esources mgmt. J. - Journal of Marketing - DataBase - Decision Sciences - Journal of MIS - ICIS - Journal of MIS - ICIS - J. of Strategic Inf. Systems - Decision Sciences - DATA BASE Advances - MIS Quarterly - ICIS - Logistics and transportation review - Information Resources mgmt. J. - Transportation Journal - Organization Science - Information and Management - European J. of P. and Supply Mgmt. - European Journal of Information Systems - Telecommunications Policy - HICCS - PACIS - Information Systems Research
Research approach
By testing a hypothesized relationship between a finite set of factors and adoption, the variance in adoption/nonadoption is explained. Researchers conducted case studies [26], [52], [64], [6] and [33], and/or relied on theories and empirical studies [55], [30], [60], [61], [34] and [13]) for identifying the independent variables. Case study strategies were adopted in the early stages for understanding the broad and complex IOS phenomena [64]. This knowledge was foreseen to contribute to the posing of causal questions which were tested across a large sample. Statistical techniques are used for establishing the validity of the tested relationship. The generalizability of the results is assumed to be broader than the context from which it has arrived. Either a new research model is developed [66], [30], [60], [39], [63], and [33] or existing research frameworks are adapted [20], [59], [34] and [13]. Although researchers recognize, adapt and use research frameworks such as those of O’Callaghan [55], Iacovou et al. [35] and Hart and Saunders [33], there has not been a framework that is recognized as a standard. If it is accepted that different frameworks are required for different purposes, then the interrelations among frameworks should also be understood.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
The dependent variable ’adoption’ operationalized in different ways:
has
been
1. adoption decision (e.g. [66]) 2. extent of adoption (e.g. [25]) 3. critical success factors of implementation (e.g. [11]) 4. adoption of multiple telecommunication technologies (e.g. [30]). There are also differences in choosing the research sample: 1. Industry wise (e.g. [55]) vs. multiple industry wise/geography/society wise (e.g. [26]) vs. network/technology wise (e.g. [66]) 2. Adopters/Implementers (e.g. [6]) vs. non-adopters (e.g. [66]) and in conducting the study: 3. An instance ([60] and [61]) vs. longitudinal [8] vs. case study accounts [52]. There are many possible combinations, but one that is frequently used is: sampling both adopters and non adopters, across multiple industries or a region or a society, using survey at an instance as the research method, for studying the factors that affect adoption decision.
4.4
Analyzing the application of theories
The most frequently used theoretical perspective is diffusion of innovation (DOI). IOS is viewed as an innovation. DOI is the “process by which an innovation is communicated through certain channels over time among the members of the social system” [65]. The theory was originally developed for studying individual adoption decisions concerning well-defined innovations (e.g. television, genetically modified corn seeds) [46]. The review and analysis work of Tornatzky and Klein [69] (innovation characteristics and their effect on innovation adoption/implementation) forms the basis for much of this research. Three innovation characteristics i) relative advantage ii) compatibility and iii) complexity (that Tornatzky and Klein find to have a consistent relationship with innovation adoption) are tested many times in the IOS context [61], [68]. Characteristics identified from this theory are usually hypothesized together with characteristics identified from other theories such as socio-political theories [53] and critical mass theory [9]. In a review of DOI research studies, Prescott and Conger [62] concluded that critical mass and influence theories are relevant for studying IOS adoption, as is adapted DOI theory. However, DOI is
criticized “for trading simplicity and generalizability against accuracy by using simple metaphors of force” [46]. A socio-political perspective is sometimes used for studying IOS adoption. It argues that “firms form interorganizational linkages primarily to gain control over critical resources and thereby reduce uncertainty in their acquisition” [65]. Three forces studied in the IOS context are i) power [33] ii) dependence [33] and iii) trust [54]. IOS are historically implemented in hub and spoke formations involving a large buyer and a number of small suppliers. The hub firm initiates the IOS and exercises its purchasing power to encourage the spoke organizations to transact electronically. The role of this exercising of power in IOS adoption is studied. Power can be exercised coercively and persuasively. Whilst the coercive exercise of power enforces adoption, persuasive exercise is found to enhance trust among suppliers that results in increased usage [33]. Trust or ‘favorite transaction climate’ is assumed to facilitate IOS adoption. Critical mass theory is recognized for its potential to explain the adoption decision process [9], [62] and [49] . It views adoption of IOS as a collective action. The decision to adopt or join the collective action is influenced by the number of people who already have joined the system. Neo et al. [53] find that peer influence is an important factor affecting the timing of Trade net system adoption in Singapore. Bouchard [9] and Mahler and Rogers [49] find that the decision to adopt an IOS is influenced by the adoption decisions of others in the network.
4.5
Critique of the stage
The predominant studying of subjective perceptions/secondary attributes [24] and not just the objective features/primary attributes does not match well with the positivist ideology. Downs and Mohr [24] explain that characteristics of an innovation such as costs, complexity and relative advantage are perceived subjectively. For instance, an investment of 100,000 dollars might be perceived by one organization as expensive, while another might perceive it as trivial. Thus they argue that generalization of relations between innovation characteristics and innovation adoption is a ‘fruitless’ or ‘pre-mature’ activity. As an alternative they suggest the studying of an innovation in a single organization. Tornatzky and Klein [69] address this critique by putting forth a philosophical question about perceptional judgments of an innovation. “In prosaic terms, are we (and our perceptions) are more alike or
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
more different?” They assume that “we” are more alike (in contrast to Downs and Mohr). Hence, they contend that perceived innovation characteristics can predict the adoption and implementation. This stage of the field is based on the ideology of Tornatzky and Klein.
5 Understanding Stage 5.1 Overview of stage Researchers find previous explanations of IOS adoption simplistic and inadequate. The complex nature of IOS is explained and the need to study its adoption in a context that is broader than just the organization is highlighted [22]. The factor approach that has traditionally been used for studying IOS adoption is found unsuitable [42]. Researchers call for (in the terms of Markus and Robey [47]) multi-level analysis using processual approaches whilst recognizing emergent causal agency. The adoption process is explained in terms of patterns ([21] [22] [23]) and analytical descriptions [43]. This stage remains the current trend.
5.2
Assumptions held
IOS are assumed to be complex, learning intensive and socially constructed systems [46]. Instead of regarding adoption as the product of a set of contextual factors, the affect is assumed to be bi-directional in nature (Kurnia and Johnston 2002). Adoption is understood as an on-going process and not as an instance. The adoption is studied within the larger ‘network’ (Cunningham and Tynan 1993) context and not just the organizational or dyadic context.
5.3
Research approach
A process approach is used for studying bi-directional interactions over a period of time across multiple levels of analysis in a network. Case study or action research is used as the research method for obtaining data [41]. Usually the adoption is studied in one or more industries. In-depth case studies or action research usually restrict the study to one industry [29]. The results obtained take the form of a pattern or an analytical description. Patterns are a form of abstraction which can be tested across contexts for validity. Hence, they are generalizable. Analytical descriptions are useful for collating the rich data and understanding its implications. The cumulative nature of analytical descriptions is, however, unclear. According to Kurnia and Johnston [42], factor and process based approaches complement one another. The relationships identified using factor approach are necessary but not sufficient to guarantee success. The process approach is of help in understanding the dynamics of complex phenomena.
However, it is less prescriptive. When the assumptions are explicitly recognized, the approaches can be used in combination.
5.4
Analyzing the application of theories
Structuration theory, which has become popular for studying Information Systems implementation [56], has been adapted for studying IOS adoption [37]. The theory provides the breadth required for studying multiple analysis levels and the interactivity for inter-relating the various levels. According to structuration theory, action is both constrained and enabled by structure (which exists only as memory traces). Gregor and Johnston [28] and Johnston and Gregor [37] operationalized this theory in the IOS context, arguing that the activities of a firm can be both constrained and enabled by industry structure. The structure of an industry can be understood by mapping the relationships between firms. The theory Reviewed papers in the Understanding stage 1. Cunningham and Tynan 2. Damsgaard 3. Damsgard 4. Chan and Swatman 5. Damsgaard and Lyytinen 6. Kumar et al. 7. Mak and Johnston 8. Chan and Swatman 9. Gregor and Menzies 10. Gregor and Johnston 11. Johnston and Gregor 12. Damsgaard and Lyytinen 13. Kurnia and Johnston 14. Kurnia and Johnston 15. Damsgaard and Lyytinen 16. Damsgaard and Lyytinen 17. Kurnia and Johnston
vol. 13 1993 1997 1999 May 1998 1998 June 1998 Jan. 1999 1999 1999 2000 vol. 9, 2000 2000 Dec. 2000 vol. 6, 2001 Jan. 2001 vol 17, 2001 Jan. 2002
I. J. of Information Management HICCS Book BLED Intl. Conf. on EC. J. of Strategic Information Systems MIS Quarterly HICCS ECIS Collaborative E.C. tech. and research. ECIS E. Journal of Information Systems Book J. of Strategic Information Systems Supply Chain Management IFIP 8.6 The Information Society HICCS
is mainly used for understanding the adoption of IOS at an industry level. This theory development is influenced by the work of Damsgaard and Lyytinen ([21] [22]), who identify the need for multi-level analysis, processual approach and emergent causal agency. They use a broader multi theoretical perspective which they classify along the dimensions of i) micro ii) meso and iii) macro (denoting individual/organizational, inter-organizational and institutional contexts respectively). Network theory, which focuses on buyer-seller relationships and interaction, has also been used in the IOS adoption field. Like structuration theory, it takes into account the environment in which the interaction
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
takes place and reciprocal effects. It also characterizes the exchange episodes which the relationship is based on.
this stage ([44], [27], [42]). A summary of the three stage analyses is presented in table 1.0
5.5
Although widely-accepted IOS adoption theory has still to emerge, there is a cumulative widening of knowledge, as long as the reader is prepared to accept the many inconsistencies, differences of standpoint and contradictions. Differing perspectives on IOS adoption have steadily been added to the literature. To the initial relatively simple rationalizations (based on short case studies and interviews with executives) was added the recognition of the organizational context in which adoption takes place. Next followed the inclusion of socio-political issues such as power and trust [33] and critical mass theory [9] in the probing stage. This was further complemented by the recognition of the impact of the larger context within which the adoption takes place [22]. Each change of perspective serves to add increasingly better structured detail to earlier perspectives. Chwelos [13], for instance, develops a parsimonious model of EDI adoption based on previous results. Those in the understanding stage inter-relate many of the established findings and provide a more inclusive picture of adoption.
Critique of the stage
This approach, despite providing deeper understanding, contributes in a limited way to cumulative knowledge development. Kurnia and Johnston [41] recognize this limitation which they term “reduced ability to make general statements”. The knowledge has to be abstracted at some level and the relationships defined. This approach is useful to explain what happened, how and why it happened but it is of limited use to suggest how it should happen. The prescriptive “should” is nevertheless important because it indicates learning. The application of this learning can result in improved efficiency and effectiveness.
6
Discussion and Conclusion
The IOS adoption field emerged in the early 80’s from the discussions about the role of IT as a strategic tool. Then the discussions focused on the new inter organizational nature of IT. Researchers rationalized the role of inter organizational systems on the basis of a few apparent successes. Such rationalizations were made to provide guidance to practicing managers. Researchers such as Bakos and Treacy [3] and Clemons and Kimbrough [18] called for a theory based cumulative research tradition highlighting the anecdotal nature of analyses in the rationalizing stage. The research that follows probed both IOS adopters and non-adopters rigorously across scientifically valid samples. The studies in the probing stage are heavily influenced by the American MIS tradition (positivistic research methods supporting factor models and hypotheses). They also centered around one particular technology: EDI. This research tradition accepts the notion of cumulative scientific advancement, and though the researchers concerned clearly learned from each other, and adapted and debated each others’ models and hypotheses, no unified and universally accepted theory of IOS adoption emerged. Process theorists question the adequacy of factor based analyses for explaining the rich and complex nature of IOS. Process theorists attempt to understand the complex and dynamic nature of IOS by closely interacting with the adoption context. Though the factor studies continue, the trend is towards more contextually aware studies, with theories and methods which allow enable richer descriptive analysis, sometimes at the expense of the ability to be normative or predictive. The classic debate about inter-relating process and factor based studies is very much applicable in the IOS field at
Many inconsistencies and disagreements nevertheless remain. For instance different researchers at different points of the evolution of the field display conflicting understandings of adoption. In the rationalizing stage, adoption is largely assumed and even the organizational context in which it takes place is poorly recognized. In the probing stage, adoption is studied as “... the decision to commit resources to the innovation.” [30]. However, in the understanding stage, adoption is studied as an evolving “pattern” [22]. Moreover the field reflects a much more basic diversity of philosophical stance (positivist, interpretive), research method (statistical, case) and theory (framework, factor model, pattern) which is reflected throughout the IS discipline, and is certainly irreconcilable. With the rapid growth of the internet, it seems that the topic of IOS adoption will remain an important one for the foreseeable future, though many of the parameters may change quickly. What lessons can be learned from this historical evaluation if we focus, for example, on the adoption of Business to Business market places and eprocurement systems? We might expect the business press aimed at managers to help us define the trends and important research areas and topics. Various research methods and theoretical approaches are available, perhaps to support different purposes at different stages of the topics’ evolution. Managers needing topical advice or answers to problems require normative models
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
with limited variables and context and high communication potential. Researchers can build on these approaches by relating them to established theories and using more rigorous methods, but at the cost of unanimity and communicability. The wider the context, the wider the descriptive power of the resulting theory and the greater the cost to predictive and normative
capability. At some point it becomes necessary to translate the wide range of theoretical considerations and debates that constitute an emerging field back into terms which are relevant to practitioners. A further consideration becomes the transferability of knowledge: how can we investigate a new field without forgetting what we learned from the previous related field?
Stage classification Rationa lizing
Analyses Classification
Assumptions held
i) IOS viewed ma inly from the initiator’s perspective and ii) Causal a genc y: technology imperative.
Researc h approac h used
Re lies on a necdotal data.
Theories used
Porters five forces and transaction cost theory
Probing i) Buyer-supplier context in which IOS is adopted is recognized ii) a ll contexts a re similar iii) adoption as a n instance and iv) effect of IOS on the e nvironment is unidirectiona l Positivist studies; fac tor approach, surve ys and mostly snapshots. Tra nsaction cost theory, Diffusion of Innovation theory, Critical mass theory, soc io-political theory
Understanding i) IOS is dynamic in nature ii) larger network context in which IOS is adopted is recognized iii) e very conte xt is unique a nd iv) adoption as a process Interpretive studies; process approac h, in-depth ana lysis and longitudinal studies Transaction cost theory, Diffusion of Innova tion theory, Critica l mass theory, sociopolitica l theory
Table 1.0: Summary of our analyses
[7]
Acknowledgements The authors acknowledge the financial support of the PITNIT project (funded by Danish Research Agency Grant number 9900102). The authors thank Peter Axel Nielsen for his reflections on the paper.
[8] [9] [10]
7
Literature List
[1]
Astley, W.G. and Van de Ven, A.H., “Central Perspectives and Debates in Organizational Theory”, Administrative Science Quarterly, vol. 28, pp. 245-273, 1983. Bakos, J.Y. and Kemerer, C.F., “Recent Applications of Economic Theory in Information Technology Research”, Decision Support Systems, vol. 8, pp. 365-386, 1992. Bakos, J.Y. and Treacy, M.E., “Information Technology and Corporate Strategy: A Research Perspective”, MIS Quarterly, vol. 10, issue 2, pp. 107-119, June 1986. Bakos, Y. and E. Brynjolffson (1993). "From vendors to partners: Information technology, incentives, and the optimal number of suppliers.", Journal of Management Information Systems, vol. 10, issue 2, pp. 37-53. Barua, A. and Lee, B., “An Economic Analysis of the Introduction of an Electronic Data Interchange System”, Information Systems Research, vol. 8, no. 4, Dec. 1997. Benjamin, R.I., De Long, D.W. and Morton, M.S.S., ”Electronic Data Interchange: How Much Competitive Advantage”, Long Range Planning, vol. 23, no. 1, pp. 2940, 1990.
[2] [3] [4]
[5] [6]
[11]
[12] [13] [14] [15]
[16]
Benjamin, R.I, Rockart, J.F., Morton, M.S.S. and Wyman, J., “Information Technology: A Strategic Opportunity”, Sloan Management Review, vol. 25, no. 3, Spring 1984. Bergeron, F. and Raymond, L., “The Advantages of Electronic Data Interchange”, DATABASE, Fall 1992 Bouchard, L., “Decision Criteria in the Adoption of EDI”, Presented at International Conference in Information Systems, 1993. Cash, J. I. and B. R. Konsynski (1985). "IS Redraws Competitive Boundaries.", Harvard Business Review, vol. 64, issue 2, pp. 134-142. Cavaye, A.L.M. and Cragg, P.B., “Factors Contributing to the Success of Customer Oriented Interorganizational Systems”, Journal of Strategic Information Systems, vol. 4, no. 1, pp. 13-30, 1995. Cheon, M. J., V. Grover, et al. (1993). "The evolution of empirical research in IS - A study in IS maturity.", Information and Management, Issue 24, pp.107-119. Chwelos, P, Benbasat, I. And Dexter, A.S., “Empirical Test of an EDI Adoption Model”, Information Systems Research, Vol. 12, Issue 3, 2001. Clemons, E.K., “Information Systems for Sustainable Competitive Advantage”, Information and Management, vol. 11, pp. 131-136, 1986. Clemons, E.K., Reddi, S.P. and Row, M.C., “The Impact of Information Technology on the Organization of Economic Activity: The “Move to the Middle” Hypothesis”, Journal of Management Information Systems, vol. 10, issue 2, pp. 9-31, 1993. Clemons, E.K. and Kleindorfer, P.R., “An Economic Analysis of Interorganizational Information Technology”, Decision Support Systems, vol. 8, pp. 431-446, 1992.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
[17] Clemons, E.K. and Row, M.C., “Structural Differences Among Firms: A Potential Source of Competitive Advantage in the Application of Information Technology”, International Conference of Information Systems, Dec. 1987. [18] Clemons, E.K. and Kimbrough, S.O., “Information Systems, Telecommunications, and their Effects on Industrial Organizations”, Proceeding of the 7th International Conference of Information Systems, 1986. [19] Clemons, E.K. and McFarlan, F.W., “Telecom: Hook Up or Lose Out”, Harvard Business Review, no. 4, July – August 1986. [20] Crum, M.R., Premkumar, G. and Ramamurthy, K., “An Assessment of Motor Carrier Adoption, Use and Satifaction with EDI, Motor Carrier EDI, pp. 44-57, summer 1996. [21] Damsgaard, J. (1997), “The Danish EDI Bandwagon Gaining Momentum, Proceedings of the 30th Annual Hawaii International Conference on System Sciences. [22] Damsgaard, J. and Lyytinen, K. (1998), “Contours of Diffusion of Electronic Data Interchange in Finland: Overcoming Technological Barriers and Collaborating to Make it Happen”, Journal of Strategic Information Systems, Vol. 7, pp. 275-297. [23] Damsgaard, J. and Lyytinen, K, “The Dynamics of Factors Explaining EDI Diffusion in Hong Kong in the Late 1990’s”, Proceedings of the Fourth Pacific Asia Conference on Information Systems, Hong Kong SAR, pp. 1061-1074, 2000. [24] Downs, G.W., Jr. and Mohr, L.B., “Conceptual Issues in the Study of Innovation”, Administrative Science Quarterly, vol. 21, Dec. 1976. [25] Drury, D.H. and Farhoomand, A., “Innovation Adoption of EDI”, Information Resources Management Journal, vol. 9, no. 3, Summer 1996. [26] Emmelhainz, M.A., “Strategic Issues of EDI Implementation”, Journal of Business Logistics, vol. 9, no. 2, 1988. [27] Falconer, D. J. and D. R. Mackay (1999)., Ontological problems of pluralist research methodologies., Proceedings of AIS AMCIS 99: 1999 Americas Conference on Information Systems, Association of Information Systems. [28] Gregor, S. and Johnston, R.B., “Developing an Understanding of Interorganizational Systems: Arguments for Multi Level Analysis and Structuration Theory”, Proceedings of the 8th European Conference on Information Systems, 2000. [29] Gregor, S. and Menzies, D., “Electronic Data Interchange and Supply Chain Management: A Case Study of the Beef Industry”, The Third Collaborative Electronic Commerce Technology and Research, Wellington, 1999. [30] Grover, V. and Goslar, M.D., “The Initiation, Adoption and Implementation of Telecommunications Technologies in U.S. Organizations”, Journal of Management Information Systems, vol. 10, issue 1, 1993. [31] Grover, V., “An Empirically Derived Model for the Adoption of Customer-Based Interorganizational
[32] [33]
[34] [35] [36] [37]
[38]
[39]
Systems”, Decision Sciences, vol. 24, no. 3, May /June 1993. Hafner, A.B., “Pareto's Principle: The 80-20 Rule”, http://library.shu.edu/HafnerAW/awh-th-mathpareto.htm, Site accessed on 22nd April 2002. Hart, P. and Saunders, C., “Power and trust: Critical Factors in the Adoption and Use of Electronic Data Interchange”, Organization Science, vol. 8, no. 1, Jan.Feb. 1997. Heck, E.V. and Ribbers, P.M., “The Adoption and Impact of EDI in Dutch SME’s”, Proceedings of the 32nd Hawaii International Conference on System Sciences, 1999. Iacovou, C. L., I. Benbasat, et al. (1995)., "Electronic Data Interchange and Small Organizations: Adoption and Impact if Technology.", MIS Quarterly, pp. 465-485. Ives, B. and Learmonth, G.P., “The Information System as a Competitive Weapon”, Communications of the ACM, vol. 27, no. 12, pp. 1193-1201, Dec. 1987. Johnston, R.B. and Gregor, S., “A Theory of IndustryLevel Activity for Understanding the Adoption of Interorganizational Systems”, European Journal of Information Systems, vol. 9, pp. 243-251, 2000. Johnston, R.B. and Mak, H.C., “An Emerging Vision of Internet-Enabled Supply Chain Electronic Commerce”, International Journal of Electronic Commerce, vol. 4, Issue 4, 2000. Iacovou, C.L, Benbasat, I. And Dexter, A.S., ”Electronic Data Interchange and Small Organizations: Adoption and Impact if Technology”, MIS Quarterly, pp. 465-485, Dec. 1995.
[40] K. Kumar, H. van Dissel and P. Bielli, The Merchant of Prato revisited: Towards a third rationality of Information Systems, MIS Quarterly 22:2, 1998, pp. 199-226. [41] Kurnia, S. and Johnston, R.B., “A Review of Approaches to EC-enabled IOS Adoption Studies”, Proceedings of the 35th Annual Hawaii International Conference on System Sciences (HICCS -35’02), 2002. [42] Kurnia, S. and Johnston, R.B., “The Need for a Processual View of Inter-Organizational Systems Adoption”, Journal of Strategic Information Systems, vol. 9, pp. 295-319, 2000. [43] Kurnia, S. and Johnston, R.B., “Adoption of Efficient Consumer Response: The Issue of Mutuality”, Supply Chain Management: An International Journal, Vol. 6, No. 5, pp. 230-241, 2001. [44] Lee, A. S. (1991). "Integrating Positivist and Interpretive Approaches to Organizational Research.", Organization Science, Vol.2, issue 4, pp. 342-365. [45] Ling, C.Y., “Model of Factors Influences on Electronic Commerce Adoption and Diffusion in Small and Medium Sized Enterprises”, Pacific Asia Conference in Information Systems, 2000. [46] Lyytinen, K. and Damsgaard, J., “What’s Wrong with the Diffusion of Innovation Theory?”, In Proceedings of the IFIP TC8 Working Group 8.6 Conference, Diffusing Software Product and Process Innovations, April 7-10, pp. 173-190, Banff, Canada, 2001.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Appendix 12.2 Research Publication Two – the ECIS 2003 Research Philosophies in the IOS Adoption Field By Ramanathan Somasundaram and Jan Karlsbjerg
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Appendix 12.2 Research Publication Two – ECIS 2003 Research Philosophies in the IOS Adoption Field By Ramanathan Somasundaram and Jan Karlsbjerg
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Research Philosophies in IOS Research
Research Philosophies in the IOS Adoption Field Ramanathan Somasundaram Department of Computer Science Aalborg University Fredrik Bajersvej 7E DK-9220 Aalborg OE Denmark Email:
[email protected] Tel: +45 9635 8852 Fax: +45 9815 9889 Jan Karlsbjerg Department of Computer Science Aalborg University Fredrik Bajersvej 7E DK-9220 Aalborg OE Denmark Email:
[email protected] Tel: +45 9635 8852 Fax: +45 9815 9889 Abstract This paper pursues two objectives. First it depicts the status of research philosophies of the IOS adoption field. Second, it gives an overview of the discussions within the field of IS about interrelating research philosophies and analyzes to which extent this has happened in the IOS adoption field thus far. In depicting the philosophical status of the field we apply Orlikowski and Baroudi’s (1991) framework that classifies IS research as relying on a foundation of assumptions from either positivism, interpretivism or critical theory. We found the framework well suited for our review of the field, although we did not find any instances where critical theory had been used as the basic assumption in an IOS adoption study. Our findings show that there are several clear openings for contributions that apply new research perspectives to the field of IOS or combine multiple research perspectives for added insight into the dynamics of IOS adoption. Keywords IOS, inter-organizational systems, research philosophies, scientific progress, positivism, interpretivism, critical theory
Introduction This paper presents an analysis of the progression of science with the field of adoption of inter-organizational systems (IOS) from the field’s inception in the mid 1980s until the present day. We have conducted an extensive review of IOS adoption related articles and
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Page 1
Research Philosophies in IOS Research
analyzed their research philosophies, i.e. the underlying ontology and epistemology as well as the methodology applied. We take this perspective to ensure a depth/relevance in the analysis, which we believe has been lacking in previous work. Chua (1986) presented a framework for analyzing research approaches in his field of accounting. The framework classifies research philosophies as under three which are i) positivist ii) interpretive and iii) critical. Each of the philosophies are detailed further as under assumptions, beliefs about physical and social reality, beliefs about knowledge (includes both epistemology and methodology), beliefs about theory and practice and through an empirical example. Orlikowski and Baroudi’s (1991) later adopted the framework for analyzing the field of IS. In so doing they applied the framework to articles from a broad range of subjects, but from a small number of publication outlets. Our work is distinct from previous work (Lee 1991; Orlikowski and Baroudi 1991; Landry and Banville 1992; Benbasat and Weber 1996; Lee 1999; Mingers 2001) in three ways. First, we analyze scientific progression in a narrowly defined IOS adoption field rather than attempt an analysis of such a wide field as all of IS as we feel that such wide scope analyses necessarily lead to broad, conceptual arguments that are not easily applicable for new research projects. In contrast, we are able to depict the specifics of scientific progression in the field and support our arguments with a number of concrete findings and citations within a narrow focus. Second, traditional literature reviews are based on a selection of a small number of recognized publication outlets for a period of time (e.g. Orlikowski and Baroudi (1991) and Cheon et al. (1993)). The disadvantages of such an approach are i) the list of chosen outlets is often disputed and ii) more importantly, publications in conferences and lesser deemed journals that have contributed to the scientific development are left out. We have applied a more comprehensive sample selection approach that includes publications within the field from as early as 1966 (Kaufman 1966) to as recent as 2002 (Kurnia and Johnston 2002) from journals, conferences, and book chapters. The process that we adopted for selecting the literature sample is explained in the methodology section. Third, we support our analysis with the very details of scientific progression synthesized from literatures in the IOS adoption field. This is in contrast with previous work where progression has been analyzed by studying a few static variables. E.g. Orlikowski and Baroudi (1991) identify positivism as the dominant research paradigm by reviewing 155 articles in the IS field, and Cheon et al. (1993) test generic variables such as “research methods” and “research types” for explaining IS maturity. These studies however provide little insight about the dynamics of progression. We hold that insight into the dynamics is important for making predictions and daring to be normative about the future of a research field. The remainder of the paper is organized as follows. Section two presents the process by which we selected our sample of papers for the review. Section three presents the research philosophy framework (Chua 1986; Orlikowski and Baroudi 1991) and analyzes the reviewed papers as under each of the three research philosophies. This is followed by a holistic discussion of the analyses, and finally section five concludes the paper where we sum up our findings and identify promising avenues for further research.
Review Process A sample of 73 papers that study IOS adoption was analyzed for a review of the IOS adoption field. This review is documented in detail in a separate paper (Somasundaram and Rose 2003). The present paper extends the analysis of the literature review. The papers in the
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Page 2
Research Philosophies in IOS Research
sample were classified as under positivist, interpretive or critical philosophies. Then, the scientific progression as under each of the philosophies is analyzed. A discussion about the state of and the prospect of inter-relating philosophies in the IOS adoption field follows. We arrived at the sample using the snow-balling technique. Moriarty and Bateson (1982) define the use of snow-balling techniques for arriving at a sample. They present three snowballing techniques which are i) single-stage snowball ii) multiple-stage snowballing and iii) exhaustive snowballing. We have applied the exhaustive snowballing which prescribes following all leads obtained in the first sampling stage to reach a sample for the second stage, and continuing this process until no new leads are generated. We relied on four papers across time-period as our starting points. They are i) a review paper by Kurnia and Johnston (2002) where the authors evaluate the emergence of process oriented research in the IOS adoption field ii) a paper by Damsgaard and Lyytinen (1998) which calls for a processual and multi-level analysis iii) a classic factor based research by Hart and Saunders (1997) and iv) one of the first instances of theorizing in the IOS area by Bakos and Treacy (1986). We followed the literature lists in the four above mentioned papers (first stage snowballing), but thoroughly read only those that studied IOS adoption at the organizational and inter-organizational level of analysis. Then we followed the literature list of the second round of papers further and so on until we found no new papers. We did not adopt citation search for arriving at our sample because the term “adoption” is often confused with the terms “diffusion” and “implementation”. Furthermore, the results that we got did not fit into our restricted focus when we searched ABI/Inform using the following key words: IOS, adoption, organization and inter-organization. Had we chosen to arrive at the sample using key word search, we would have ignored the research work (prior to Bakos and Treacy (1986)) where IOS adoption is indirectly studied under the key words “IT” and “competitive advantage”. Apart from the IOS adoption papers, we also reviewed related conceptual papers such as that of Cooper and Zmud (1990) where adoption is defined as a stage in the implementation process, and Kumar et al. (1998) where trust and co-operation are identified as the third underlying rationality. The discussion under each of the philosophies serves two purposes. Firstly, it temporally explains the progression in the IOS adoption science within each philosophy strand. Secondly, it analyzes the consistency with which the science has progressed within each strand. We evaluate the possibility of interrelating philosophies in the IOS adoption field in the discussion section.
Orlikowski and Baroudi’s Framework Applied to IOS Adoption Research Orlikowski and Baroudi (1991) discuss a range of philosophical assumptions available to study IS phenomena in order to “encourage greater debate and mindfulness around the approach we adopt when we embark on research investigations”. They operationalize this objective by adopting Chua’s (1986) classification of research epistemologies as either i) positivist ii) interpretive or iii) critical studies. While Orlikowski and Baroudi use Chua’s classification mainly to discuss a range of philosophical assumptions per se, we use this classification for analyzing the application of these philosophical assumptions in the IOS adoption field. We rely upon our literature survey data for conducting our analysis and the following discussion.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Page 3
Research Philosophies in IOS Research
Positivist Philosophy in IOS Adoption Research The central principles of positivist philosophy as identified from Orlikowski and Baroudi (1991) are i.
The world that we study is “objective” and exists “independent of humans”.
ii.
The researcher’s duty is to explain this “physical and social world” through universal laws or principles.
iii.
Deductions made from the laws or principles can be used to explain “events or actions”. When aware of the laws or principles researchers can predict and control events or actions. The rigor, validity and replicability criteria for conducting and evaluating research according to this philosophy have been institutionalized in the field of IS since its origin of the hardware disciplines of systems construction. All of our surveyed work published in renowned outlets until the mid 90’s apply positivist philosophy. Orlikowski and Baroudi (1991) epistemologically classify the positivist papers in their review sample as either i) descriptive or ii) theoretically grounded. The ratio of descriptive to theoretically grounded papers in their sample of 155 papers is approximately 1:3. Their work does not account for the time at which descriptive and theoretically grounded papers were published. Our analysis shows a similar approximated ratio between the descriptive and theoretically grounded work in the positivist IOS context. IOS adoption emerged as a field from the descriptive work as the impact of IOS is implicitly discussed under the “IT and competitiveness” umbrella based on anecdotal evidence. Most of the descriptive work is published in the mid 1980’s. Bakos and Treacy (1986) along with a few others (e.g. Clemons and Kimbrough (1986)) lay the base for a theory based cumulative tradition highlighting the then descriptive and anecdotal nature of the research. The volume of published descriptive work gradually decreases from this time as theoretically grounded papers become main stream. The well established ideology of logical empiricism is used as the logic for cumulating research work in the positivist perspective by theory refinement through falsification. The theoretically grounded IOS papers implement logical empiricism in a standardized pattern that resembles framework development through: established findings – sample selection – statistical analysis – findings and discussion. The IOS adoption field has had difficulties in arriving at a standard, encompassing research framework or theory. There are, however, a few influential frameworks for studying IOS adoption that are frequently quoted in the literature, such as an empirically derived detailed adoption model (Grover 1993), a small business adoption model (Iacovou, Benbasat et al. 1995), an EDI adoption and use model (Hart and Saunders 1997) and a research model (Premkumar, Ramamurthy et al. 1997). Researchers following this ideology usually select one of these frameworks, in many cases adapt it to suit their situation (e.g. (Heck and Ribbers 1999)) and test its applicability. The papers in our sample show a variety of sampling sizes and techniques with sample sizes ranging from 100 phone calls (Cooper and Zmud 1990) to 5000 questionnaires (O'Callaghan, Kaufmann et al. 1992), and return rates ranging from 11.6% in a double-round survey study (Bergeron and Raymond 1992) to 62% in a phone survey where the participants had agreed in advance to participate. Figure 1 presents a summary of the number of usable responses in the quantitative positivist studies in our sample.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Page 4
Research Philosophies in IOS Research
5 4 3 2
Usable responses
1 0 50-99 100149
150199
200249
250299
300349
350399
Figure 1. Usable responses in quantitative positivist studies in our sample A model gains acceptance and becomes a standard explanatory model for IOS adoption when it passes the falsification test in a large number of studies. No model has yet been tested enough for it to emerge as the one standard model. Moreover, the cumulative knowledge development has been hampered by inconsistent findings between studies. For example, while Iacavou et al. (1995) and Grover (1993) find relative advantage as having a positive impact on IOS adoption, Premkumar et al. (1995) and Saunders et al. (1992) do not find a significant relationship. It is not just the independent variables and their relationship with the dependent variable that vary among models. The dependent variable itself varies as well. The dependent variables that have been studied include the following: i.
Intent/decision to adopt
ii.
Rate of adoption
iii.
Extent of adoption
iv.
Internal effects of adoption
v.
Proactive vs. reactive decision mode for adoption
vi.
Timing of adoption
While this shows the richness of IOS adoption studies, it is important to recognize the implications of such diversity in cumulative knowledge development. The positivist philosophy is criticized for overly abstracting or oversimplifying the complex IOS phenomena through a few unilateral relationships. Damsgaard and Lyytinen (1998) mention that unlike simple innovations such as television sets, IOSs represent a rich and complex technology whose adoption happens over a period of time as a result of interactions among multiple parties. Thus they deem the factor based approach with its snapshots of adoption processes as inadequate for richly explaining the complex IOS adoption phenomena.
Interpretive Philosophy in Inter-Organizational Systems (IOS) Adoption Research The central principles of interpretive philosophy as identified from Orlikowski and Baroudi (1991) are
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Page 5
Research Philosophies in IOS Research
i.
The world/reality is not objective; instead it is presumed to be socially constructed.
ii.
A researcher’s role is to understand the inter-subjective meanings embedded in social life and explain the actions of those constructing the reality.
iii.
While this approach is appropriate for studying complex phenomena in depth, it is restrictive regarding breadth/generalizability.
The interpretive philosophy has been gaining acceptance among the larger IS community in the recent years. In 1993 the high-ranking IS journal MIS Quarterly explicitly changed its editorial policy to include interpretive research work (DeSanctis 1993 quoted in Walsham (1995)). However, tools and techniques for carrying out interpretive studies have not yet reached the same widespread use as their positivist counterparts. Only a few institutions, mostly European based, provide training in using interpretive tools and techniques. Table 1 illustrates the historical dominance of positivist research philosophy in our sample, and figure 2 illustrates the development over time as interpretivist studies gain acceptance in the scientific outlets.
Research philosophy
Frequency
Percent
Positivist
56
76.7
•
“descriptive”
(26)
(35.6)
•
Theoretically grounded
(30)
(41.1)
Interpretive
17
23.3
Critical
0
0
Total
73
100%
Table 1. Articles classified by research philosophy (modeled from Orlikowski & Baroudi (1991))
16 14 12 10 8 6 4 2 0
Interpretive studies Positivist studies
19651969
19751979
19851989
19951999
Figure 2. Distribution of positivist and interpretive studies in our sample over time One of the first published works that applies interpretive philosophy in the study of IOS adoption is the work of Damsgaard (1997). He categorizes his field study data into diffusion patterns for demonstrating the complex nature of EDI. The appropriateness of using this process approach for studying IOS adoption is analyzed in depth by Kurnia and Johnston Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Page 6
Research Philosophies in IOS Research
(2000; 2001). They argue that a factor based approach is inadequate for studying the complex IOS adoption process. Capability of organization, nature of technology and external factors do not uni-directionally determine the action that leads to the adoption of IOSs. Instead, the adoption of an IOS is seen as an interactive process where the causal notion is bi-directional in nature; the factors affect the adoption of IOS and get affected in the process. It is inadequate just to study the causal notions at the organizational level. Instead the effects of industry and remote environment in the adoption of IOS are also to be understood. Finally, the IOS adoption is not an instantaneous act, but one that takes place over a period of time. Hence, longitudinal studies that study the dynamics of interactions are preferred over snapshot factor based approaches. Orlikowski and Baroudi (1991) extend Chua’s framework by classifying interpretive researchers as under two categories; “weak” and “strong” constructionists. Researchers adopting the weak constructionist view attempt to provide an intuitive explanation of an act or an experience. In the strong constructionist view, the neutrality of the researcher in providing an unbiased descriptive account is questioned. Thus, the researcher is expected to analyze the influences of his or her biases on the descriptive accounts provided. The “strong” constructionist philosophy is argued as a replacement for the positivist unlike the “weak” view which is touted as complementing the positivist. Our research shows two schools of thoughts among the papers that study IOS adoption from an interpretive philosophy. Both of them fall under the “weak” constructionist category and they both encourage the researcher to approach the context with a framework in mind, with the objective of the study being refinement of the framework. The differences however arise from their beliefs about the extent to which factor based approaches can explain IOS adoption. Those in the first school of thought (Kurnia and Johnston 2000) who are aligned closer to the positivist philosophy view their interpretive work as a ‘second-order’ model that details the dynamics of interactions conceptualized in the ‘first-order’ factor model. Those in the second school of thought (Damsgaard and Lyytinen 1998) view factor based approaches as inadequate for studying the complex IOS adoption phenomena. The interpretive philosophy is frequently criticized by its opponents for producing inconclusive results. After studying an interpretive paper a reader is often left wondering what such analytical descriptions mean or how they are useful. Kurnia and Johnston (2000) state that “While this approach promises to give greater depth of understanding of dynamic and complex interactions of organizations within the industry in IOS adoptions, one must acknowledge that it provides a reduced ability to make general statements. Furthermore due to the complexity and richness of the analysis it may create barriers to the interpretation of the findings particularly for practitioners.”
Critical Philosophy in Inter-Organizational Systems (IOS) Adoption Research The central principles of critical philosophy as identified by Orlikowski and Baroudi (1991) are i.
Social reality is historically constituted, hence human beings, organizations and societies are not confined to existence in a particular state.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Page 7
Research Philosophies in IOS Research
ii.
The role of a researcher is to expose the hidden contradictions and unfulfilled potentiality in the societal order and initiate changes in the social relations and practices.
iii.
“Critical theorists do not share common philosophical standards for the evaluation of theories. What is acceptable theory or explanation is still debatable. This ambiguity of evaluation may be difficult for proponents of the dominant research tradition to accept, given their experience with positivism’s relatively unambiguous criteria for what constitutes valid knowledge.” (Orlikowski and Baroudi 1991 p.23).
We did not find papers in our sample that can absolutely be categorized under critical philosophy. However, we did find applications of a critical philosophy concept “totality” in our sample. As distinctive features of critical philosophy were not found in those papers we have categorized the papers under the other philosophies. The “totality” concept denotes that a phenomenon does not exist as an isolated element; instead it is related to its context through multiple relations which are essentially interdependent. Johnston and Gregor (2000) use an adapted version of this concept for developing a Structuration type IOS adoption theory. In this theory, a firm and its actions are rationalized as within its immediate industry environment and external environment. Damsgaard and Lyytinen (1998) use micro, meso and macro concepts and their interrelations for rationalizing an outcome. Their work could have been categorized as under critical philosophy had they focused on the structural contradictions between buyers and suppliers and initiated changes for altering the status quo. Their objective seems to be rooted in the implementation science instead.
DISCUSSION As it can be seen from the sample quite a few interpretive papers have been published in journals such as Journal of Strategic Information Systems (e.g. (Damsgaard and Lyytinen 1998)) and in conferences such as HICSS (e.g. (Kurnia and Johnston 2002)), while at the same time, a classical positivist work about EDI adoption is published in the journal Information System Research (Chwelos, Benbasat et al. 2001). With IOS adoption research poised to take off to new heights in the e-business age, which of the philosophies should one adopt? Can one integrate the philosophies instead of choosing one over other? A number of approaches have been suggested to do just that. Kurnia and Johnston (2000) propose that positivist factor based studies be conducted first to understand the relationship between multiple independent variables and IOS adoption – the dependent variable. “Second-order” interpretive studies should then be conducted to improve the understanding of the complex bi-directional relations among the independent and dependent variables in the positivist research framework and to refine the research framework for better representing reality. According to this framework, scientific progression would take place as the refined framework is tested using factor based methods and then refined using interpretive methods and so on and so forth. In contrast to Kurnia and Johnston, Lee (1991) suggests that researchers first conduct interpretive studies to reach a rich understanding. From this understanding, a list of variables may be chosen to be tested in positivist studies across multiple settings. Lee presents a threelevel understanding framework to operationalize his suggestion. The first level is the subjective understanding held by the practitioners who create the reality. The second level is the interpretive understanding that the researcher obtains by interacting with the subjects in
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Page 8
Research Philosophies in IOS Research
focus. The third level is the positivist understanding which also belongs to the researcher. This understanding is created and tested across a large sample in order to explain the studied empirical reality. Like Kurnia and Johnston’s framework, Lee’s framework is iterative. Both of the frameworks suggest knowledge accumulation through cumulativeness either within the same study or among multiple studies. Mingers (2001) provides the philosophical reasoning for Kurnia and Johnston’s and Lee’s suggestions. He argues that research paradigms are simply “constructions of thoughts” and that it’s a fallacy to hold that the world must conform to these paradigms. Hence, their definitions should not be considered as static, but rather as evolving. He further argues that paradigms are “permeable at the edges” and thus not incommensurable. Mingers’ thinking is countered by the work of Falconer and Mackay (1999) who argue that cross-paradigmatic research is “ill-founded” and that multi-paradigm proposals that operate at the epistemological level fail to recognize the paradigmatic differences at the deeper ontological level. The central disagreement in the discussion is at the ontological level, about whether or not different ontologies can be mixed. Orlikowski and Baroudi (1991) provide a partial resolution to this discussion by classifying and characterizing interpretive philosophy as under weak and strong constructionists. They argue that while positivist and weak constructionist research can be inter-related and cumulated upon, positivist and strong constructionist research cannot because they are conceptual opposites. If we are to adopt Orlikowski and Baroudi’s position, the IOS adoption field is rather fortunate. In our sample, all of the interpretive work adopted weak constructionist ideology and thus Kurnia and Johnston’s (2000) suggestion for integrating positivist and interpretive research seems a relevant option for the IOS adoption field. With the emergence of interpretivist research, we predict the application of strong constructionist work in the IOS adoption field in the future. Judging from the discussions in the IS field, the critical philosophy and a hybrid of the critical and interpretive philosophies might also be applied for studying IOS adoption in future studies. Our stance on the philosophy debate is as follows; each of the philosophies has its strengths and weaknesses. While through applying interpretive philosophy one can understand the dynamics associated with the adoption process in detail, interpretivism is ineffective for creating context free rules. Positivist philosophy in contrast tests relationships between two or more abstract variables across contexts with the objective of creating universal rules, but in the pursuit of generality the insights lose contact with the individual contexts. We believe researchers should select a philosophy that is appropriate for achieving their objectives. We argue in the line of Mingers (2001), Lee (1991) and Kurnia and Johnston (2000) that knowledge accumulated in different philosophies can be inter-related and built upon. It is not optimal for science if the philosophy strands continue to develop independently. Specifically, we advocate Lee’s subjective-interpretive-positivist understanding cycle for cumulative knowledge development. In our review of the IOS adoption field, however, we have not been able to find any completed cycles of knowledge generation among researchers applying the two research philosophies. Interpretivist researchers often quote positivist research as part of the previous work in the field, and their research addresses what they believe to be shortcomings in the positivist research. But we have not found evidence that the insights produced in interpretivist research is being discussed or applied in subsequent positivist research.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Page 9
Research Philosophies in IOS Research
CONCLUSION Our objective in writing this paper has been two-fold. First to depict the scientific progress within positivist, interpretive or critical philosophy strands in the IOS adoption field and second to provide an overview of the discussions about inter-relating philosophies in the IS field. We apply Orlikowski and Baroudi’s (1991) framework that classifies research as relying on a foundation of assumptions from either positivism, interpretivism or critical theory. We found the framework well suited for our review of the field since it facilitated a coherent presentation of the discussion about interrelating philosophies through a review of literatures in the larger IS and the organizational science fields, thus achieving our second objective. While positivist studies have dominated the field of IOS adoption research since its emergence as a separate field in the mid 1980s, interpretivist studies are gaining ground in both conference proceedings and journals, and they can no longer be ignored by the traditional positivist researchers. For the sake of the scientific progress of the field more work must be done toward combining the insights from positivist and interpretivist studies. From the same inclusive philosophy we also encourage research in IOS adoption using the critical theory tradition as we feel this will advance the scientific field.
References Bakos, J. Y. and M. E. Treacy (1986). “Information Technology and Corporate Strategy: A Research Perspective.” MIS Quarterly 10(2): 107-99. Bergeron, F. and L. Raymond (1992). “The Advantages of Electronic Data Interchange.” DATABASE 23(4): 19-31. Chua, W. F. (1986). “Radical Developments in Accounting Thought.” The Accounting Review 61(4): 601-32. Chwelos, P., I. Benbasat, et al. (2001). “Empirical Test of an EDI Adoption Model.” Information Systems Research 12(3): 304-21. Clemons, E. K. and S. O. Kimbrough (1986). Information Systems, Telecommunications, and their Effects on Industrial Organizations. 7th International Conference of Information Systems. Cooper, R. B. and R. W. Zmud (1990). “Information Technology Implementation Research: A Technological Diffusion Research.” Management Science 36(2). Damsgaard, J. (1997). The Danish EDI Bandwagon Gaining Momentum. Proceedings of the 30th Annual Hawaii International Conference on System Sciences, Hawaii, USA. Damsgaard, J. and K. Lyytinen (1998). “Contours of diffusion of electronic data interchange in Finland; Overcoming technological barriers and collaborating to make it happen.” Journal of Stategic Information Systems 7(4): 275-97. Falconer, D. J. and D. R. Mackay (1999). Ontological problems of pluralist research methodologies. Proceedings of AIS AMCIS 99: 1999 Americas Conference on Information Systems, Association of Information Systems. Grover, V. (1993). “An Empirically Derived Model for the Adoption of Customer-Based Interorganizational Systems.” Decision Sciences 24(3). Hart, P. and C. Saunders (1997). “Power and trust: Critical Factors in the Adoption and Use of Electronic Data Interchange.” Organization Science 8(1).
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Page 10
Research Philosophies in IOS Research
Heck, E. V. and P. M. Ribbers (1999). The Adoption and Impact of EDI in Dutch SME's. Proceedings of the 32nd Hawaii International Conference on System Sciences, Hawaii. Iacovou, C. L., I. Benbasat, et al. (1995). “Electronic Data Interchange and Small Organizations: Adoption and Impact if Technology.” MIS Quarterly: 465-85. Johnston, R. B. and S. Gregor (2000). “A Theory of Industry-Level Activity for Understanding the Adoption of Interorganizational Systems.” European Journal of Information Systems 9: 243-51. Kumar, K., V. H. G. Dissel, et al. (1998). “The Merchant of Prato - Revisited: Toward a Third Rationality of Information Systems.” MIS Quarterly 22(2): 199-226. Kurnia, S. and R. B. Johnston (2000). “The Need for a Processual View of InterOrganizational Systems Adoption.” Journal of Strategic Information Systems 9: 295319. Kurnia, S. and R. B. Johnston (2001). “Adoption of Efficient Consumer Response: The Issue of Mutuality.” Supply Chain Management: An International Journal 6(5): 230-41. Kurnia, S. and R. B. Johnston (2002). A Review of Approaches to EC-enabled IOS Adoption Studies. Proceedings of the 35th Annual Hawaii International Conference on System Sciences, Hawaii. Lee, A. S. (1991). “Integrating Positivist and Interpretive Approaches to Organizational Research.” Organization Science 2(4): 342-65. Mingers, J. (2001). “Combining IS research methods: towards a pluralist methodology.” Information Systems Research 12(3): 240-59. Moriarty, R. T. and J. E. G. Bateson (1982). “Exploring Complex Decision Making Units: A New Approach.” Journal of Marketing Research 19(2): 182-91. O'Callaghan, R., P. J. Kaufmann, et al. (1992). “Adoption Correlates and Share Effects of Electronic Data Interchange Systems in Marketing Channels.” Journal of Marketing 56(2). Orlikowski, W. J. and J. Baroudi (1991). “Studying Information Technology in Organizations: Research Approaches and Assumptions.” Information Systems Research 2(1): 1-28. Premkumar, G. and K. Ramamurthy (1995). “The Role of Interorganizational and Organizational Factors on the Decision Mode for Adoption of Interorganizational Systems.” Decision Sciences 26(3). Premkumar, G., K. Ramamurthy, et al. (1997). “Determinants of EDI Adoption in the Transportation Industry.” European Journal of Information Systems 6: 107-21. Saunders, C. S. S. and S. Clark (1992). “EDI Adoption and Implementation: A Focus on Interorganizational Linkages.” Information Resources Management Journal 5(1). Somasundaram, R. and J. Rose (2003). Rationalizing, Probing, Understanding: The Evolution of the Inter-Organizational Systems Adoption Field. Proceedings of the Thirty-Sixth Annual Hawaii International Conference on System Sciences (HICSS36), Hawaii, USA. Walsham, G. (1995). “The Emergence of Interpretivism in IS Research.” Information Systems Research 6(4): 376-94.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Page 11
King, J. L. (1983). Centralized Versus Decentralized Computing: Organizational Considerations and Management Options. Computing Surveys, 15(4). Kumar, K., V. H. G. Dissel and P. Bielli (1998). The Merchant of Prato - Revisited: Toward a Third Rationality of Information Systems. MIS Quarterly, 22(2): 199-226. MacManus, S. (2002). Understanding the Incremental Nature of E-Procurement Implementation at the State and Local Levels. Journal of Public Procurement, 2(1): 5-28. McCue, C. P. and J. T. Pitzer (2000). Centralized vs. Decenntralized Purchasing: Current Trends in Governmental Procurement Practices. Journal of Public Budgeting, Accounting & Financial Management, 12(3). NIGP (1996). National Association of State Purchasing Officials/National Association of State Information Resource Executives Joint Force on Information Technology and Procurement Reform. Stepney, P. (2000). E-governance: Weber's Revenge? Political Studies Association - UK 50th Conference, London. Thai, K. V. and R. Grimm (2000). Government Procurement: Past and Current Developments. Journal of Public Budgeting, Accounting & Financial Management, 12(2): 231 - 247. Vagstad, S. (2000). Centralized vs. Decentralized Procurement: Does Dispersed Information Call for Decentralized Decision-Making. International Journal of Industrial Organization, 18. Zulfiqar, K. A., S. L. Pan, J. N. Lee and J. C. Huang (2001). E-Government: An Exploratory Study of On-line Electronic Procurement Systems [Case Study]. The 9th European Conference on Information Systems, Bled, Slovenia.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Appendix 12.3 Research Publication Three – the ECIS 2004 Diffusion of E-Procurement in the Public Sector – Revisiting Centralization versus Decentralization Debates as a Twist in the Tale By Ramanathan Somasundaram
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
DIFFUSION OF E-PROCUREMENT IN THE PUBLIC SECTOR – REVISITING CENTRALIZATION VERSUS DECENTRALIZATION DEBATES AS A TWIST IN THE TALE Somasundaram, Ramanathan, Aalborg University, Nørre Voldgade 8 3, 1358 København K, Denmark,
[email protected]
Abstract Several governments are in the process of implementing e-procurement. In the process they face several challenges. The diffusion of e-procurement in the Danish public sector is researched using case study research methodology to study the challenges faced. The results of the study are applied in this paper for explaining some of the challenges faced using the legacy centralization versus decentralization debates regarding computing and organization; which forms the tale. These debates in no doubt help in kick starting the research on the diffusion of e-procurement in the public sector. However, the phenomenon under study is complex and unique in several respects that the debates alone are inadequate for fully explaining the challenges. The paper presents a question as the twist in the tale answering which helps in understanding the nature of challenges much better. The question is “To what extent is government a single organization?” Keywords: E-procurement, Public sector, E-government, Diffusion, Implementation, InterOrganizational Systems, Electronic Markets.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
1
INTRODUCTION
“A twist in the tale” is a collection of short stories written by Jeffrey Archer. The stories are written such that the reader realizes the true meaning of the story while reading the last few lines. Logic similar to that of the Archer’s is adopted for structuring this paper. The traditional centralization versus decentralization debates regarding computing and organization are applied in this paper for explaining some of the challenges faced during the diffusion of e-procurement in the public sector; which forms the tale. The paper presents a question as the twist in the tale answering which helps in understanding the nature of challenges much better.
2
THE STATE OF E-PROCUREMENT IN THE PUBLIC SECTOR
The public sector spends between 10 and 20 percent of its earnings in procurement activities (Thai and Grimm 2000; Jones 2002; Anderson et al. 2003; Goerdeler 2003). This is a whopping amount when converted to dollar terms. European Union for instance procures for more than 720 billion Euros annually (EU&PP 2003). E-procurement promises significant savings through increased efficiency and effectiveness. Several government organizations that represent regions and countries have embarked on e-procurement for realizing the promised savings. E-procurement in the public sector is internationally emerging. Countries such as Denmark, Norway, Finland, Ireland, the United Kingdom, Spain, Germany, Portugal, Italy, Singapore, Brazil and the regions in the USA and Australia have embarked on e-procurement. Most or even all of the countries are still in the implementation stages. Not all countries diffuse e-procurement similarly; instead there are several diffusion patterns. For instance, the Danish central government has chosen a private emarket as the infrastructure for e-procurement (efkous 2003). The Spanish ministry of public administration has taken the role of defining functional, technical and organizational specifications (Juan 2002). The German government has invested 4.5 million Euro in developing a “flag ship” project e-vergabe, which it touts as the “model for public procurement in Europe” (Goerdeler 2003).
3
E-PROCUREMENT DEFINITION
E-procurement in the context of this paper does not imply just the technical system but also the organization surrounding it. “Build it they will come” or “IT as a magic bullet” are not the kind of assumptions held while using e-procurement. An e-procurement project is regarded successful only when the end users in public organizations procure electronically. MacManus’s (2002) recognition of the differences in meaning between “procurement” and “purchasing” is of relevance here. Quoting the dictionary of purchasing terms, MacManus (2002) explains procurement as denoting “…the combined functions of purchasing, inventory control, traffic and transportation, receiving and inspection, storekeeping and salvage and disposal operations” (NIGP 1996). Purchasing on the other hand is narrowly defined as “the act and the function of responsibility for the acquisition of equipments, materials, supplies and services. [Purchasing] describes determining the need, selecting the supplier, arriving at a fair and reasonable price and terms, preparing the contract or purchasing order, and following up to ensure timely delivery” (NIGP 1996). Contemporary e-procurement systems in the public sector facilitate the narrow set of activities defined as under purchasing and not that of the broader procurement. E-procurement as in the NIGP definition sense is more of a vision than it is reality.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
4
PROBLEM PRESENTATION
E-procurement in the public sector has a lot of promise. A government can negotiate good deals with suppliers when it is able to accumulate all of its purchasing and negotiate as an entity. The negotiated agreements when made available in the e-procurement system can be accessed by the end users by logging into the system via the ubiquitous internet. The end users can place the order electronically utilizing the negotiated deals. The e-procurement system when interconnected with an organization’s internal financial system minimizes the transaction processing costs. E-procurement via automating several administrative procedures and by enhanced monitoring abilities minimizes opportunistic behavior among purchasing officials. There however are several obstacles that are to be overcome for realizing the promise. Firstly, how should the e-procurement system be developed? For instance, should the government embark on a flag-ship project or should it let the concept evolve or should it outsource the project to private organizations or should it restrict its involvement only to the role of defining specifications? Secondly, an organization is able to realize the full benefits of procurement only when its internal financial system is interconnected with the e-procurement system. The interconnection however is a challenge as there are disparate internal information systems in a government. Thirdly, several governmental organizations have decentralized procurement practices adopting the trendy new public management (NPM) rationale (Anderson et al. 2003). The decentralization in some organizations is so deep that each individual does his or her own procurement. For establishing e-procurement there definitely is a need for centralizing decision making. The challenge in the centralizing is that power when let gone is difficult to get back. Finally, government organizations perform a diverse set of functions. The adequacy of a standardized procurement infrastructure to satisfy such diverse needs comes into the limelight. Though e-procurement has only recently become a research topic, there are several research strands that the research community can rely upon explaining the above described challenges. The strands that we can rely upon include i) public administration ii) information systems iii) procurement science iv) electronic markets and v) Inter Organizational Systems (IOS) adoption science. A few research studies in the recent years have attempted explaining the e-procurement phenomenon. Zulfiqar et al. (2001) explains the challenges faced during the implementation of e-procurement in the Singaporean context via an explorative study. Anderson et al. (2003) enquire the use of e-procurement in the Danish local governmental organizations through a statistical study. They make a methodological contribution in the process by questioning the notion of querying the “professional buyer”. There are several other papers that tackle issues relevant to the research area. For instance, Vagstad (2000) makes an economic analysis querying the effect of local intelligence on decentralizing decision making in the government context. Jones (2002) provides an excellent descriptive account of purchasing practices in the Singaporean government. Gordon and Walsh (1997) analyze in details the challenge of outsourcing information technology in governmental organizations. E-procurement research area at this stage requires an explorative study that not just analyzes the development insightfully but also lays the ground work for theory based accumulative work. This paper analyzes the results of an on-going research studying the diffusion of e-procurement in the Danish public sector using the legacy debates on centralization versus decentralization of computing and the impact of computing on centralizing versus decentralizing decision making in organizations.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
5
STRUCTURE OF THE PAPER
The case study research method adopted is explained in details in the next section. Then, relevant details about the Danish e-procurement scene are presented. Three types of centralization versus decentralization debates are identified in the e-procurement context in the section that follows. They are •
centralization vs. decentralization of decision making at an organizational level
•
centralization vs. decentralization of decision making at an inter-organizational level and
•
centralization vs. decentralization of computing infrastructure
Accumulated knowledge on each of the three debates is applied for analyzing the nature of the challenges faced in the e-procurement context next. An insight drawn from the analyses is presented in the final section as the twist in the tale.
6
METHODOLOGY
This paper results from an on-going three year long study. The study’s objective is to research the diffusion of e-procurement in the Danish public sector. The objective of the study so far has been to document and analyze the Danish e-procurement effort based on an explorative study. An explorative study is adopted for there is little theory based knowledge available on the e-procurement subject. The case study methodology is chosen for it is suitable for studying newly emerging research areas. Eisenhardt (1989) states that case study research “…given the strengths of this theory-building approach and its independence from prior literature and past empirical observation, it is particularly suited to new research areas or research areas for which existing theory seems inadequate” (p. 548). The case study method can be adopted i) to provide description ii) to test theory and iii) or to generate theory (Eisenhardt 1989). The method is used in the context of this study primarily to provide description and to generate theory. The research is a community level study in the sense several types of actors’ involvement in the Danish e-procurement project were studied just as the one adopted by Kumar et al. (1998). The types include i) buyers ii) sellers iii) infrastructure providers iv) trade associations and v) the strategic planners. Semi-structured interview is the prime method adopted for studying the actors’ involvement. A set of question areas is usually e-mailed to the interviewee pre-interview. These question areas are identified from a detailed literature review of several research strands. The interview is done with the objective of learning about the question areas. The interviewer at the same time was open to new leads arising from the conversation. The knowledge obtained from a conversation is taken forward to the next conversation and thus the Danish procurement context was researched. Apart from the interviews, secondary sources such as home pages, presentations, news papers and internal reports were sought.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
7
THE DANISH E-PROCUREMENT SCENE
7.1
The Country and Its Administration
Denmark is a country with a population of 5.4 million. The government’s expenditure accounts for 25% of its GDP. It is one of the least corrupt countries in the world. Global Corruption Report (2003) ranks it the second least corrupt country in the world (Anonymous 2002). The Danish government is tri-layered. The layers are as per the hierarchy i) ministries ii) regional bodies (amter) and iii) local bodies (kommune). There are 18 ministries, 14 county organizations and 275 local bodies. The local bodies account for 50% of the nation’s expenditure. They however are largely self sufficient for they collect 33% of the nation’s tax revenues (sources; Kl.dk, Denmark.dk, folketing.dk).
7.2
Initiating E-procurement Project
Denmark’s yearly procurement volume amounts to app. 15 billion USD of which a billion USD worth goods and services are immediately suitable for e-procurement. It is the first country to have embarked on e-procurement in Europe (DOIP accessed on Nov. 2003). The ministry of science, technology and innovation came up with the idea of e-procurement as an officer involved in the process explains; “Basically it was three four years ago1 we started seeing a marketplace (fails to recollect the name) but we looked at that and got the idea that may be we could make a marketplace for the public sector as well. And then we started; we presented it for the minister and we presented for the government and they agreed that we should try something like it. Because, it was the opinion that public sector could save some money by doing e-commerce and hardly any in the public sector had started using ecommerce. So, we thought that making the infrastructure ready with a portal, we could push the development of e-commerce… That is the whole idea. Because when we ask the ministries and the communities why they were not buying things electronically, they would say well it is too expensive, it is too difficult, what systems should we choose? The public sector cannot use this and a lot of excuses. If we make sure that they have the infrastructure and the right conditions at least they don’t have that excuse.” The government called for proposals for developing and hosting electronic trading infrastructure over which public bodies can procure. Out of five submissions, Gatetrade, a consortium formed by four large Danish organizations (Maersk data, Tele Denmark, Post Denmark and Danske Bank) was selected as the winner on the 21st of March 2001. The ministry suggested the use of Gatetrade as the trading infrastructure to the rest of the government.
7.3
About Gatetrade
The portal went live on the 3rd of Jan. 2002. Seven pilot organizations traded over its infrastructure in its early days. The portal was open to all since mid 2002. The infrastructure is open not just for the public sector but also for the private sector. It has progressively increased the number of customers and the line items traded over its infrastructure. As of Oct. 2003, it has over 1400 accessing customers and a million line items. Despite which the organization is not yet breaking even. The prime reason for the organization not breaking even is that the local and regional bodies have not traded over its infrastructure as it was foreseen during the conceptualization stages. Except for a couple of ministries, 1
The interview was conducted during March 2003.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
the rest have taken steps towards trading over its infrastructure. The owners are strongly backing the initiative. Gatetrade offers three types of services on the buy side. They are i) Gatetrade direct ii) Gatetrade match and iii) Gatetrade collect. Gatetrade direct is the simplest solution of the three. It is a basic browser based solution. It is the most commonly adopted solution so far. Gatetrade match is a slightly advanced solution for the connecting organizations can download trade documents electronically and thereby seamless interconnect. Gatetrade is actively marketing this solution currently to its customer base. Gatetrade collect is the most advanced solution of all in terms of integration possibilities. Organizations that have access to the Gatetrade collect solution usually integrate their ERP systems with that of the marketplace. The end users punch out of their ERP systems and collect the data that they require from the market place and bring it back via the established integration. It is hence the solution is termed as Gatetrade collect. Only the organizations with large transaction volume find this solution feasible.
7.4
SKI (National Procurement Agency)
SKI was established as a joint venture between the ministry of finance (55%) and the national association of local authorities in Denmark (45%) a decade or so back. It negotiates frame agreements on behalf of all Danish governmental organizations. The buying organizations can choose using SKI negotiated agreements instead of self negotiating. SKI however has not been able to get the volume discounts as a representative of a large buying organization would be able to for it has not been able to convince the end users in utilizing the negotiated agreements. The sellers are unwilling to provide the same kind of deal that they would provide to a large customer due to this lack of utilization. SKI is thus facing the typical “catch-22” challenge. The SKI negotiated agreements are made available in Gatetrade. The end users can thus be aware of the agreements and utilize the negotiated deals by just a click of a button. SKI works closely with Gatetrade to the extent that some SKI executives are even in Gatetrade’s steering committee. A SKI executive explains their collaboration with Gatetrade as follows;
“…actually because it is such a big and important project, our owners in the Danish state wanted us to be a part of it and be involved in it, and that means that we are part of the steering committee and part of some of the project teams that involve DOIP, and therefore we have tight relations with Gatetrade.net”.
7.5
Grass Root Initiatives
Several bottom-up initiatives have propped up while the top-down Gatetrade has been initiated. Two of the prominent grass root initiatives are explained in this section. The first is Kubus (www.kubus.dk). It is a small sized innovation oriented company that was started in 1988. It is owned and run by two brothers; Søren and Steen Rågård. During the mid-90’s the company embarked on developing content management systems. Then onwards it has developed a rack of systems. Kubus Tradebuilder is an e-procurement related solution that won the best e-business system prize (e-handelsprisen) awarded by a consortium of leading organizations for the 2003 year. Søren
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Rågard, the technical director of the company explains as to how Kubus got into developing the Trade builder system;
“In about the last quarter of year 2000, we got involved in two Kommune (municipalities). They were doing e-procurement or procurement together where the institutions and schools were buying at the same partners. They wanted to make the catalogue they had in paper; they wanted to make it electronic… and then we focused more on the (frame) agreements you have…. All agreements, all those you cannot procure on the Internet, also them you have to call him on the phone, those you have to print out some reports and fill out with a pen; 100%, one place you find it all. So, you don’t have the portal of services you can buy on the Internet. But you also have all the other agreements and the things you have to do at one place.” The trade builder solution is built on the recognition that there is a need for an electronic solution for managing frame agreements regardless of whether the transaction is carried out electronically or not. Kubus hosts a supporting .net infrastructure via which suppliers and the end users can interact about experiences in using the product during the post sale phases. Thus, while Gatetrade is building upon its transaction focus towards other aspects, Kubus is building upon the content management and the post sale aspects of procurement. KMD’s (Kommune Data) web indkøb (purchase) is the second grass root initiative. Local bodies (Kommune) collectively started KMD for addressing their similar IT needs. KMD Ø is a financial system that is run in most of the local bodies. The web indkøb system is tightly integrated with the local bodies’ internal financial system. The local bodies apparently have long sought KMD for a procurement module. A chief officer in KMD explains the history behind developing the web indkøb system as follows; “One of the key demands from the municipalities at that time and still today is; KMD you must develop a procurement system which is closely integrated to our financial system. Because the whole process is about ordering, about the invoice; all related back to our financial controlling system…This is our situation. We are building iprocurement system that is at a minimum (here meaning very) closely related to the financial. Because we know that is what the municipalities want and that is what they will pay for”
8
CENTRALIZATION VERSUS DECENTRALIZATION DEBATES IN THE EPROCUREMENT CONTEXT
The exploratory study highlighted several challenges that are to be overcome for diffusing eprocurement. For example, how can procurement related decision making be centralized at an organizational level? How can SKI negotiate better deals and as well encourage the use of the negotiated deals? Does the Danish government require a representative such as SKI at all for negotiating frame agreements? How can Gatetrade obtain adequate participation so it attains selfsustainability? How best to incorporate the innovative ideas that grass root initiatives such as Kubus and KMD bring about?
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
The diffusion of e-procurement in the public sector is a sensitive topic. Discussing each of the above mentioned questions can result in several opinions. Just opinions alone are inadequate for scientifically advance the research field. A theory based explanation is instead required. The following historical centralization versus decentralization debates are used for understanding the challenges in perspective. •
centralization vs. decentralization of decision making at an organizational level
•
Centralization vs. decentralization of decision making at an inter-organizational level and
•
Centralization vs. decentralization of computing infrastructure
A key insight realized from the discussion forms the twist in the tale. Discussion under each of the debates in the Danish e-procurement context is presented in this section.
8.1
Centralization vs. Decentralization of Decision Making at an Organizational Level
The centralization versus decentralization debate at an organizational level is about who in the organization will decide on the goods and services that an organization requires and from which supplier should they be procured from. McCue and Pitzer (2000) explain the various combinations along which purchasing can be centralized. In one extreme, there is a centralized purchasing authority to whom end users send their requests. The centralized authority authorizes the requests, identifies the suppliers, negotiates prices and makes the purchasing decision. In the decentralized extreme, end users make their own purchasing decision. Procurement authority is both centralized and decentralized among the Danish governmental organizations. Procurement is a centralized activity in the Danish tax office and it has remained that way for the last 40 years (source: interview with a purchasing officer in Aarhus tax office). In contrast, procurement is a highly decentralized activity in the Danish universities. It however used to be centralized like a decade back but since then it has gradually become decentralized. A secretary in a Danish university explains procurement practice as follows; “We thought so (that agreements made by a central purchasing body would be cheaper). But in a couple of years, we started checking on the prices ourselves. If we buy furniture directly from furniture stores, we can get a lot cheaper than if we buy from the university because her salary needs to be paid. If they get 500 DKK table and sell it for 550 DKK. Then the 50 is for her salary... It is what they give and what they take… We would not allow that. So we buy it ourselves.” Both the centralized and the decentralized structures have their respective advantages and disadvantages. An organization is able to accumulate its purchase and negotiate volume discounts involving professional expertise when the decision authority is centralized. However, the central authority can take a good decision only when it is well aware of the end user’s requirements. Gurbaxani and Whang (1991) term the cost of informing the central authority and the lack of it as decision information costs. Decision information costs tend to be high when the purchasing authority is centralized and low when decentralized. Organizations are not able to obtain volume discounts when purchasing authority is decentralized. The end users when given the responsibility can act in a self serving manner instead of serving the organization, the risk of which is termed as the agency costs. The agency costs can be minimized through monitoring, which costs as well.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
The use of IT for automating procurement procedures alters the arguments for centralization and decentralization. Those arguing for centralization can cite the role of IT in reducing the decision information costs. The end users can inform their requirements via a workflow system based on which the central purchasing authority can negotiate frame agreements. The end users can access negotiated frame agreements posted in the intranet or in an e-market simply via a desktop. The central purchasing authority can use business intelligence tools for learning about the needs of an organization and thereby is in a position to negotiate better frame agreements. It is thus the implementation of eprocurement system supports the centralization. Those arguing for decentralization can cite the role of IT in minimizing the cost of monitoring and thereby the agency costs. The effect of IT in centralization versus decentralization of organizational structures has been a lengthy one. Gurbaxani and Whang (1991) provide a seemingly good solution to the debate. They recommend choosing the organizational structure in which the internal co-ordination costs, which is a sum of agency and decision information costs, are at a minimum. The centralization of decision making at the organizational level is recommended in this paper for e-procurement is expected to reduce decision information costs much more than it would the agency costs. Such a claim is made despite Denmark being one of the least corrupt countries in the world. Centralization however is a challenging task for power when let gone is difficult to get back (King 1983).
8.2
Centralization versus decentralization of decision making at an inter-organizational level:
Government organizations can negotiate better frame agreements when they accumulate their purchasing volumes via joint ventures. Such a trend is prevalent. Thirty three states in the US formed an alliance termed “Western State Contracting Alliance” (WSCA) whose motto is “every body benefits from the use of cumulative volume discount contracts” (http://www.aboutwsca.org/). Bartle and Korosec (2003) identify several such efforts made by the US states. In the Danish context, Kubus embarked on building the trade builder system when two local bodies that procure as one entity sought its assistance for developing a catalogue management system. SKI was formed to negotiate agreements on behalf of the whole Danish government. The science behind this debate is quite similar to that of the previous debate. It is just the level of analysis that changes. Those arguing for centralization cite the advantages of reduced decision information costs. Those arguing for decentralization unlike the previous have a much stronger argument. Firstly, the local bodies would like to support their local economy by awarding government contracts to organizations based in their area. To remind the reader, local bodies obtain as much as 66% of their revenues from local taxes and the Danish government’s expenditure accounts for 25% of the nation’s GDP. The local bodies thus would want to maintain their autonomy or enhance it. Secondly, only the large suppliers have the ability to provide for the requirements of a large buyer such as that of a government. Small suppliers when unable to compete with the large might not survive especially when they are highly dependent on government orders. Government traditionally has taken efforts to encourage small and medium sized companies and not discourage their existence. Thirdly, adopting new public management (NPM) logic authority has been decentralized during the last decade to local governments such that service is generated as close to the citizens (Anderson et al. 2003). An attempt to centralize procurement is against this ideology. Finally, the local bodies query the notion of being monitored at the governance level. This debate is much more complex than the previous one. It is difficult to apply the solution that Gurbaxani and Whang (1991) suggest for resolving this debate. Local bodies for instance query if at
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
all they should be monitored. The issue is not just about efficiency but also about effectiveness. Governance is not just about making economically rational decisions as the Cybernetic theory explains. There are as well the adverse effects associated with rationalization that are to be taken into consideration (Stepney 2000).
8.3
Centralization vs. decentralization of computing infrastructure
The debate about whether to centralize or decentralize the computing infrastructure is a traditional one. King’s (1983) summary of the debate is very representative of the topic. The debate has originated in the e-procurement context as well. In the Danish context, should all governmental organizations adopt the centralized Gatetrade infrastructure or should they all develop custom solutions. The two sides of the debate are hereby presented; Those supporting Gatetrade as the standard argue as follows; we should regard government as a single entity. Applications required for e-procurement are quite similar. Why then should multiple infrastructures for the same purpose be developed by the same entity? E-procurement infrastructure is not only costly to develop but also to maintain and further advance. It is prudent that investments be made in one sophisticated system that seamlessly connects with all of the government. This view assumes that all involved actors are rational and that they all share the same ideas on improving government performance. The other side supporting decentralized computing has the following arguments; unlike private enterprises for which efficiency is the sole evaluation criteria, governments as well have to take effectiveness into consideration. Government is such a complex entity that one system cannot possibly satisfy all of its requirements. As an end user in a large Danish university puts it
“There are 22 departments. It is like 22 different firms. Some are old fashioned .We have a lot of responsibilities here. We can buy all our things. In some departments it is institute leader like Peter who is buying but not here. There are different rules dependent on where you are hired.” The proposed centralized system does not satisfactorily interconnect with our internal financial system for now2. When supporting a single system we do not encourage the emergence of innovative grass root initiatives such as that of Kubus. This is not good for the nation as a whole. There clearly is a need for an organization specific or a community specific system. The fact that the proposed standard is owned by a private organization does not help the cause either. King (1983) explains that the extent to which the computing infrastructure is centralized or decentralized is not about “which way is best?” It usually is “whose way is it going to be?” In the Danish context, controls have increasingly been decentralized during the last decade. Thus proposing a centralized IT infrastructure does not fit well with the norms. The legacy KMD Ø financial system and the contemporary Kubus system as well are to be taken into account while deciding on the debate. There however is no disagreeing that the Danish government should enhance its efficiency on procurement activities. Just allowing the systems to evolve and hope that the issues will resolve it self 2
A similar problem is faced in the Singaporean context as well describes Zulfiqar et al. (2001).
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
is not a recommended option. Increased efficiency is possible only when the diverse systems interconnect with one another. Hence, it is recommended that a committee is formed in where the involved actors interact and decide upon the best mode for proceeding further.
9
A TWIST IN THE TALE
The tale so far has been that the public sector faces several challenges while implementing eprocurement. Several of these challenges are about centralization versus decentralization of decision making and computing as it is explained in this paper. Three categories of debates identified in the paper based on a case study of the Danish public sector have to be resolved satisfactorily for the successful diffusion of e-procurement; centralization versus decentralization of decision making at an organizational and inter-organizational levels and about the computing infrastructure. The centralization-decentralization debates have been on-going for several decades. The knowledge about these debates is applied for understanding the nature of the challenges faced while implementing eprocurement. In no doubt the accumulated knowledge has helped in kick starting the diffusion research on e-procurement in the public sector. Several public sector specific issues have been included for enriching the debate such as the local bodies’ desire to be autonomous, the importance of effectiveness and not just efficiency and the need for encouraging small suppliers. But why then do these debates originate? The answer to this question which forms the twist in the tale is “these debates originate due to differences in perception about the extent to which government is a single organization.”
10
LITERATURE LIST
Anderson, K. V., N. C. Juul and J. K. Pedersen (2003). Fractional Institutional Endeavors and e_procurement in Local Government. 16th Bled Electronic Commerce Conference, Bled, Slovenia. Anonymous (2002). Global Corruption Report. http://www.globalcorruptionreport.org/download/gcr2003/24_Data_and_research.pdf. Bartle, J. R. and R. L. Korosec (2003). A Review of State Procurement and Contracting. Journal of Public Procurement, 3(2): 192-214. DOIP (accessed on Nov. 2003). http://www.doip.dk/default.asp. efkous (2003). Baggrundesmateriale - DOIP, http;//www.efokus.dk/. Eisenhardt, K. M. (1989). Building Theories from Case Study Research. Academy of Management Review, 14: 532-550. EU&PP (2003). Public Procurement in the European Union. Goerdeler, A. (2003). Electronic Public Procurement in Germany. Business Briefing: Global Purchasing & Supply Chain Strategies. Gordon, M. L. and T. P. Walsh (1997). Outsourcing Technology in Government: Owned, Controlled, or Regulated Institutions. Journal of Goverment Information, 24(4): 267 - 283. Grieger, M. (2003). Electronic Marketplaces: A Literature Review and a Call for Supply Chain Management Research. European Journal of Operational Research, 144: 280-294. Gurbaxani, V. and S. Whang (1991). The Impact of Information Systems on Organizatoons and Markets. Communications of the ACM, 34(1). Jones, D. S. (2002). Procurement Practices in the Singapore Civil Service: Balancing Control and Delegation. Journal of Public Procurement, 2(1): 29-53. Juan, F. M. M. (2002). Expert Meeting on the Context of the Study on Trans Border Public Procurement. Presentation.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
King, J. L. (1983). Centralized Versus Decentralized Computing: Organizational Considerations and Management Options. Computing Surveys, 15(4). Kumar, K., V. H. G. Dissel and P. Bielli (1998). The Merchant of Prato - Revisited: Toward a Third Rationality of Information Systems. MIS Quarterly, 22(2): 199-226. MacManus, S. (2002). Understanding the Incremental Nature of E-Procurement Implementation at the State and Local Levels. Journal of Public Procurement, 2(1): 5-28. McCue, C. P. and J. T. Pitzer (2000). Centralized vs. Decenntralized Purchasing: Current Trends in Governmental Procurement Practices. Journal of Public Budgeting, Accounting & Financial Management, 12(3). NIGP (1996). National Association of State Purchasing Officials/National Association of State Information Resource Executives Joint Force on Information Technology and Procurement Reform. Stepney, P. (2000). E-governance: Weber's Revenge? Political Studies Association - UK 50th Conference, London. Thai, K. V. and R. Grimm (2000). Government Procurement: Past and Current Developments. Journal of Public Budgeting, Accounting & Financial Management, 12(2): 231 - 247. Vagstad, S. (2000). Centralized vs. Decentralized Procurement: Does Dispersed Information Call for Decentralized Decision-Making. International Journal of Industrial Organization, 18. Zulfiqar, K. A., S. L. Pan, J. N. Lee and J. C. Huang (2001). E-Government: An Exploratory Study of On-line Electronic Procurement Systems [Case Study]. The 9th European Conference on Information Systems, Bled, Slovenia.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Appendix 12.4 Research Publication Four – ECEG 2004/EJEG 2005 Government’s Role in the Diffusion of E-Procurement in the Public Sector By Ramanathan Somasundaram and Jan Damsgaard
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Government’s Role in the Diffusion of E-Procurement in the Public Sector Ramanathan Somasundaram Aalborg University Frederik Bajers Vej 7, Dk-9100, Aalborg, Denmark
[email protected] Jan Damsgaard Copenhagen Business School Howitzvej 60, DK-2000 Frederiksberg, Denmark.
[email protected]
Abstract The role played by governmental institutions for accelerating the diffusion of e-procurement in the public sector is analyzed in this paper. Such an analysis is interesting for institutions encouraging the diffusion of e-government are not objective third party intermediaries instead are part of the government. The paper is written based on a single embedded case study carried out to enquire the challenges faced by the Danish public sector in the diffusion of eprocurement. The actions taken by the ministry of science, technology and innovation in Denmark are analyzed as under the following sections; knowledge building, knowledge deployment, subsidy, mobilization, standard setting and innovation directive. The analysis shows that as public administration is politically managed, government is mainly able to influence and not regulate both the supply and the demand sides. A regulatory action can be misinterpreted as an effort to alter power structures within the public administration.
Key words: e-procurement, e-government, public sector, diffusion, policy, inter-organizational systems and institutions.
1
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
1. Introduction E-government is a buzzword that has gained prominence in the last few years. There are several competing definitions of e-government. Zulfiqar et al. (2001) provide a summary of e-government definitions. Researchers participating in the first Scandinavian workshop on e-government held at Ørebro, Sweden in Feb. 2004 and in the e-government workshop at the London School of Economics, March 2004 discussed the topic. The discussions did not lead to a clear-cut definition. They instead resulted in broadly interpretable notions such as “we do not know what it is yet” (Taran 2003) and “there is nothing as such as egovernment; it is the diffusion of contemporary IT in the public sector we study”. Most agree that public e-procurement is a central theme in e-government. The attention of this paper is limited to the diffusion of e-procurement in the public sector. Specifically, the role played by institutions in encouraging diffusion is analyzed. Of the several e-government initiatives, the rationalization of e-procurement functionality holds enormous promise. Denmark, a country with around 5.5 million population, procures goods and services for app. 100 billion DKK annually (app. 15 billion USD) (DOIP.dk). A report by KPMG consulting (2000) identifies 8 billion worth of goods and services of the 100 immediately suitable for e-procurement. Market research reports show that businesses via e-procurement can save between 10 and 50% (Peria 2003). The savings are made mainly from better sourcing decisions and via reduced administrative costs (Ageshin 2001). Several studies show that a significant percentage of savings to be obtained from e-procurement arises from better sourcing decisions than via reduced administrative costs (Baker 1999). The public sector characterized by high purchasing volume, maverick buying and the lack of transparency stands to benefit enormously from e-procurement. The private sector driven by the desire for maintaining competitive advantage and by the need to maintain profitability has taken rapid strides in using eprocurement (Krysiak et al. 2003). E-procurement in the public sector is being implemented worldwide and much money is spent to build up and implement eprocurement solutions. Yet we know surprisingly little about how government engage itself in furthering public e-procurement, let alone what lessons can be learned from their involvement. E-procurement in this paper is regarded an innovation, diffusion of which brings good to a society. As per the institutional point of view, society comprised of institutions modulates the diffusion of innovation and gets reformed as innovations are diffused. Institution as per King et al. (1994) is “any standing social entity that exerts influence and regulation over other social entities”. An innovation need not just be an object but could also be an idea or a practice (Rogers 1995). The role played by governmental institutions has been the focus in most diffusion of innovation (DOI) research. Damsgaard and Lyytinen (2001) differently analyze the role played by industry associations in the diffusion of EDI. Speilman (2002) looks at the role played by profit oriented multi national enterprise in the diffusion of agricultural technologies.
2
Diffusion of Electronic Public Procurement in Denmark - Exam Copy A characteristic common to all of the above is that the entity exerting influence or regulation more often than not is a governing third party. The actions of industry associations, ministerial bodies and technology standards organizations affect others more than they affect self. The scenario however is different in the e-government context for the entity upon which influence and regulation is exerted upon and the entity that exerts more often than not belong to the same public body. Those affected by actions recommended or mandated for accelerating the diffusion of e-government tend to critically evaluate the actions as efforts taken towards altering the administration of the public sector. Given the emerging nature of the study area, there is a lack of documentation of the challenges that public sector institutions encounter while attempting to accelerate the diffusion of e-government. Those implementing e-government would find tools and techniques for analyzing their activities and a critical analysis of their action beneficial. Institutions involved in the diffusion of e-procurement in the public sector i)
ii) iii) iv) v)
vi)
The World Bank, Inter American Development Bank and the Asian Development Bank have formed a workgroup to provide guidance in implementing e-procurement to mainly developing countries (Mdb-egp.org). The European Union has been actively persuading its member nations to implement eprocurement (Simap.eu.int). Private owned businesses when involved in hosting e-government procurement infrastructure actively market their products and services (Hoodevik et al.). The role of consultants in informing the public sector about savings that can potentially be obtained via e-procurement can in no way be ignored (KPMG 2000) and In some countries, a part of the government such as the ministry of finance in the Singaporean government (Zulfiqar et al. 2001) is self driven towards the diffusion of eprocurement. Solution providers such as PeopleSoft (2001) and IBM (IBM-IEG) have taken special interests in informing the public sector about e-government’s potential
Figure 1: Institutions involved in the diffusion of e-procurement in the public sector
Documented knowledge regarding the diffusion of e-procurement in the public sector is mostly anecdotal in nature. The main source of information on the current state of happenings is obtained from magazines ((Emarketservices.com; Kablenet.com), status reports provided by governmental organizations (Anonymous 2001; Eprocurementscotland.com) and presentations (Court 2003). Research oriented publications on the topic is sparse and so far appeared mainly in conferences and as working papers ((Anderson et al. 2003; Coulthard and Castleman 2001; Heneriksen et al. 2004; Ramanathan 2004; Tonkin 2001; Zulfiqar et al. 2001). While these papers most certainly have made insightful contributions, they do not analyze the role of an institution in the diffusion of e-procurement. Figure one lists institutions involved in the diffusion of e-procurement in the public sector. The focus in this paper is on the role played by governmental organizations for accelerating the diffusion of e-procurement in the public sector. Dimensions for
3
Diffusion of Electronic Public Procurement in Denmark - Exam Copy institutional intervention as defined by King et al. (1994) is used as the framework for analyzing specific actions that institutions might engage in. The X-axis of the framework is defined by supply push and demand pull and the Y-axis is defined by influence and regulation. The research methodology adopted for collecting data about the diffusion of eprocurement in the Danish public sector is explained first. Second, information about the Danish public sector that is relevant for this paper is presented. Third, variables that define King et al. (1994) institutional intervention framework are explained. Fourth, actions that the ministry of science, technology and innovation has taken are analyzed within the framework’s province.
2. Research Methodology This paper results from a single embedded case study (Eisenhardt 1989; Yin 1994) carried out to enquire the diffusion of e-procurement in the Danish public sector. The study was embarked upon during August 2001 and it runs for duration of three years. The following are the central objectives of the study; i) ii)
to provide rich descriptive data on the challenges faced in the diffusion of eprocurement and to identify actors involved and define their involvement in the diffusion of eprocurement.
The study enquired buyers, sellers, policy makers, solution providers and governmental associations about their involvement in the diffusion of e-procurement. The enquiry was made using semi-structured interviews. The questions posed during the interviews were informed by a detailed literature review and secondary research about the organization enquired. The areas covered by the literature review include i) inter-organizational systems adoption ii) public procurement iii) e-procurement in the private sector and iv) public management research. The interviewer willfully queried further when reasoning not deducible from the literature resulted. As the involvement of national level policy makers, organizational level decision makers and the end users in the Danish government is analyzed, the study is multi-level in nature. The research design is similar to Kumar et al.’s (1998) analysis of SPRINTEL System in the Italian district of Prato.
3. E-procurement in the Danish Public Sector In terms of population Denmark accounts for 1.4 percent of Europe. The public sector plays an important role in the Danish economy. One third of the workforce is employed in the public sector (Denmark.dk). The government’s expenditure amounts to around 25% of the GDP. Danish per Capita GDP is one of the highest at $29,700 USD (USAdk.org). In the 2002 year, Global Corruption Report (Lamsdorff 2002) ranks Denmark as the second least corrupt country. It is a welfare state with the notion of collectiveness prevalent. For instance, Denmark was one of the few founding members in developing
4
Diffusion of Electronic Public Procurement in Denmark - Exam Copy the world’s first mobile system (Nordic Mobile Telephone system) with roaming, which then evolved to GSM (Sintef.no). The Danish public sector administration is tri layered. The three layers are i) municipality (Kommune) ii) county (Amt) and iii) the state (ministry). There are 275 local bodies, 14 regions and 18 ministries. The local bodies account for around 50% of the nation’s expenditure. They are to a large extent autonomous as local government taxes account for 33% of the nation’s tax yield. Local bodies are selected by the people and do not as such serve the state administration. Denmark is one of the early movers in implementing e-procurement in the public sector. Singapore, Australia and some states in the United States were ahead of Denmark in implementing e-procurement. Australia with its Transigo initiative in 1995 is probably the first country to have embarked on e-procurement (Coulthard and Castleman 2001). The ministry of science, technology and innovation through open competition selected a private owned electronic marketplace as the infrastructure for public procurement in 2001. The e-marketplace, Gatetrade, is equally owned by four large Danish companies; Maersk Data, Tele Denmark, Danske Bank and Post Denmark. The Danish government has not financially invested in Gatetrade. Gatetrade charges its customers a transaction fee for documents exchanged via its infrastructure. It has added a portfolio of value added services such as IT and business consulting in the recent times. Gatetrade is an open emarket that is used by both the public and the private sectors. It has been operational since Jan. 2002 (Gatetrade.com). Those involved in Gatetrade’s selection support their decision with the following reasoning; first, the state by channeling its purchase via Gatetrade will encourage the diffusion of e-commerce in the Danish society. Second, the state by accumulating purchasing volume can negotiate better frame agreements and thus enhance its operational efficiency. SKI (national procurement agency) has been negotiating frame agreements for the Danish state for over a decade. SKI negotiated agreements when made available in Gatetrade can be accessed and utilized by all governmental organizations with much ease. Third, administrative costs will go down via the enhanced use of ecommerce. Fourth, interoperability can be enhanced when all governmental organizations and sellers adopt Gatetrade as the standard trading infrastructure (Anonymous; Kiærbye). By early 2004, the Danish public sector however has not channeled via Gatetrade as much of its procurement as envisioned. Problems for Gatetrade are aggregated in that its four large owners are yet to fully utilize the e-market. Moreover, hope that the Danish business community would extensively trade via e-market has not materialized. Gatetrade has declared a loss of 32 million DKK (app. 5 million USD) for the 2003 year as against 44.2 million DKK (app. 7 million USD) for the pervious year. The owners however have declared their firm support via investing 160 million DKK (app. 27 million USD) in the initiative. They invested 60 million DKK (app. 10 million USD) as early as July 2003 (Anonymous 2003a). Gatetrade during early 2004 declared that trade via its infrastructure has quadrupled during 2003 ((Anonymous 2004b). The director Steen Gade expects Gatetrade to break even around the 2007-2008 period (Anonymous 2003b). The
5
Diffusion of Electronic Public Procurement in Denmark - Exam Copy Danish state, Gatetrade’s owners and the Danish business community in general account for 40, 30 and 30 percentage of trade conducted via Gatetrade (Anonymous 2004b). Several other initiatives are underway with which public sector organizations are able to conduct e-procurement. KommuneData’s (KMD) web purchasing module is an option widely chosen by local governmental organizations. KMD was formed by the national association of local authorities and the association of county councils in 1972 to serve local governmental organization’s IT needs. KMD promises seamless integration between its financial system and the web-purchasing module (KMD.dk). Kubus is an innovative software development firm that focuses on the contract management aspects of e-procurement. It won a national prestigious e-business prize for the 2002 and 2003 years (Anonymous 2003e) for its trade builder system. A small number of municipalities are committed towards adopting Kubus (Kubus.dk) solutions. This paper focuses on the uptake of e-procurement in the Danish public sector. Specifically efforts of those who encourage the diffusion of e-procurement are analyzed. King et al.’s (1994) institutional intervention framework is adopted in this paper. Damsgaard and Lyytinen (2001) have shown the framework’s potential through analyzing intermediating organizations’ role in the diffusion of EDI. A point that requires mentioning in this regard is that those encouraging the diffusion of e-procurement are very much a part of the public sector that they administer and not an intermediary. The implications of such are discussed in the forthcoming sections. The dimensions of institutional intervention as defined by King et al.’s (1994) are introduced next. Specific actions taken by those encouraging Gatetrade are analyzed then.
Supply push
4. The dimensions of institutional intervention
Demand pull
Influence
King et al. (1994) in their institutional intervention framework define the dimensions within which specific actions taken by an institution for encouraging the diffusion of innovation can be analyzed. The Y-axis of the framework is marked by regulation and influence. “Regulation by institutions is the indirect or direct Figure 2: Dimensions of institutional intervention intervention of behavior those under the institution’s influence, with the specific objective of modifying the behavior through sanction or other affirmative means” (King et al. 1994). The context within which regulatory and influencing actions than an institution undertakes is captured by supply push and demand pull variables in the X axis. There is both supply side and demand side when discussing the diffusion of any innovation. An institution can choose to influence or regulate either the demand side or the supply side or both via specific actions. The Regulation
6
Diffusion of Electronic Public Procurement in Denmark - Exam Copy dimensions of institutional intervention are graphically presented as a 2x2 matrix in figure two. Public sector organizations world wide are in the process of implementing eprocurement. So far the e-procurement subject just as other e-government subjects has been researched in a limited manner. Quite a lot of research done on e-procurement is anecdotal in nature (Coulthard and Castleman 2001; Heneriksen et al. 2004). However, formal research methods are increasingly being adopted. So far, case study research seems to have been the dominant research approach (Ramanathan 2004; Tonkin 2001; Zulfiqar et al. 2001). Statistical studies however are being done (Anderson et al. 2003). The area under study is complex for diffusion is usually researched at a national level with embedded units of analysis.
5. Analyzing government’s actions to encourage the diffusion of eprocurement The following specific actions that King et al. (1994) identify are used for analyzing the efforts of those encouraging the diffusion of e-procurement: knowledge building, knowledge deployment, subsidy, mobilization, standard setting and innovation directive. Each specific action is first described in general terms and then effort made by the Danish government in carrying out the action is analyzed. Finally, a conjecture is derived from the Danish experience, which can be applied elsewhere in the diffusion of e-procurement in the public sector. Knowledge building: As the private sector has tried and tested e-procurement since late 90’s there is not a need for funding basic technological research in the e-procurement case. Government however can play an important role in encouraging the use of eprocurement technologies in the public sector. The ministry of science, technology and innovation realizing e-procurement’s potential in the public sector has selected Gatetrade, a private e-market, as the standard for public procurement. It could however have adopted a different strategy such as investing in a custom developed e-procurement solution for the public sector as Singapore (GEBIZ.sg) and Germany (E-vergabe.de) have done. The Danish effort is certainly internationally acclaimed as the model for private-public partnership (Anonymous 2003d; Calway 2003). The ministry has been quite successful in encouraging private companies to invest in developing an e-market for the public sector. Gatetrade owners have so far invested 160 million DKK (app. 28 million USD) in developing the infrastructure. The e-market technology while advanced is incompatible with financial systems installed in a significant percent of government organizations. This incompatibility seems a critical reason behind resistance towards procuring via Gatetrade. It is said that a significant percent of savings from e-procurement is obtained from reduced administrative costs. Trade related data has to be manually entered when financial system is incompatible with the e-procurement module.
7
Diffusion of Electronic Public Procurement in Denmark - Exam Copy The ministry of science, technology and innovation has not adequately engaged local organizations and regions while selecting Gatetrade. The ministry proactively selected Gatetrade attempting to push the diffusion of e-procurement. Local and regional organizations that resist cite that they should have been consulted while selecting national e-procurement infrastructure. Conjecture one: A part of the government has to take proactive efforts for encouraging the diffusion of e-procurement in the public sector. In the process however, it has to adequately involve all segments of public administration while deciding upon the trading infrastructure. Knowledge deployment: Institution’s role in nurturing the development of infrastructure and skills required for the adoption and the utilization of an innovation is discussed in this section. Infrastructure required for the utilization of e-procurement includes computers and a reliable internet connection. The digital task force that comprises employees from several Danish governmental organizations mentions in its e-government strategy document (Anonymous 2004a) that IT infrastructure in the Danish public sector is quite advanced. The public sector employees use e-mail, internet and computers in general extensively in their everyday operations. Employees engaged in e-procurement however have to learn using the systems functionality and get accustomed to the system’s graphical user interface. The Danish government via choosing private/independently managed solution providers has limited its involvement in educating employees about solutions on offer. Solution providers such as Gatetrade, KMD and Kubus train their customers either for free or for a charge. Conjecture two: A government’s involvement in educating employees is minimized when the development and the hosting of e-procurement solutions are outsourced. Subsidy: Government via subsidies can encourage the supply side to invest in the development of an innovation and can minimize uncertainty perceived by the demand side while deciding on adopting an innovation. Subsidies play an important role in minimizing threshold perceived by decision makers on both the supply and the demand sides to act towards the development and deployment of innovations (Granovetter 1978). In networked innovations such as the Internet and telephones, subsidies can be crucial for attracting a critical mass of users (Oliver et al. 1985). E-procurement is a networked innovation for adequate participation from both buyers and suppliers is required for sustainable trading activity to occur via a technology standard. The Danish government has not provided financial subsidies to any of the supply organizations such as Gatetrade, Kubus and KMD. Businesses invested in the development of e-procurement infrastructure hoping to profit from government’s large purchasing volume. The Danish government via its outsourcing strategy shares subsidizing responsibilities with the infrastructure providers for attracting demand side participation. Gatetrade fully subsidized the use of its infrastructure to seven pilot organizations for the first six months of its inception. Up until now suppliers that have
8
Diffusion of Electronic Public Procurement in Denmark - Exam Copy negotiated frame agreements with SKI can host their catalogues for free in Gatetrade subsidized by the Danish government. Conjecture three: Government via outsourcing the development and hosting of eprocurement solutions can share subsidizing responsibilities with private investors. Given the networked nature of e-procurement, it is recommended that subsidies be provided to both the supply side and the demand side for attracting a critical mass of users and participation. Mobilization: Efforts taken by institutions in informing user population about an innovation is discussed under mobilization. Mobilization efforts usually propagate the benefits of adopting an innovation. A population is informed via various modes such as seminars, conferences, publications and workshops. The ministry of science, technology and innovation, local government association (Kommune Landsforeningen) and regional government association (Amts råd foreningen) have via reports and seminars been informing governmental organizations the benefits of using e-procurement. The message appears to have been effectively delivered. The ministry of science, technology and innovation has been advocating the use of Gatetrade as the technology standard for all public procurement. It does however appear that a section of the Danish government is aligned differently to a decentralized ideology than the centralized that Gatetrade represents. Gatetrade just as other solution providers has been marketing its product and services. E-procurement is successfully diffused only when both organizational level decision makers and end users within the organization are convinced to adopt. Conjecture four: Governmental associations can via active mobilizations efforts have a positive influence in the diffusion of e-procurement. It should however be noted that eprocurement is successfully diffused only when both organizational level decision makers and end users are convinced to adopt. Standard setting: During the early stages of the diffusion of an innovation several standards compete to become the de facto. Those adopting an innovation regard the lack of a de facto standard a barrier towards adoption. In some instances such as in the case of operating systems, Microsoft being widely used has emerged as a de facto standard. There are other innovations such as the EDI in where government or a non-profit organization had to intervene to control promulgation in the number of standards. The timing of intervention is important for an innovation when regulated early can stall technological progress. A delayed intervention on the other hand can be ineffective as investments made in adopting a standard can cause organizations to resist regulation efforts (Damsgaard and Truex 2000; Kindleberger 1983). When several e-procurement standards exist, suppliers may have to prepare their catalogues to suit the different standards. Small and medium sized enterprises would find providing catalogues in several standards resource demanding and hence lack competitiveness as against large enterprises. SME’s can be quite critical when the public
9
Diffusion of Electronic Public Procurement in Denmark - Exam Copy sector demands catalogues in several standards. Public sector by adopting several eprocurement standards runs the risk of reinventing the wheel. Public sector however is not just efficiency driven but also ideology driven. Local governmental organizations can resist the adoption of a centralized e-procurement solution attempting to maintain their autonomy. The ministry of science, technology and innovation has selected Gatetrade as the standard and recommends that all public sector organizations use Gatetrade. A part of the public sector however questions if government via selecting Gatetrade discourages the development of innovative e-procurement solutions. Kubus is the case under discussion in this regard in the Danish context. If all governmental organizations had adopted Gatetrade, then Kubus would not have obtained the support required for its existence. Kubus has proven its innovative capability by winning e-handelprisen for the 2002 and 2003 years. Conjecture five: A solution selected as the standard should be provided as a recommended choice and not imposed upon. Innovation directive: Innovation directive is a command provided by the government to the supply side and the demand side for investing in the development or use of an innovation. The command can take several forms such as i) mandating the use of technology in a governmental agency or an industry ii) requiring that organizations use a certain percentage of their income in research and development activities and iii) directing organizations to use a product or a process. The ministry of finance in an effort to encourage the use of e-procurement has deducted a sum of money from budget allocated to some governmental organizations equivalent to savings expected from the use. It however is unclear as to how savings expected from eprocurement is calculated. Organizations have found through experience that savings that management and IT consultants foresee from the use of technologies is often hyped. Tonkin (2001) finds that organizations have in several instances first decided upon their e-government strategy and then rationalized their investment decision. For the German Bund Online project the rationale for calculating savings expected from e-government has not been declared. The state of Victoria in Australia has revised savings expected from e-procurement post implementation from 12 to 11 million dollars per annum (Tonkin 2001). It is to be recognized that the savings figure is often an anticipated one and our understanding of costs and benefits associated with the diffusion of an innovation is evolving is nature. The budget cut made the ministry of finance can result in a backlash when organizations post-implementation do not save as much as expected of them. Conjecture six: Innovation directives such as that of reducing budget allocations to governmental organizations can backfire when organizations post implementation realize that savings obtained do not match the budget cuts.
10
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
6. Conclusion This paper analyses actions taken by the Danish government for encouraging the diffusion of e-procurement in the public sector. The dimensions of institutional intervention defined by King el al. (1994) are used as a framework for performing the analysis. The data for the study was collected via a single embedded case study design. A part of the Danish government, ministry of science, technology and innovation, has played an active role in encouraging the diffusion of e-procurement. The ministry partnered with a few other governmental institutions and suggested that the Danish public sector trade via an e-market developed and hosted by a consortia of private companies; Gatetrade. The strategy to outsource the development and hosting of e-procurement has had success in that a part of the public sector – ministries - has joined Gatetrade and is increasingly utilizing the infrastructure. However, the local and regional bodies, which account for more than half the nation’s expenditure have not used Gatetrade actively. The analysis shows that as public administration is politically managed, government is mainly able to influence and not regulate both the supply and the demand sides. A regulatory action can be misinterpreted as an effort to alter power structures within the public administration.
7. Literature list Ageshin, E.A. (2001) "E-procurement at Work: A Case Study," Production and Inventory Management Journal Vol. 42, 1. Anderson, K.V., Juul, N.C., and Pedersen, J.K. (2003) "Fractional Institutional Endeavors and e_procurement in Local Government," 16th Bled Electronic Commerce Conference, Bled, Slovenia. Anonymous (2000) "The E-jump: A Discussion Paper on Electronic Procurement in Public Institutions," KPMG Consulting Anonymous (2001) "Strategy for the Implementation of E-Procurement in the Irish Public Sector", http://www.finance.gov.ie/documents/publications/other/ExecutiveSummary.pdf; last accessed on Feb 15 2004. Anonymous (2003a) "60 million kr til e-handel plads", Fyens Stifstidende. Anonymous (2003b) "Elektronisk Indkøb Portal giver overskud i 2005", Erhvervs bladet. Anonymous (2003c) "e-Procurement/Denmark: Successful public/private procurement initiative in Denmark," http://europa.eu.int/ISPO/ida/jsps/index.jsp?fuseAction=showDocument&documentID=693&pare nt=chapter&preChapterID=0-140-196-197 last accessed on May 15th 2004. Anonymous (2003d) "Modernizing Public Procurement", http://www.isc.ie/downloads/info.pdf; last accessed on March 23 2004. Anonymous (2003e) "Odder rykker front på nettet", Computer world. Anonymous (2004a) "The Danish eGovernment Strategy 2004 - 2006", http://www.e.gov.dk/sitemod/upload/Root/projekter/strategy_2004_06_en1.pdf; last accessed on May 15 2004. Anonymous (2004b) "E-handel på Gatetrade firedoblet på et år", Erhvervs Bladet. Baker, J. (1999) ePurchasing: Big Benefits, Tough Choices in: P2P Newsletter, 1999. Calway, R. (2003) "Galway City Council Report", http://www.isc.ie/downloads/info.pdf; last accessed on Feb. 12 2004. Coulthard, D., and Castleman, T. (2001) "Electronic Procurement in Government: More Complicated Than Just Good Business," European Conference on Information Systems. Court, P. (2003) Development of E-Procurement in the Public Sector - Making it Happen, CICA Annual Convention, 2003.
11
Diffusion of Electronic Public Procurement in Denmark - Exam Copy Damsgaard, J., and Lyytinen, K. (2001) "The Role of Intermediating Institutions in the Diffusion of Electronic Data Interchange (EDI): How Industry Associations Intervened in Denmark, Finland, and Hong Kong," The Information Society Vol. 17, pp 195-210. Damsgaard, J., and Truex, D. (2000) "Binary trading relations and the limits of EDI standards: The Procrustean bed of standards," European Journal of Information Systems Vol. 9, pp 173-188. Denmark.dk (xxxx) "The Danish Ministry of Foreign Affairs", http://www.denmark.dk/servlet/page?_pageid=85&_dad=portal30&_schema=PORTAL30&_fsitei d=175&_fid=12424w; last accessed on May 7th 2004. DOIP.dk (xxxx) "The Public Procurement Portal", http://www.doip.dk/default.asp; last accessed on April 12 2004. Eisenhardt, K.M. (1989) "Building Theories from Case Study Research," Academy of Management Review Vol. 14, pp 532-550. Emarketservices.com (xxxx) "E-market Database", http://www.emarketservices.com/; last accessed on March 12th 2004. Eprocurementscotland.com http://www.eprocurementscotland.com/; last accessed on April 23 2004. E-vergabe.de (xxxx) "The German Procurement Initiative", http://www.evergabe-online.de/; last accessed on April 12 2004. Folketinget.dk (xxxx) "Danish Folketinget - Danish parliment", http://www.folketinget.dk/; last accessed on April 2nd 2004. Gatetrade.com (xxxx) "A Danish e-market", http://www.gatetrade.com; last accessed on April 7 2004. GEBIZ.sg (xxxx) "Singaporean e-procurement initiative", http://www.gebiz.gov.sg/; last accessed on March 12 2004. Granovetter, M. (1978) "Threshold Models of Collective Behavior," American Journal of Sociology Vol. 83 (May), pp 1420-1443. Heneriksen, H.Z., Mahnke, V., and Hansen, J.M. (2004) "Public eProcurement Adoption: Economic and Political Rationality," Proceedings of the 37th Hawaii International Conference on System Sciences. Hp.dk (xxxx) "e-government framework", http://www.hp.dk/offentlig/egf.pdf; last accessed on 15th April 2004. Kablenet.com (xxxx) "Government Computing News", http://www.kablenet.com/; last accessed on May 15 2004. Kiærbye, H. (xxxx) "Helle Kiærbye Explains the Success of Denmark's Public Procurement Portal", http://europa.eu.int/ISPO/ida/export/files/egovo/1410.pdf; last accessed on March 12 2004. Kindleberger, C.P. (1983) "Standards as Public, Collective and Private Goods," KYKLOS Vol. 36. King, J.L., Gurbaxani, V., Kraemer, K., McFarlan, W., Raman, and Yap (1994) "Institutional Factors in Information Technology Innovation," Information Systems Research Vol. 5, 2. Kl.dk (xxxx) "Kommune LandsForeningen (Local bodies association)", http://www.kl.dk/; last accessed on Feb 20 2004. KMD.dk (xxxx) "Kommune Data - A Danisg Software Development Firm", http://www.kmd.dk/; last accessed on Feb. 8 2004. Kraaijenbrink, J. (2002) "Centralization Revisited? Problems on Implementing Integrated Service Delivery in the Netherlands," EGOV 2002. Krysiak, M., Tucker, C., Spitzer, D., and Holland, K. (2003) "E-Procurement; State Government Learns from the Private Sector," in: Digital Government; Principles and Best Practices, A. Pavlichev (ed.), Idea Group Inc., Hershey PA, USA. Kubus.dk (xxxx) "A Danish Software Development Firm", http://www.kubus.dk/go.asp?LangRef=1; last accessed on March 22 2004. Kumar, K., Dissel, V.H.G., and Bielli, P. (1998) "The Merchant of Prato - Revisited: Toward a Third Rationality of Information Systems," MIS Quarterly Vol. 22, 2, June, pp 199-226.
12
Diffusion of Electronic Public Procurement in Denmark - Exam Copy Lamsdorff, J.G. (2002) "Corruption Perceptions Index", http://www.globalcorruptionreport.org/download/gcr2003/24_Data_and_research.pdf; last accessed on April 4 2004. Oliver, P., Marwell, G., and Teixeira, R. (1985) "A Theory of the Critical Mass. I. Interdependence, Group Heterogeneity and the Production of Collective Action," American Journal of Sociology Vol. 91, 3, Nov., pp 522-556. Peria, F. (2003) E-procurement can have a substantial and positive impact on your bottom line in: CIO, 2003. Ramanathan, S. (2004) "Diffusion of eProcurement in the Public Sector: Revisiting Centralization Versus Decentralization Debates as a Twist in the Tale," European Conference on Information Systems. Rogers, E.M. (1995) "Diffusion of Innovation", (4th ed.) The Free Press, New York. Sell, A., Patakorpi, E., Anckar, B., and Walden, P. (2004) "Outsourcing Public Sector Services to Electronic Commerce Players," Hawaii International Conference on System Sciences. Sintef.no (xxxx) "Nordic Mobile Telephone System", http://www.sintef.no/eway/default0.asp?pid=199&oid=0&trg=LeftPage_4143&MainPage_3296= 4143:0::0:0:0:3296;::0:0:0&LeftPage_4143=3004:25979::0:3550:1:3296;4143;::10:0:0&lang=e ng; last accessed on Jan 26 2004. SKI.dk (xxxx) "The National Procurement Agency", http://www.ski.dk/; last accessed on Feb. 17 2004. Speilman, D. (2002) "The Diffusion of Agricultural Technology and the Role of Multinational Enterprise: Issues for Developing Economies," Dissertation Proposal, Department of Economics, American University Taran (2003) "E-Government: What Is It, or What Could It Be?" http://www.knowprose.com/mtentries/000375.html; last accessed on l April 14 2004. Tonkin, C. (2001) "E-Procurement in the Public Sector; Story, Myth and Legend," The Policy Institute, Trinity College, Dublin USA-dk.org (xxxx) "American Danish Business Council", http://www.usadk.org/html/denmark.html; last accessed on Feb. 12 2004. Yin, R.K. (1994) "Case Study Research: Design and Methods", (2nd ed.) Sage Publications, Thousand Oaks, California. Zulfiqar, K.A., Pan, S.L., Lee, J.N., and Huang, J.C. (2001) "E-Government: An Exploratory Study of On-line Electronic Procurement Systems [Case Study]," The 9th European Conference on Information Systems, Bled, Slovenia.
13
Appendix 12.5 Research Publication Five – the BLED 2004 Operationalizing Critical Mass as the Dependent Variable for Researching the Diffusion of EMarketplaces – its Implications By Ramanathan Somasundaram
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
17th Bled eCommerce Conference eGlobal Bled, Slovenia, June 21 - 23, 2004
Operationalizing critical mass as the dependent variable for researching the diffusion of emarketplaces – its implications Ramanathan Somasundaram Department of Computer Science, Aalborg University, Denmark
[email protected] Abstract E-market in this paper is regarded an innovation. Traditionally, researchers studying the diffusion of information technology innovations regard “adoption” of technologies by individual and organizations the dependent variable. This paper breaking away from tradition quips what if we regard critical mass the dependent variable? Critical mass seems an appealing dependent variable for an e-market can attain critical mass only when adequate number of end users adopt and participate. However, critical mass can be regarded the dependent variable only when we are able to predictively measure critical mass. Potential adopters’ expectation regarding an e-market’s profitability is proposed a measure for critical mass. Critical mass is explained here a moving target. The exact point in time at which an e-market would attain critical mass cannot be specified just as the market equilibrium in reality cannot be specified. The dynamics of attaining critical mass and not critical mass per se should be of interest just as it is about market equilibrium. Can critical mass theory as applied in this paper be used for researching the emergence of open source communities? When then does an open source community attain critical mass? Key words; Inter Organizational Systems, IOS, diffusion, adoption, implementation, electronic markets, critical mass and critical mass theory.
1 Introduction The term “critical mass” is often casually used not just by the business community but also among academics. Try presenting “critical mass” notion to an academic audience, almost every one regardless of their scientific understanding of the topic will have a say on it. Such a request is NOT made for venting out one’s experience. The critical mass term just as “strategy”, “governance” and “business model” is widely used and those aware of it have a subjective understanding of what it implies. Critical mass originates from nuclear physics in where it denotes the minimum amount of nuclear material that must be present for a self sustaining nuclear fission reaction to occur (Oliver et al. 1985).
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Social science researchers have applied the critical mass notion for explaining the diffusion of innovations (Rogers 1995). The notion is recognized to have high relevance with regards to innovations that are networked in nature such as telephones, e-markets, electronic mail, Internet and groupware systems (Katzand Shapiro 1986; Shapiroand Varian 1998). In the e-market context, the term denotes adequate levels of participation that can be measured for instance by the number of participants, catalogues and transactions, attaining which is what they aim for. An e-market is assumed to continue existing once it is in being due to network externalities; all or nothing proposition by Markus (1987) explains the rationale.
2 Literature analysis Unlike in physics where the exact instance at which a nuclear reaction would take off is predicted, in the social sciences area the critical mass term has been used in the “we know when it has happened” sense (Artleand Averous 1975; Markus 1987). The critical mass and the effect of it have been analyzed in the information systems area mostly in the form of economic modeling work (Clemonsand Kleindorfer 1992; Wangand Seidmann 1995).
Economic modeling strand
Explaining the growth curve of a networked technology post occurrence
Field study that analyze the effect of critical mass in the adoption decision
Economic modeling informed by field study data
e.g. Clemons and Kleindorfer 1992, Wang and Seidman 1995 and Kauffman and Wang 2002
Gurbaxani 1990, Rai et al. 1998 and Dutta and Roy 2003
Bouchard 1995, Damsgaard and Lyytinen 1998 and Lou et al. 2000
Teng et al. 2002; the author work like to position his work in this strand.
Table 1: Critical mass as studied in the information systems area
In another strand, researchers have attempted explaining the diffusion of networked information technologies such as the Internet and the BITNET (Duttaand ROy 2003; Gurbaxani 1990; Rai et al. 1998). These analyses are based on data that represent the event as it has occurred. The most representative explanation is the one that is best able to overlap the growth curve. Just as Rogers’ (1995), their models assume population a static variable and inflection point is defined by a percentage of the population which varies from 16% as in the case of Rogers (1995) to 37% to 50% as in Rai et al. (1998) case. They follow up their explanatory modeling with a detailed analysis of factors underlying their respective network’s actual growth curve. A few field study based analysis have included critical mass as an independent variable that affects one’s decision to adopt a networked technology (Bouchard 1993; Damsgaardand Lyytinen 1998; Lou et al. 2000). Teng et al. (2002)’s work marks the emergence of a trend in where economic modeling based on field study data is being done. In the recent years, there has been an increase in the attention provided to researching the role of critical mass in the diffusion of networked information technologies; see for example the research by (Duttaand ROy 2003; 2 Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Kauffmanand Wang 2002; Lou et al. 2000; Teng et al. 2002). An overview of research strands that study critical mass in the information systems area is provided in table 2. The above mentioned papers complement well or even extend knowledge accumulated in the inter-organizational systems (IOS) adoption field. The adoption of IOS has been researched as early as 1966 (Kaufman). A large number of field studies have been done in the IOS adoption area, most of which have studied the EDI technology. Researchers via statistical studies have researched the strength of relationship between factors such as the complexity, relative advantage, power and trust and the adoption of IOS (Chwelos et al. 2001; Hartand Saunders 1997). A process approach to researching the diffusion of networked information technologies has gained prominence since the late 90’s (Damsgaardand Lyytinen 1998; Kurniaand Johnston 2000). Ramanathan and Rose (2003) have analyzed the nature of knowledge accumulation in the IOS adoption field.
3 Problem formulation The analysis in this paper is focused on contributing towards researching the diffusion of one type of IOS; e-market. E-market in this paper is regarded an innovation. Traditionally, researchers studying the diffusion of networked information technology innovations regard the “adoption” of technologies by individual and organizations the dependent variable (see details of review paper by (Prescottand Conger 1995)). This paper breaking away from tradition quips what if we regard critical mass the dependent variable? Critical mass seems an appealing dependent variable for an e-market can attain critical mass only when adequate number of end users adopt and participate. A key requirement for regarding critical mass the dependent variable is predictively defining and measuring critical mass with respect to the diffusion of an innovation. Most studies treat critical mass vaguely or identify an innovation as having attained critical mass. Markus (1987), while conceptually analyzing the diffusion of interactive media highlights the need for operationalizing and measuring critical mass. Bouchard (1993) and Mahlers and Rogers (1999) explain the influence that a critical mass of users of an interactive technology has on one’s decision to adopt the technology. Their focus however lies on the adoption decision than on operationalizing critical mass. Gurbaxani (1990) and Rai et al. (1998) retrospectively analyze the diffusion of BITNET and the Internet respectively. Both of them attempt defining a suitable logic for explaining the growth of their respective networks and thereby provide base for developing predictive models. In the process they seek information on the inflection point at which growth rate is at the maximum; critical mass instance. While such retrospective analyses are certainly insightful and even required, would providing a measure for critical mass not be even more helpful in identifying the inflection point before the event has happened. Research community with prior knowledge of the measure can normatively guide practitioners towards attaining critical mass; thus making their work relevant. This paper calls for researching the diffusion of e-markets regarding critical mass the 3 Diffusion of Electronic Public Procurement in Denmark - Exam Copy
dependent variable. It clarifies further the ambiguous notion of critical mass. Furthermore, a measure for critical mass is described in the e-market context. The implications of regarding critical mass the dependent variable for research are discussed. This paper while conceptual in nature has resulted from an embedded case study (Yin 1994) carried out for researching the diffusion of e-procurement in the Danish public sector. The Danish ministry of science, technology and innovation has recommended the use of a private owned e-market for public procurement. However, the governmental organizations have not participated adequately as it was expected. Several stakeholders – buyers, sellers, competitors and trade associations – were queried about their involvement in the diffusion of eprocurement in the Danish public sector. The query was informed mainly by IOS adoption literatures. After having embarked upon the study, the author learnt about critical mass theory that has its origins in Sociology (Oliver et al. 1985) and its application for researching the diffusion of interactive media (Markus 1987). The ideas presented here are informed by critical mass theory related literatures and from the qualitative study done for researching the diffusion of e-procurement in the Danish public sector.
4 Argumentation details The following is the line of argument adopted for writing the paper; e-market is first recognized as a subset of IOS. This is done for the IOS area having been researched for almost four decades has a lot to contribute for researching the diffusion of e-markets. Second, the inadequacies of regarding “adoption” – identified as the most commonly regarded dependent variable for researching the diffusion of IOS – are highlighted. Such is done based on anecdotal data and via a literature review. Third, Oliver et al. (1985)’s critical mass theory in where critical mass is regarded the concept researched and a few relevant notions are explained. The new nature of knowledge that can be generated from applying these theoretical notions for researching the diffusion of e-markets is highlighted. Fourth, the critical mass is defined as a variable that is perceived by potential adopters in a population. The potential adopter’s perception of an e-market’s profitability is defined as the measure for critical mass. The research implications of regarding critical mass the dependent variable are highlighted in the final section.
5 E-market as a subset of IOS An e-market is an IOS that facilitates buyers and sellers to exchange information about market prices and product offerings (Bakos 1991). EDI is an IOS that has been extensively researched during the late 80’s and the early 90’s. There are three parties (buyers, sellers and market host) involved in diffusing independently managed e-markets. There however are only two parties actively involved in the implementation of EDI. An e-market’s host is a thinking body with survival instincts. In research so far, the role of host in the implementation of e-markets is poorly recognized.
4 Diffusion of Electronic Public Procurement in Denmark - Exam Copy
An e-market developed by a single buyer or seller for the purposes of automating interactions with its counterparts is different from EDI in that the medium has changed. Network effects for such would be quite similar to that of EDI. An open e-market, be it horizontal or vertical that allows multiple buyers and sellers to interact among one another, would have much higher network effects when successful. There are different types of e-markets. Kaplan and Sawhney (2000) conceptualized four types using a 2x2 matrix whose dimensions include “how businesses buy” and “what businesses buy”. The understanding about e-markets since then has evolved. Researchers such as Segev et al. (1999), Lennstrand (2001), Mahadevan (2002), Choudhury et al. (1998) and Sawy (2001) have both deductively and inductively conceptualized the dimensions that define an emarket. The dimensions along which an e-market can be classified are so many i) relationship orientation ii) revenue sources iii) transaction focus iv) ownership bias and v) market orientation and each has its subdivisions – that there can potentially be hundreds of types of e-market. Of the several types that exist, this paper has high relevance to those e-markets (type = open) that enrolls multiple buyers and sellers. A single buyer (e-procurement) or seller owned (web shops) emarket can be regarded as an automation effort by a proprietary. The challenges faced by open and proprietary e-markets for attaining critical mass are different.
6 Inadequacies of regarding adoption the dependent variable Several buying and selling organizations joined/adopted Gatetrade (www.gatetrade.net) during 2001 when it was initialized. Gatetrade however did not experience a proportional surge in trading volumes as the adopters hardly participated. The adopter’s use of the e-market infrastructure is vital for Gatetrade for it gets most of its revenue by charging for the transactions carried over its infrastructure. However, not all e-markets earn their revenues by charging for transactions. For instance, Scanmarket (www.scanmarket.com) leases out the use of its infrastructure to customers for a standard fee. The adopter’s use of the infrastructure has no implications for Scanmarket in terms of revenue. Scanmarket through its business model thereby shifts the onus of using the e-market to adopters. The challenge for Scanmarket thus is to attract adequate number of adopters. An e-market is successfully diffused only when it is adopted and used. Hence, diffusion research should regard adoption as a phase in the implementation process and not as the objective in itself. Such line of thinking is gaining hold in the diffusion research in the recent years. DIGIT (Diffusion Interest Group in Information Technology) 2003 workshop for instance is titled “beyond acceptance: investigating post adoption phenomena”. A research strand is emerging that analyzes “assimilation gaps” that exist between the adoption and the use of an innovation (Fichmanand Kemerer 1999). IOS adoption research has mostly queried factors that affect an organization’s decision to adopt. A hot topic in the subject used to be “why do small and medium sized enterprises resist adopting EDI?” (e.g. (Iacovou et al. 1995)). The underlying assumption behind such an enquiry is that a change agent when aware of the factors that cause SME’s to resist adopting EDI can remedy the situation 5 Diffusion of Electronic Public Procurement in Denmark - Exam Copy
and thereby enhance diffusion. It is not just the organizational and the innovation characteristics that affect one’s decision to adopt but also a decision maker’s perception of how many relevant others have adopted (Bouchard 1993) and how many others would adopt (Mahlerand Rogers 1999). Damsgaard and Lyytinen (1998) explain that decision makers act strategically when they choose to adopt or not. Kurnia and Johnston (2000) explain that decision makers are not unidirectionally affected by organizational, inter-organizational and technology factors but they via their actions have an effect on these factors vice versa. By focusing on an individual or an organization’s adoption decision, IS research community looses opportunity to learn about the population (environment) to which adopter’s belong. For instance, if adopter population can be characterized with assumptions regarding risk, resource or interest distribution, then diffusion can be predicted as an aggregated outcome. Mahajan et al. (1990) quote aggregated diffusion modeling work done in the marketing area. Such line of work has not been done in the IOS area to the author’s knowledge. The implications of regarding critical mass the dependent variable are graphically explained below in figure one. Nature of technology
External environment
Capability of organization
Action
E-market /IOS host
Influences via marketing
Unit of analysis Perceived cost variables
Perceived benefit variables
Outcome A. A typical adoption model with the factor approach* Nature of technology
Action guided by decision rule “act when perceived benefit > perceived cost”
Capability of organization
Number of adopters External environment
Action
Interorganizational environment Unit of analysis
Outcome
Unit of analysis
B. The processual approach*
Population characteristics – risk, resource and interest distribution
Critical mass /outcome C. The critical mass approach
* A and B are obtained from Kurnia and Johnston (2000)
Figure 1 – The critical mass approach graphically depicted
7 Regarding critical mass the dependent variable Regarding critical mass the dependent variable is appealing for the term is often associated with self-sustenance. The term has high relevance to phenomena that generate network externalities (e.g. message groups, online chat, open source communities and e-markets). For example, it is much easier to find a playing partner in one of the active Yahoo chess groups than in a lesser active online chess group. One who wants to play chess online would prefer joining Yahoo chess groups to the lesser active ones; in the process increasing activity in Yahoo and thereby its lead over the others. Yahoo chess groups can thus said to have attained critical mass. E-market is similar to that of an online chess group in that a 6 Diffusion of Electronic Public Procurement in Denmark - Exam Copy
buyer would find an e-market with a large number of sellers attractive and for sellers the vice versa. It needs to be clarified here that the terms “active” and “large” are perceived by potential adopters. What is large for some one in theory can be small for another. The challenge for an online chess community or an emarket is to cross the point beyond which a significant percentage of potential adopter’s would perceive it to have attained critical mass. Oliver et al. (1985) propose a theory of critical mass in collective action. Their objective is to explain as to why a small percentage of population takes the risk of contributing towards a collective good while a large percentage adopts a wait and see approach. Oliver et al.’s (1985) research objective when operationalized in the e-market context would explain as to why an organization risks joining an emarket that is yet to attain critical mass. An underlying assumption in the emarket context is that joining an e-market involves asset specific investments, which can take several forms such as the installation of technical infrastructure, training and redesigning organizational and inter organizational trade procedures. Thus if an organization joins an e-market that fails to attain critical mass then the organization risks loosing its asset specific investments. A rational decision would thus be to wait until an e-market attains critical mass. But if every one waits then how would e-markets attain critical mass? In reality however, e-markets such as Elemica (www.elemica.com) have emerged into existence. Critical mass theory explains the conditions under which collective action would and would not emerge (Oliverand Marwell 2001). The three central claims of critical mass theory are as follows; i) the likelihood of a collective good coming into existence is higher when heterogeneity within a potential adopter population is higher. Heterogeneity is characterized by interests and resources. For instance one with high level of interests and resources is most likely to contribute ii) Decision makers act sequentially while deciding upon whether or not to adopt a collective good. The theory assumes decision makers fully informed of other’s action while deciding. iii) Collective action can be classified under two types based on their production function; accelerative and decelerating. Production function is a curve that explains the relationship between one’s contribution towards a collective good and the likelihood of the good coming into existence. Early contributions in an accelerative production function increase the likely hood of a good coming into existence very little. The curve however grows steeply once a large percent of potential adopters find the collective good likely to exist. The decelerating curve in contrast grows steeply at first and then tapers off gradually. Markus (1987) applies critical mass theory for explaining the diffusion of interactive media. She finds interactive media diffusing in an accelerative pattern. She assumes interdependence as reciprocal instead of sequential. Reciprocal interdependence implies that one’s decision to adopt influences those who are yet to adopt while at the same time one gets influenced by the perceived likelihood of the actions of those who are yet to adopt; watch while being watched. When critical mass is regarded the dependent variable, several research issues comes into notice which when addressed help in better explaining the diffusion phenomenon. A conceptual explanation of how such is possible is hereby provided; Granovetter (1978) explains in the rioting context that the threshold to 7 Diffusion of Electronic Public Procurement in Denmark - Exam Copy
act for potential adopter population is distributed. One decides to act/adopt when he or she perceives the benefit of participation exceeding the cost of participation. In tandem with Oliver et al.’s (1985) conceptualization and Rai et al. (1998) findings, the potential adopter population is assumed heterogeneous. An e-market when having a transaction oriented business model would attain critical mass when adequate numbers of trading partners (both buyers and sellers) join/adopt and trade/participate/use. The population adopting an e-market and the population participating/using the e-market differs largely with a slight overlap. Adopters of an e-market usually are decision makers who are in the management end of organizational hierarchy. The users of an e-market on the other end are operation oriented employees like secretaries and procurement officers. Factors that affect one’s adoption of an e-market and one’s use of it would likely differ. Adopters and users meta-processing engine however remains the same; both act when they perceive the benefit of participation exceeding the cost of participation. Both the adopters and the users consider the cost-benefit equation of alternatives while considering action. A potential adopter evaluating joining a transaction oriented e-market would for instance look at the option of developing an eprocurement solution or explore other types of e-markets. A user would consider the traditional ways of working a viable alternative to trading via an e-market. Adopting an e-market does not automatically imply that it will be used. Potential participants would resist using an e-market when they find the adopted e-market unattractive (explained by cost-benefit equation). Adopter’s would tend to market their choice with vigor when stakes involved are high. For instance if an organization chooses to license the use of an e-market for a standard fee, it will actively market or even coerce usage so it can benefit from the investment. The extent to which an adopter is able to persuade or coerce usage is mediated by politics within an organization. Regarding critical mass the dependent variable in the manner described above denotes a paradigmatic shift in diffusion research due to the following; variables that affect one’s decision to adopt and use are processed through a decision maker’s cost-benefit processing engine. This is unlike the contemporary way of regarding independent variables directly affecting one’s decision to adopt. A key implication of the proposed shift is that research findings would have much higher relevance. Such line of thinking, though not explicitly recognized, is getting hold in researching the emergence of open source communities. There have been quite a few explanations on the costs and benefits of participating in an open source community (Karim and Bob 2003, Philippe 2003 and Andrea and Rossi 2003). This conceptualization explicitly acknowledges that decision makers perceive costs and benefits. An e-market can act opportunistically by altering potential adopter’s perception to its advantage through for instance marketing efforts. On the other hand, research efforts can inform potential adopters about evaluating an e-market. Oliver et al.’s (1985) critical mass theory, applied in the way described, is a suitable candidate for researching the diffusion of complex and networked technologies for it satisfies several assumptions held by researchers in studying the subject. Tornatzky and Klein (1982) recommend that the innovation diffusion research should predict, study both the adoption and the extent of use, use 8 Diffusion of Electronic Public Procurement in Denmark - Exam Copy
replicable and compare research approaches and query a decision maker before he adopts an innovation. Critical mass theory inherently is predictive in that it explains the conditions under which critical mass in collective action is attained. The author has explained earlier in this section as to how an individual’s adoption and his or hers extent of use are required for an e-market to attain critical mass. Factors that affect one’s perceived costs and benefits of an innovation can be acquired using replicable and comparable research approach while one considers adopting an innovation. Critical mass theory is truly a multi-level theory for it explains the relationship between an individual’s adoption decision and the emergence of a community. The theory assumes adopter’s deciding on an innovation in terms of time either sequentially or as Markus (1987) explains reciprocally. Critical mass theory thus has the potential to satisfy process theorists’ ((Kurniaand Johnston 2000; Lyytinenand Damsgaard 2001) call for researching the diffusion of complex and networked technologies over a period of time. The theory addresses Lyytinen and Damsgaard’s (2001) claim that one acts strategically while deciding to adopt and that complex technologies are learning intensive by explaining one’s decision to adopt in terms of his or hers perceived costs and benefits. Critical mass theory assumes potential adopter population distributed in terms of interests and resources. For applying the theory one has to explain resources and interests via contextual measures. It is through measures one can analyze the distribution of interests and resources. Critical mass theory regards the diffusion of other innovations, upon which the diffusion of an innovation is contingent (Mahajan et al. 1990), a resource. A resourceful person who is interested is likely to participate in an innovation for he or she perceives the benefits higher than the costs of participation. However, one would opt out from participating in an innovation when he or she perceives the costs of participation higher than the benefits after having adopted. Several suppliers joined Gatetrade when it started expecting to benefit from selling a lot via the medium. However, when they did not sell as much as they had expected they perceived the costs higher than the benefits and hence they dropped out. This explains that innovation need not necessarily traverse through distinct stages as Rogers (1995) assumes instead it can flow back and forth with feedback mechanisms as Lyytinen and Damsgaard (2001) explains. Critical mass theory as explained above is an aggregate diffusion model (Mahajan et al. 1990) that explains the diffusion of an innovation as a result of several individual/entities decision to adopt and participate.
8 Operationalizing critical mass in the e-market context Critical mass can be regarded the dependent variable only when we are able to predictively measure critical mass. Critical mass in nuclear physics is the amount of radioactive material required for a self-sustaining nuclear explosion. The term is vaguely measured in social sciences. The usual measure adopted is “we know when it has happened”. A measure is provided in this section with which one can predict as to when a three-party B2B e-market attains critical mass.
9 Diffusion of Electronic Public Procurement in Denmark - Exam Copy
An innovation is said to have attained critical mass when it is widely used. The validity of such an association is much higher in the case of networked technologies (e.g. e-market) for one’s benefit from participation depends on who else is part of the network and to what extent the network is used (Rohlfs 1974). A network host would collect service charges for having developed the network and for maintenance from the participants for their use. This paper assumes linearity between participation and service charges collected by the host. The measure for participation depends on an e-market’s business model. Participation for instance is measured by the number of transactions when an e-market collects revenues for the number of documents exchanged over its network. When an e-market charges a standard license fee for the use of its infrastructure then the participation is measured by the number of participants. An e-market, regardless of the business model, would continue to exist/would have self-sustained when it breaks even. The hosting of a B2B e-market is a costly activity. Hence, it is reasonable to assume that an e-market requires adequate participation for breaking even. An organization needs to make asset specific investments while joining an emarket. It invests not just in technology such as for integrating its internal financial systems with the e-market but also adapt its organizational and more importantly its inter-organizational procedures. It is vital for an organization that the e-market in which it has joined self sustains for if otherwise it risks loosing its investments. The risk of investing in an e-market is higher during the early stages than in the later stages of its inception. Due to the risk factor, organizations joining an e-market would find investing during the early stages costlier than investing in the later stages. Moreover, an organization inducing (persuade or coerce) its trading partners to adopt a specific e-market is as well an asset specific investment. An organization invests lesser for inducing its trading partners when a higher percentage of the population has adopted an e-market. Sociologists make simplistic assumptions regarding threshold and critical mass. Oliver et al. (1985) explain critical mass theory through a scenario in where a population of 1000 requires collecting 100,000 to hire a lawyer for preventing the closure of a school in their neighborhood. In the example, the number representing critical mass is 100,000. Everyone while deciding whether or not to contribute towards the figure has perfect information regarding how big the kitty has grown. In reality, there is no such static number defining critical mass in the diffusion of innovations. Critical mass instead is a perceptual variable. Members of a population perceive costs and benefits of participating in an e-market differently. For example, several organizations that joined Gatetrade during the early stages of its existence did so for they expected trading over it would be the norm. While regarding their action strategic they expected higher benefits due to first mover advantages. Potential adopters’ expectation regarding an e-market’s profitability is proposed as a measure for critical mass. Expectation here denotes a potential adopter’s beliefs on an e-market’s future. His perception about when an e-market would attain profitability accurately reflects his beliefs. One’s decision to join an emarket is guided by his or her beliefs. If one expects an e-market to do very well, then he or she would want to benefit by being a part of it. If one is unsure about joining an e-market, then he or she would adopt a wait and see approach. The 10 Diffusion of Electronic Public Procurement in Denmark - Exam Copy
duration until which one waits depends on the extent of uncertainty that he or she faces. It is the distribution of such beliefs among potential adopter population and the pattern in which individual adopter’s decision is aggregated that determines whether or not an e-market would attain critical mass. As per the proposed construct, an e-market need not break even for it to attain critical mass. Instead, it is adequate when an e-market is able to attract large enough percentage (number defined by an e-market’s business model e.g. whose transaction when carried out is adequate for an e-market to break even) of a population that regards trading over the e-market as way forward. The change agent should channel its marketing efforts in attracting the participation of those who have high regards about the emarket. As the percentage of population that has adopted grows, then the threshold of the remaining population decreases. Granovetter (1978) and Oliver et al. (1985) recommend such a form of aggregation. An e-market attaining critical mass is a process that is emerging. E-market, a thinking actor with survival instincts, is expected to gain efficiency through experience. Mass media and the word of mouth; variables from Bass model (1990) play a critical role in defining potential adopter’s expectation. These dynamically changing perceptions measured via potential adopters’ perception on an e-market’s profitability can be used for predicting as to when an e-market would attain critical mass. Those participating in an e-market would want an emarket to attain critical mass to the extent they have invested in an e-market. They would via persuasion or coercion alter a potential adopter’s expectation regarding an e-market. It is quite possible for one after having invested in an e-market to develop low expectations regarding an e-markets future resulting in him or her opting out. Such however is less likely to happen when an e-market has attained critical ma(Granovetter 1978). Critical mass as explained here is a moving target. The exact point in time at which an e-market would attain critical mass cannot be specified just as the market equilibrium in reality cannot be specified. However, measuring critical mass via one’s expectation about an e-market’s profitability allows for learning about factors that affect the process of attaining critical mass. The change agent when being aware of the factors can influence the process just as the potential adopters when being aware of it can evaluate an e-market. Critical mass just as market equilibrium itself should not be of interest instead it is the dynamics of attaining that should be.
9 Research implications An individual/organization’s decision to adopt is the unit of analysis while regarding adoption the dependent variable. When critical mass is regarded the dependent variable the system and the emergence of it is brought into focus. A few prominent implications for research community caused by the shift in focus are hereby explained. First, measures are required for explaining threshold distribution in a population. Second, e-markets from different parts of the world are increasingly getting interconnected. Because to this a Danish buying organization is now able to log on to a Scandinavian e-market and find a supplier from Japan who is registered with a 11 Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Japanese e-market without even having to shift graphical user interface. Is the Japanese firm part of the Scandinavian e-market’s adopter population? How is one to define population and the effect of its growth in the diffusion of e-markets? Marketing literatures, especially the ones that discuss telecom marketing, could be helpful. Third, risk and the word of mouth are explained as assumptions explaining aggregation dynamics in the diffusion of e-markets. Could there be more such assumptions? Fourth, can critical mass theory as applied in this paper be used for researching the emergence of open source communities? When then does an open source community attain critical mass? Literature List Artle, R., and Averous, C., (1975): The Telephone System as a Public Good: Static and Dynamic Aspects, Bell Journal of Economics Vol. 4, 1, pp 89-100. Bakos, J.Y., (1991): A Strategic Analysis of Electronic Marketplaces, MIS Quarterly Vol. 15, 3, pp 295-310. Bouchard, L., (1993): Decision Criteria in the Adoption of EDI, "International Conference in Information Systems," Orlando, Florida, pp. 365-376. Choudhury, V., Hartzel, K.S., and Konsynski, B.R., (1998): Uses and Consequences of Electronic Markets: An Empirical Investigation in the Aircraft Parts Industry, MIS Quarterly Vol. December. Chwelos, P., Benbasat, I., and Dexter, A.S., (2001): Empirical Test of an EDI Adoption Model, Information Systems Research Vol. 12, 3, pp 304-321. Clemons, E.K., and Kleindorfer, P.R., (1992): An Economic Analysis of Interorganizational Information Technology, Decision Support Systems Vol. 8, 5, pp 431-446. Damsgaard, J., and Lyytinen, K., (1998): Contours of diffusion of electronic data interchange in Finland; Overcoming technological barriers and collaborating to make it happen, Journal of Stategic Information Systems Vol. 7, 4, pp 275-297. Dutta, A., and ROy, R., (2003): Anticipating Internet Diffusion, Communications of the ACM Vol. 46, 2. Fichman, R.G., and Kemerer, C.F., (1999): The Ilusory Diffusion of Innovation: An Examination of Assimilation Gaps, Information Systems Research Vol. 10, 3, pp 255-275. Granovetter, M., (1978): Threshold Models of Collective Behavior, American Journal of Sociology Vol. 83 (May), pp 1420-1443. Gurbaxani, V., (1990): Diffusion in Computing Networks: The Case of BITNET, Communication of the ACM Vol. 33, 12, pp 65-75. Hart, P., and Saunders, C., (1997): Power and trust: Critical Factors in the Adoption and Use of Electronic Data Interchange, Organization Science Vol. 8, 1. Iacovou, C.L., Benbasat, I., and Dexter, A.S., (1995): Electronic Data Interchange and Small Organizations: Adoption and Impact if Technology, MIS Quarterly Vol. pp 465-485. Kaplan, S., and Sawhney, M., (2000): E-Hubs: The New B2B Marketplaces, Harvard Business Review Vol. Katz, M.L., and Shapiro, C., (1986): Technology Adoption in the Presence of Network Externalities, The Journal of Political Economy Vol. 94, 4, pp 822-841. Kauffman, R.J., and Wang, Y.-M., (2002): The Network Externalities Hypothesis and Competitive Network Growth, Journal of Organizational Computing and Electronic Commerce Vol. 12, 1, pp 59-83. 12 Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Kaufman, F., (1966): Data Systems that Cross Company Boundaries, Harvard Business Review Vol. pp 141-155. Kurnia, S., and Johnston, R.B., (2000): The Need for a Processual View of InterOrganizational Systems Adoption, Journal of Strategic Information Systems Vol. 9, pp 295-319. Lennstrand, B., Frey, M., and Johansen, M., (2001): Analyzing B2B eMarkets the Impact of Product and Industry Characteristics on Value Creation and Business Strategies, "ITS Asia-Indian Ocean Regional Conference." Lou, H., Luo, W., and Strong, D., (2000): Perceived Critical Mass Effect on Groupware Acceptance, European Journal of Information Systems Vol. 9, pp 91103. Lyytinen, K., and Damsgaard, J., (2001): What's Wrong with the Diffusion of Innovation Theory?, "Proceedings of the IFIP TC8 Working Group 8.6 Conference, Diffusing Software Product and Process Innovations," Banff, Canada, pp. 173-190. Mahadevan, B., (2002): Emerging Market Mechanisms in Business-to-Business E-Commerce: A Framework, "International Conference on e-Business, eEducation, e-Science, and e-Medicine on the Internet (SSGRR 2002's)," Rome, Italy. Mahajan, V., Muller, E., and Bass, F.M., (1990): New Product Diffusion Models in Marketing; A Review and Directions for Research, Journal of Marketing Vol. 54, pp 1-26. Mahler, A., and Rogers, E.M., (1999): The Diffusion of Interactive Communication Innovations and the Critical Mass: The Adoption of Telecommunications Services by German Banks, Telecommunications Policy. Vol. 23, pp 719-740. Markus, L.M., (1987): Toward a "Critical Mass" Theory of Interactive Media, Communication Research Vol. 14, 5, pp 491-511. Oliver, P., and Marwell, G., (2001): Whatever Happened to Critical Mass Theory? A Retrospective and Assesment, Sociological Theory Vol. 19, 3, pp 293-311. Oliver, P., Marwell, G., and Teixeira, R., (1985): A Theory of the Critical Mass. I. Interdependence, Group Heterogeneity and the Production of Collective Action, American Journal of Sociology Vol. 91, 3, pp 522-556. Prescott, M.B., and Conger, S.A., (1995): Information Technology Innovations: A Classification by IT Locus of Impact and Research Approach, DATA BASE Advances Vol. 26, 2 & 3. Rai, A., Ravichandran, T., and Samaddar, S., (1998): How to Anticipate the Internet's Global Diffusion, Communication of the ACM Vol. 41, 10. Ramanathan, S., and Rose, J., (2003): Rationalizing, Probing, Understanding: the Evolution of the Inter-Organizational Systems Adoption Field, "Hawaii International Conference on Systems Sciences," Hawaii. Rogers, E.M. "Diffusion of Innovation," (4th ed.) The Free Press, New York, 1995. Rohlfs, J., (1974): A Theory of Interdependent Demand for a Communications Service, The Bell Journal of Economics and Management Science Vol. 5, 1, pp 16 - 37. Sawy, O.E., (2001): "Identifying Structural Models of B2B Procurement Exchanges," Marshall School of Business. Segev, A., Gebauer, J., and Farber, F., (1999): Internet-Based Electronic Markets, International Journal of Electronic Markets Vol. 9, 3. 13 Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Shapiro, C., and Varian, H. "Information Rules: A Strategic Guide to the Network Economy Harvard Business School Press, 1998. Teng, T.C.J., Grover, V., and Guttler, W., (2002): Information Technology Innovations: General Diffusion Patterns and Its Relationships to Innovation Characteristics, IEEE Transactions on Engineering Management Vol. 49, 1. Tornatzky, L.G., and Klein, K.J., (1982): Innovation Characteristics and Innovation Adoption-Implementation: A Meta Analysis of Findings, IEEE Transactions on Engineering Management Vol. EM-29, 1. Wang, E.T.G., and Seidmann, A., (1995): Electronic Data Interchange: Competitive Externalities and Strategic Implementation Policies, Management Science Vol. 41, 3, pp 401-418. Yin, R.K. "Case Study Research: Design and Methods," (2nd ed.) Sage Publications, Thousand Oaks, California, 1994.
14 Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Appendix 12.6 IFIP 9.4 Newsletter Diffusion of E-Procurement in the Public Sector – What can we Learn from the Danish Experience? By Ramanathan Somasundaram
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
employed in the public sector as the work force in Denmark due to ageing population would significantly be less in a few years time. There is a high cost involved in printing these documents and in transporting them both within and across an organization. When e-procurement is done, efficiency is obtained not just via reduced printing and transportation costs but also via reduced process cycle time that results due to real time transmission. Maverick buying characterizes most if not all governmental organizations. In other words, the public sector lacks data on what they buy, from whom they buy and for how much they buy. Government organizations via e-procurement can obtain high quality data on their purchasing activities. Equipped with such data, government is well positioned to negotiate better frame agreements with suppliers. It is thus via reduced purchasing costs public sector efficiency can be enhanced. Corruption in the Danish public sector is almost non existent. In the global corruption report, Denmark is ranked the second least corrupt country. Corruption however is prevalent in developing and under developed countries, especially in the public sector. Development assistance provided by organizations such as the World Bank often does not achieve the intended purpose due to corruption. E-procurement via enhancing transparency and accountability in the purchasing process is expected to reduce corruption. While benefits obtained from reduced corruption are hard to quantify, it can be confidently said that their impact would be significant given the dire need for development assistance. E-procurement in the Danish public sector The public sector in Denmark is tri layered. The layers are i) Kommune (local authority) ii) Amt (regions) and iii) ministry (ministry). There are 275 local governmental organizations, 14 regions and 18 ministries. The local governmental organizations are autonomous to an extent for they collect revenues equivalent 2/3rd of their expenses via local taxes. Information about the extent of autonomy is important in that it in informal terms determines the authority that the central government has over the local governments. The Danish government as an entity procures goods and services for 100 billion DKK (app. 15 billion USD). A report by KPMG finds 8 billion worth of goods and services immediately suitable for e-procurement. The ministry of science, technology and innovation initiated e-procurement in the Danish public sector. A member involved in initiating the project reasons their decision as follows; “Basically it was three four years ago 1 we started seeing a marketplace (fails to recollect the name) but we looked at that and got the idea that may be we could make a marketplace for the public sector as well. And then we started; we presented it for the minister and we presented for the government and they agreed that we should try something like it. Because, it was the opinion that public sector could save some money by doing e-commerce and hardly any in the public sector had started using e-commerce. 1
The interview was conducted during March 2003.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
So, we thought that making the infrastructure ready with a portal, we could push the development of e-commerce… That is the whole idea. Because when we ask the ministries and the communities why they were not buying things electronically, they would say well it is too expensive, it is too difficult, what systems should we choose? The public sector cannot use this and a lot of excuses. If we make sure that they have the infrastructure and the right conditions at least they don’t have that excuse.” The ministry invited proposals for developing and hosting a procurement portal over which public sector companies can trade. Gatetrade, an e-market equally owned by four large Danish enterprises, was identified the winner on the 21st of March 2001. The ministry of science, technology and innovation suggested to governmental organizations that they use Gatetrade as the infrastructure to carry out their procurement. Gatetrade officially started its operations on the 1st of Jan. 2002. The infrastructure is open for use by both the public and the private sector. As of early 2004, the public sector provided 40% of its revenues. While Gatetrade has most certainly come a long way in terms of its product and service offerings and in attracting users, it is very much a loss making organization. Gatetrade’s owners have so far invested 160 million DKK (app. 27 million USD. Gatetrade posted a loss of 32 million DKK (app. 5 million USD) for the 2003 year as against 44.2 million DKK (8 million USD) for the 2002 year. Ministries in the Danish government and a few large governmental organizations such as Copenhagen University and Copenhagen Kommune have so far used Gatetrade. Kommune Data (KMD), an organization formed by the national association of local authorities and the association of county councils in 1972 to serve local governmental organization’s IT needs, is emerging as Gatetrade’s competitor. Most of the 275 local authorities have implemented KMD’s financial system. KMD is developing a web procurement module that seamlessly integrates with its financial system as chief product officer in KMD explains, “One of the key demands from the municipalities at that time and still today is; KMD you must develop a procurement system which is closely integrated to our financial system. Because the whole process is about ordering, about the invoice; all related back to our financial controlling system…This is our situation. We are building iprocurement system that is at a minimum (here meaning very) closely related to the financial. Because we know that is what the municipalities want and that is what they will pay for” Kubus is an innovative software development firm that has less than 10 employees. It provides contract management solutions to a few local authorities. SKI (state and local purchasing organization) negotiates frame agreements on behalf of the whole Danish government. It however has had only limited success in attracting buying organizations to use its frame agreements. An accounting manager of a large Danish organization has the following to say about SKI;
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
“Yes. That’s one of my enemies because the agreements are very poor. All my frame agreements is made with the same companies that SKI has; all my agreements is better… than they are by SKI, by the Danish government. Totally” SKI negotiated frame agreements can be accessed via Gatetrade. Purchasing activities are centralized differently within government organizations. Purchasing is a highly centralized activity in the Danish tax authority such that tax offices spread through out Denmark are obliged to use SKI negotiated frame agreements. In the other end of the spectrum, purchasing is a highly decentralized activity in Danish universities. Given the political nature of public administration, it is quite a challenge to centralize purchasing activities as the chief accounts officer in a Danish university explains; “The institution of X, the University of X is an institution the political Deans and the leaders of the institutes have very high power. It is very decentralized. It is not politically correct to have central departments taking care of things like that. I don’t think it is time for it now. But if I can make a number of successes to give the documentation that money is saved, perhaps it is better to easier to get that. I think we should do that. I don’t think we will be allowed to do that…” Key insights obtained from the study The following are the key insights obtained from researching the diffusion of eprocurement in the Danish public sector; Firstly, decision making power is taken away from organizations when procurement activity is centralized. Organizations use this power to administer their local economy via for instance incubating innovative companies such as Kubus and having a sense of control over the local economy. Small and medium sized companies would lack the infrastructure required to service large national contracts. Government via centralizing procurement would unintentionally support large enterprises over SMEs. In contrast, governments traditionally have supported SME’s. The public sector should be driven not only by efficiency but also by a larger array of variables. Secondly, the decision to select Gatetrade is made by the ministry of science, technology and innovation. The ministry had not actively engaged local authorities or KMD, the local authority IT expert while selecting Gatetrade. Had key members of the government been involved in preparing the national strategy for e-procurement, the implementation of the strategy would have been much easier. The ministry of science, technology and innovation is not a third party intermediary but is very much a part of the public administration. Its effort to implement e-procurement will be viewed in the light of ideological debates that have characterized the administration of public sector for centuries. Thirdly, outsourcing the development and hosting of e-procurement has its advantages. The Danish government has benefited from 160 million DKK investment that Gatetrade
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
owners have so far made in developing the infrastructure. As public procurement carried out over Gatetrade accounts for 40%, the Danish government has encouraged the diffusion of e-commerce in society in general. Finally, government organizations are able to get full benefits only when their financial system seamlessly integrates with e-procurement module. In the Danish government just as elsewhere several financials systems are used. Gatetrade is able to integrate well with financial systems installed in the Danish ministries. KMD is well positioned to provide a e-procurement module extension to financial systems installed in most local authorities. Technology is a critical variable in the diffusion of e-procurement in the public sector. A government should have a long term strategy for unifying islands of IT infrastructure that characterizes the public sector.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Appendix 12.7 Interview Data
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Interview data S. No. 1 2 3 4 5 6
Company name Aalborg Kommune IBX Aalborg Univ Aalborg Univ. Aarhus Købmand, Aarhus Told Skat
Type Buyer (professional) Technology provider (2 meet) Buyer (end user) Buyer (end user) Buyer (professional) Buyer (professional)
Date 25 Feb. 03 28th Feb. 03 4th Mar. 03 5th Mar. 03 6th Mar. 03 21st Mar. 03
7 8 9
Aarhus Univ Atea Ericsson
Buyer (professional) Seller - decision maker E-procurement coordinator
24th Mar. 03 27th Mar. 03 28th Mar. 03
10 11 12
Arla Zitrech Aalborg univ
E-procurement coordinator Seller – decision maker Buyer (professional)
31st Mar. 03 1st Apr. 03 8th Apr. 03
13
KBH univ
Buyer (professional)
10th Apr. 03
14 15 16 17 18
Aarhus bibliotek Biolab AH diagnostics Bie – Berntensen Fiskeri Ministry
Buyer (end user) Seller – decision maker Seller – decision maker Seller – decision maker Buyer (professional)
11th Apr. 03 14th Apr. 03 16th Apr. 03 22nd Apr. 03 22nd Apr. 03
19 20 21 22 23 24 25 26
Lundbeck SKI Science ministry Nycomed Københavns kommune Economy ministry Gentofte Kommune Herning kommune
E-procurement coordinator National procurement agency Policy maker E-procurement coordinator Buyer (professional) Buyer (professional) Buyer (professional) Buyer (professional)
7th May 03 8th May 03 14th May 03 15th May 03 27th May 03 28th May 03 28th May 03 4th June 03
27 28 29
Economy styrelsen Gatetrade Grundfos
Buyer (professional) Technology provider (2 meet) E-procurement coordinator
11th June 03 12th June 03 26th June 03
30 31 32 33
Lyreco Amtsråds foreningen KMD Kubus
Seller – decision maker Regional govt. association Technology provider Technology provider
27th June 03 2nd July 03 7th July 03 23rd July 03
th
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Appendix 12.8 Interview Questionnaire
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Interview Overview Researcher information Somasundaram Ramanathan alias Ram. SM. is a Danish government funded (grant no. 9900102) Ph.D. student at the department of computer science, Aalborg University. He completed his Master of Science degree in International Business Economics from the department of business studies, Aalborg University in June 2001. He can contacted at
[email protected] or at 9635 8894 (konto) or at 6170 2971 (mobil). Purpose To understand how government to business e-business communities emerge. Organizations that are both experienced and inexperienced in e-commerce are interviewed. Interview nature All of the questions would be open in nature (how, why etc.) The respondent can choose not to answer a particular question. There will not be any direct references to the interviewee. References while made to this interview will be abstracted to “a government organization in Denmark…” The talk will last approximately 60 minutes. Question guide 1) Information about Interviewee Position Experience 2) Organizational Information Organizational structure Purchasing (indkøb) budget Electronic commerce projects IT department 3) Purchasing Purchasing department information Types of goods purchased (cleaning materials, paper etc.) Purchasing process (e.g. centralization of the activity) 4) Internet Purchasing Extent of Internet purchasing and experiences with internet purchasing if any Marketplace participation, awareness and selection if any Future plans for purchasing over the Internet if any Awareness of other organizations buying and selling over the Internet if any Integration of internet systems with internal system (e.g. electronic invoice) if any 5) Comments that you would like to add.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Appendix 12.9 Critical Mass Game
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
The Critical Mass Game Objective To study the emergence of a networked innovation in class room context The context A bunch of potential adopters are enquired about their decision to join/adopt an e-market. Data sheets added to this document define the characteristics of the potential adopters. Through certain logic, benefit perceived by a potential adopter is differentiated. Some perceive the benefit of joining the e-market high while others perceive low (e.g. 12000$ and 3000$ for high and low respectively). Due to network externalities, a potential adopter is assumed to find the e-market more valuable as more join. The cost of joining the e-market for an adopter is assumed static (8000$). The e-market is assumed to breakeven when it earns 40,000 USD (i.e.) when it obtains five adopters. An e-market has to breakeven eventually for it to continue existing. The duration for which it can sustain depends on the losses it is able to able. Remember that some in the adopter population take more risks than the others (e.g.) one might decide to take losses in the beginning to gain long term benefits. How is the game played? Game preparation: Data sheets enclosed with this document define a population. You can define the characteristics of the population as you like (i.e.) a large number in the population could find joining the e-market beneficial and the vice versa. Or else you could introduce randomness by printing the seven data sheets many times and ask the students to choose blindly. Playing the game: Once the data sheet is distributed to potential adopters, ask them one by one (e.g. from right to left) to decide on whether or not to join the e-market. In the process make sure to inform them about long term benefits (data found down the rows). Say three have joined, then the number of buyers in an e-market is assumed to be three. The next person deciding on joining the e-market takes into account the data in the appropriate row. When all have decided, you will know whether or not the e-market has attained break-even/critical mass. Post game: Make sure to reflect upon the dynamics underlying the emergence of the emarket.
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Data sheet one Perceived benefit value starts at 3000 Standard remarks i) all values mentioned in the table can be regarded as in dollar nominations ii) the values mentioned in the table are for a year No. of buyers in market X 0
Perceived benefit value 3000
Joining cost 8000
Net benefit - 5000
Cost for X 40000
Earnings for X 0
Risk of X failing to emerge in % 100
1
3600
8000
- 4400
40000
8000
80
2
4320
8000
- 3680
40000
16000
60
3
5184
8000
- 2816
40000
24000
40
4
6221
8000
- 1779
40000
32000
20
5
7465
8000
- 535
40000
40000
0
6
8958
8000
+ 958
40000
48000
None
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Data sheet two Perceived benefit value starts at 3500 Standard remarks i) all values mentioned in the table can be regarded as in dollar nominations ii) the values mentioned in the table are for a year
No. of buyers in market X 0
Perceived benefit value 3500
Joining cost 8000
Net benefit - 4500
Cost for X 40000
Earnings for X 0
Risk of X failing to emerge in % 100
1
4200
8000
- 3800
40000
8000
80
2
5040
8000
- 2960
40000
16000
60
3
6048
8000
- 1952
40000
24000
40
4
7258
8000
- 742
40000
32000
20
5
8709
8000
+ 709
40000
40000
0
6
10451
8000
+ 3451
40000
48000
None
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Data sheet three Perceived benefit value starts at 4000 Standard remarks i) all values mentioned in the table can be regarded as in dollar nominations ii) the values mentioned in the table are for a year
No. of buyers in market X 0
Perceived benefit value 4000
Joining cost 8000
Net benefit - 4000
Cost for X 40000
Earnings for X 0
Risk of X failing to emerge in % 100
1
4800
8000
- 3200
40000
8000
80
2
5760
8000
- 2240
40000
16000
60
3
6912
8000
- 1088
40000
24000
40
4
8294
8000
+ 294
40000
32000
20
5
9953
8000
+ 1953
40000
40000
0
6
11935
8000
+ 3935
40000
48000
None
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Data sheet four Perceived benefit value starts at 5000 Standard remarks i) all values mentioned in the table can be regarded as in dollar nominations ii) the values mentioned in the table are for a year
No. of buyers in market X 0
Perceived benefit value 5000
Joining cost 8000
Net benefit - 3000
Cost for X 40000
Earnings for X 0
Risk of X failing to emerge in % 100
1
6000
8000
- 2000
40000
8000
80
2
7200
8000
- 800
40000
16000
60
3
8640
8000
+ 640
40000
24000
40
4
10368
8000
+ 2368
40000
32000
20
5
12442
8000
+ 4442
40000
40000
0
6
14930
8000
+ 6930
40000
48000
None
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Data sheet five Perceived benefit value starts at 6000 Standard remarks i) all values mentioned in the table can be regarded as in dollar nominations ii) the values mentioned in the table are for a year
No. of buyers in market X 0
Perceived benefit value 6000
Joining cost 8000
Net benefit - 2000
Cost for X 40000
Earnings for X 0
Risk of X failing to emerge in % 100
1
7200
8000
- 800
40000
8000
80
2
8640
8000
+ 640
40000
16000
60
3
10368
8000
+ 2368
40000
24000
40
4
12442
8000
+ 4442
40000
32000
20
5
14930
8000
+ 6930
40000
40000
0
6
17916
8000
+ 9916
40000
48000
None
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Data sheet six Perceived benefit value starts at 8000 Standard remarks i) all values mentioned in the table can be regarded as in dollar nominations ii) the values mentioned in the table are for a year
No. of buyers in market X 0
Perceived benefit value 8000
Joining cost 8000
Net benefit 0
Cost for X 40000
Earnings for X 0
Risk of X failing to emerge in % 100
1
9600
8000
+ 1600
40000
8000
80
2
11520
8000
+ 3520
40000
16000
60
3
13824
8000
+ 5824
40000
24000
40
4
16589
8000
+ 8589
40000
32000
20
5
19907
8000
+ 11907
40000
40000
0
6
23888
8000
+ 15888
40000
48000
None
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Data sheet seven Perceived benefit value starts at 12000 Standard remarks i) all values mentioned in the table can be regarded as in dollar nominations ii) the values mentioned in the table are for a year
No. of buyers in market X 0
Perceived benefit value 12000
Joining cost 8000
Net benefit + 4000
Cost for X 40000
Earnings for X 0
Risk of X failing to emerge in % 100
1
14400
8000
+ 6400
40000
8000
80
2
17280
8000
+ 7280
40000
16000
60
3
20736
8000
+ 12736
40000
24000
40
4
24883
8000
+ 16883
40000
32000
20
5
29860
8000
+ 21860
40000
40000
0
6
35832
8000
+ 27832
40000
48000
None
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
Appendix 12.10 A Sample Transcribed Interview (only Provided to Examiners)
Diffusion of Electronic Public Procurement in Denmark - Exam Copy
