Public Private Partnerships in Information System projects
An analysis of the effects of distinguishing characteristics of PPP-DBFMO on
the success of IS projects
N.J.J. Martens 10667091
First Supervisor: R. Kleijn & G.P.A.J. Delen Second Supervisor:
January 30th, 2015
MSc Information Sciences: Business Information Systems University of Amsterdam, Faculty of Science
Preface
This thesis is the result of my interest in information and communication technology and its applications in the public sector, an interest that also becomes apparent from my masters educations. I found the subject of Public Private Partnerships in IT projects very interesting to investigate and proved to be quite the challenge. During my bachelor and master in public administration I had already engaged on the issue of manageability of IT projects in the public sector. This thesis provides, in my opinion, one solution to the manageability problem of these IT projects. Additionally, the research for this thesis – in particular – the interviews have provided me with interesting and intriguing discussions on the subject. What also struck me was – once an interview was arranged – how enthusiastic, passionate and benevolent respondents were to talk about their activities and practical experience. This attitude had a catching effect on me and has led several times to subsequent conversations on the subject.
While writing this thesis, I was supported by a number of people in different ways. First of all I would like to thank Ronald Kleijn who, as my first thesis supervisor, greatly helped in shaping this thesis through conversations, discussions and new insights. These talks have benefited and contributed to this finished final product. Furthermore, I would like to express my appreciation for the fact that Ronald helped me by contacting his network to be interviewed on the findings of my thesis. I would also like to thank my second supervisor, Guus Delen, for his input on my thesis and contacting his former colleagues to provide expert opinions on my thesis findings.
Furthermore, I would like to thank my parents for listening to my thesis writing troubles and my sisters have also greatly helped me with talking about how to write a master thesis and provide information on writing styles and forms. Lastly, I would like to thank my girlfriend and friends who have suffered my monologues, which often led me to new insights and making notes about my thesis at the oddest moments and times, and other peculiarities that I developed while writing this thesis. A special thanks also to my roommates who, as elusive as it sounds, often came by to make me evade working on my thesis.
TABLE OF CONTENTS TABLES
FIGURES
LIST OF ABBREVIATIONS
1. INTRODUCTION 1
1.1CONTEXT AND THESIS THEME 1
1.2RESEARCH QUESTION 2
1.3RELEVANCE 3
1.4METHODOLOGY 4
1.4.1INTERVIEW DESIGN 4
1.5THESIS STRUCTURE 5
2. ASSESSING THE SUCCESS OF INFORMATION SYSTEMS PROJECTS 6
2.1THE D&M MODEL OF INFORMATION SYSTEM SUCCESS 6
2.2THE UPDATED D&M MODEL OF INFORMATION SYSTEM SUCCESS 8
3. PUBLIC PRIVATE PARTNERSHIPS 10
3.1PUBLIC PRIVATE PARTNERSHIPS IN THE DBFMO FORM 10
3.1.1PPP-DBFMO AND (IT)OUTSOURCING 11
3.2PUBLIC PRIVATE PARTNERSHIP AND ICT PROJECTS 12
3.3MANAGING INFORMATION SYSTEM SUCCESS IN IT PROJECT SOURCING 14
3.3.1CHALLENGES TO PPP-DBFMO IN TERMS OF IS PROJECTS 14
4. THE EFFECT OF PPP-DBFMO ON THE SUCCESS OF IS PROJECTS 15
4.1D&M MODEL AND DETERMINING IS SUCCESS 15
4.2DISTINGUISHING CHARACTERISTICS OF PPP-DBFMO 15
4.3THE EFFECT OF PPP-DBFMO ON IS PROJECT SUCCESS 15
4.3.1QUALITY DIMENSIONS 16
4.3.2USAGE DIMENSIONS 18
4.3.3NET BENEFITS 20
4.4FINDINGS FROM INTERVIEWS 23
4.4.1INTERVIEWS ON THE EFFECT OF PPP-DBFMO ON IS PROJECT SUCCESS 23
4.4.2ADDITIONAL INSIGHTS FROM INTERVIEWS 24
5. CONCLUSION 27 5.1CONCLUSION 27 5.2DISCUSSION 29 5.3FURTHER RESEARCH 29 LITERATURE 31 APPENDICES 34
APPENDIX A–INTERVIEW DESIGN 34
Tables
Table 2.1 IS Success Variables (from Petter, DeLone and McLean, 2013: 11) ... 9
Table 4.1 Effect of distinguishing characteristics of PPP-DBFMO on usage dimension ... 18
Table 4.2 Effect of distinguishing characteristics of PPP-DBFMO on usage dimension ... 20
Table 4.3 Effect of distinguishing characteristics of PPP-DBFMO on usage dimension ... 22
Figures Figure 1.1 Visualisation of thesis’ research model ... 5
Figure 2.1 D&M IS success model (DeLone & McLean, 1992:87) ... 7
List of Abbreviations
PPP-DBFMO Public Private Partnership in the concessional form also called Design, Build, Finance, Maintain and Operate because of the contracting out of these aspects of the project.
D&M model The DeLone & McLean model for information system success, this model resulted from a meta-analyse study of IS evaluation discourse over more than 20 years.
IS project Information System project.
IT project Information Technology project.
ITO Information Technology Outsourcing.
QD Reference to the quality dimensions of the D&M model for IS success, these are the desirable characteristics of the system, information and service quality.
UD Reference to the usage dimensions of the D&M model for IS success: (intention to) use and user satisfaction. These are the degree, manner and satisfaction of staff and customer usage of IS capabilities. ND Reference to the net benefits dimension of the D&M model for IS
success. Also described as the extent to which an IS contributes to the success of individuals, groups or organizations.
SLA Service Level Agreements.
1. Introduction
Public private partnerships are nowadays seen as the policy instrument of choice in order to attain public goals better (Klijn & van Twist, 2007). Participants in this cooperation, government and private actors, are better able to concentrate on their qualities and thus creating synergy. This chapter will focus on the context and theme of this thesis, from which a main research question and sub-questions for the structured answering are derived. Then, the relevance and methodology with a conceptual approach to finding the answer to the main question will be discussed.
1.1 Context and Thesis Theme
In the past years, Public Private Partnerships (PPP) has shown themselves to be a useful and effective tool for the outsourcing of government tasks around the world. This form of procurement is based on a result commitment, unlike in the case of public tenders, where the content of a project is drawn up in detailed and comprehensive requirements. This form of cooperation is characterized as a long-term relationship between government (public) and business (private), in which both parties maintain their identity and responsibility. Despite an allocation of tasks and extensive leeway in implementing that is given to other parties outside government, the management role remains with government. This form of cooperation ensures that the realisation of the project occurs with a clear and optimal distribution of labour and risk, which will lead to an efficiency improvement. This improvement can for instance be lower costs or better quality. A frequently occurring form of a PPP is the Design, Build, Finance,
Maintain (and Operate) (DBFMO) contract. The before mentioned elements of a project are
transferred to the supplier, including the financing of the project. Payment to the supplier occurs periodically on the basis of set achievements. Central to this form of procurement is a strong focus on the output of the project and the long contractual period (15-30 years). Due to the integrated character, the longer time span of the project and the relocation of the financing obligation to the supplier capital gains are realized (Blayse & Manley, 2004; Akintoye et al., 2005; Leiringer, 2006).
At this moment different DBFMO projects are on-going in the terrains of infrastructure and the environment, internal affairs and safety and justice. According to the progress report by the Dutch government, DBFMO has already made capital gains on an average of 10%-15% compared to traditional outsourcing. Furthermore, this report shows that these projects are delivered in time. Until now most of the PPP projects were realized in the form of infrastructural projects, however the Dutch government is investigating a more structural application of this project form. Other areas should also benefit from the creativity and innovativeness of the private sector (Eversdijk & Korsten, 2009). Fife and Hosman (2007) focus on the prospects of PPPs for ICT projects, the study of two cases provides valuable insight on the success of future PPPs in the ICT sector. Another article by Sharma (2007) focuses on best practices in PPPs in e-government services with the use of (Asian) case studies. These cases provide the reader with an overview of and insight in best practices and key learning points. The Dutch Administrative Guideline Sourcing ‘Onderweg naar Sourcing’1 focuses on Public Private
Partnerships as well (DG Organisatie en Bedrijfsvoering Rijk, 2009: 21-22). Here it is stated that a PPP is a far-reaching business partnership between a government organisation and one or multiple (a consortium) private parties. The basis of this cooperation is equivalence between government and suppliers, in the form of shared goals, resources (investments) and people in order to perform services. According to the guideline other features are the focus on partnership, tailored work, joint investment and shared risks, and to be newly developed ICT services. This last feature is where this thesis wants
to contribute with the addition on information about the improvement of the quality that PPP-DBFMO procurement can generate. This guideline additionally focuses on the control of demand for and supply of ICT services (DG Organisatie en Bedrijfsvoering Rijk, 2009: 67-73). Here some levels of coordination are identified to minimize the risk of ineffective service delivery. Furthermore, an article on iBestuur.nl by Duyneveldt and Nijman (2013)2 – IBM advisors for governmental IT projects – on
the risks and chances of PPP-DBFMO in governmental IT projects. This article shows that the possibilities that public private partnerships offer will further our society collectively, because of the potential the market holds for governmental service delivery.
Eversdijk and Korsten (2009:14) note that the public service can be improved by the introduction of market mechanisms by PPPs, however this is not an easy task. Parties in the partnership all have their own structure, culture, goals, ideas and strategies concerning the cooperation. In this partnership the synergy, between the public and private sphere, is the important added value in DBFMO as a form of PPP (Klijn & Teisman, 2003:137). This allows for the development of a product that would not have been developed without the PPP.
The focus of this thesis will be on the influence of the PPP-DBFMO procurement on the quality of information systems or technology projects of (the Dutch) government. As was shortly introduced, Public Private Partnerships have shown to be a successful way of project works. Research into the use and applicability of DBFMO for governmental Information Systems is still in its early stages. There are only a few examples of Information Systems projects procured in this PPP project form. Academic research into the best practice of PPP-DBFMO in information systems/technology projects of government and their obstacles is also still sparse. Based on the positive experiences from other areas in government, it is an interesting field of research to explore. Therefore, this thesis will focus on the effects PPP-DBFMO has on the success of IS projects that are procured in this manner and thus contribute to this field of research.
1.2 Research Question
The research question central to this thesis focuses on the potential PPP-DBFMO can have for the success of information systems projects. To assess the impact on the success of information system/technology projects, the following research question is addressed in this thesis:
RQ: “How do the distinguishing characteristics of PPP-DBFMO have an effect on the success of information systems/technology projects in the Netherlands?”
In order to answer this main research question systematically, a number of related sub questions will be used. These questions provide a guideline that will help by providing meaningful substantiation for the answer to the main question. The first sub question focuses on the assessment of success of information systems, which will provide a basis for the research in this thesis.
SQ1: “How can the success of information systems/technology projects be determined?”
Here success assessment model for information system projects will be discussed to show the important dimensions and attributes of success. For this purpose the model by DeLone and McLean (1992; 2002; 2012) on the success of information system projects may prove to be useful.
http://ibestuur.nl/partner-Then, the thesis will turn to Public Private Partnerships and provide a clear insight in the distinguishing properties of this form of outsourcing. Therefore, the second sub question reads:
SQ2: “What are Public Private Partnerships and what are the distinguishing characteristics of this project form?”
The answer to this question will first address the PPP-DBFMO form of procurement; additionally it will focus on what is distinguishing to this project form.
The final sub question will focus on the combination of the earlier questions in order to provide insight on the effect of PPP on the success of an information system project. Therefore the last subquestion of this thesis will be:
SQ3: “How do the distinguishing characteristics of Public Private Partnerships affect success in an IS/IT project?”
This question allows us to make expectations about the impact on the quality of information systems/technology projects PPP-DBFMO can have. It is important to pay special attention to the distinguishing characteristics of this project form, since these are the characteristics that will change the impact PPP-DBFMO has on the quality of IS/IT projects. In addition to the literature research qualitative sources will be used to answer this subquestion.
Using these sub questions it will be possible to provide an answer to the main research question on the success implications of Public Private Partnerships (DBFMO) for IS/IT projects of the Dutch government.
1.3 Relevance
The introduction shortly focused on the DBFMO form of PPPs and turned to the assessment of information systems/technology projects. The body of literature on research on this topic is still limited and the subject is gathering more attention.
This subject is relevant to me personally due to the subject of Public Private Partnerships in the form of DBFMO by my prior education in Public Administration. Issues of legitimately outsourcing, the procurement of government services and improvements to the service delivery of government, are interesting research material. The qualitative improvement of IS/IT projects of government will be an interesting topic to research for this thesis. Furthermore, my current education program focuses on Business Information Systems, mostly associated with Information Systems and Information Communication Technology. From both educations the subject of the possibilities of procurement through DBFMO as a form of PPP for governmental information systems is an interesting research topic. When looking at scientific articles and research reports from government this development seems to be attracting attention.
This subject is socially relevant due to the central role of public organizations and the digital governmental information system services provided for society. Changes to the service provision by government through the DBFMO procurement as a form of PPP will have consequences in society due to the embeddedness of government. Changes to this service provision will impact several relationships, such as the ones between government and citizens and a new relationship between service provider (DBFMO provider), citizens and government. Furthermore, outsourcing a
information systems/technology projects through PPP-DBFMO will also impact the position of the staff that maintains the system. Where this would previously have been public officials or subcontracted government employees, in a PPP the working relationships and positions of staff will be different.
In scientific literature, the subject of the applicability of DBFMO as a form of PPP procurement for e-government services or ICT projects has not received little attention. There is more focus on DBFMO procured projects in other areas within government, such as infrastructural projects. However, many researches mention the benefits of this form of procurement (be it in other areas), which makes it an interesting subject to also explore in other parts of government. For example, Bovaird (2004: 202) identifies resource management such as ICT as a potential purpose for PPPs. Additionally, scientific literature focuses more on IT Outsourcing – because this form of procurement has been around longer. It is shown more often that the two methods of procurement show similarities. The proposed research question will contribute to the body of literature on the possibilities of DBFMO as a form of PPP for governmental ICT project. Furthermore, this thesis will also be scientifically relevant to the IS evaluation discourse from which the IS success model is derived. This thesis will provide insight in the application and usability of the model, regarding the explanatory and predictive function.
1.4 Methodology
The research method that will be used for this thesis is a literature study and analysis supplemented by interviews to validate the findings. The thesis will start by performing a literature study on a model to determine the success of an information system, as well as a study of public private partnerships. For these literature studies different sources of information will be used, of which the most important will be academic articles. Additionally, information will be retrieved from books, newspapers and the internet to provided an extensive overview of the subject being studied.
Following the literature the findings will be analysed and the synthesis will provide us with a conceptual model, based on the findings. This model will be validated by interviews with respondents that are experts from the field of IT project management and sourcing. These interviews may also be used to acquire additional information, for instance about the issues, problems and pitfalls of DBFMO for information systems/technology projects.
1.4.1 Interview design
This section will address the initial steps of the semi-structured interview design that will be used to increase the internal validity of this thesis. As Robson (2011: 285) and Babbie (2007: 327) indicate, a semi-structured interview can contribute to the validity of research. This type of qualitative field research also allows the researcher to gain complex and in-depth responses and insights to the research matter. According to Robson (2011:87) this allows for triangulation with findings from the literature review, thus contributing to the internal validity of the thesis.
It is also important that the validity of the interviews is guaranteed, meaning that the research topics in the interview correspond to the research topics that this thesis focuses on. The further detailing of the interviews and substantiated choices regarding the validity of the interviews will be provided in Appendix A.
Figure 1.1 Visualisation of thesis’ research model
1.5 Thesis Structure
The thesis is structured in accordance with the sub questions that have been introduced earlier. The following chapter will be concerned with providing a framework that allows for the assessment of Information Systems or Technology projects. This framework is necessary for the assessment of the distinguishing characteristics of public private partnerships. Chapter 3 will provide an overview of Public Private Partnerships, which will start with providing a general overview of the topic. Then academic research on the topic stating the characteristics and distinctive features of the project form will be discussed. This will be followed by focusing on scientific and practical information about the application of PPP in ICT projects. The fourth chapter provides a brief overview of the results from the literature reviews that were done. Then the effect of the distinguishing characteristics of PPP-DBFMO will be assessed using the model that was presented in the second chapter. The final part of this chapter provides an assessment based on interviews, on the presented findings. It will also included additional insights on this subject that emerged form the interviews.
2. Assessing the success of Information Systems projects
This chapter will focus on the assessment of the success of an information system or technology project, providing an answer to the question “How can the success of information systems/technology
projects be determined?”. The model of IS success by DeLone and McLean (D&M model) will be
discussed, as a tool that allows us to assess the success of information systems. First the original model will be presented, to subsequently turn to the updated model and later work from the authors on this subject.
2.1 The D&M model of Information System Success
DeLone and McLean (1992; 2003) have performed extensive reviews of literature on the assessment of information system success. The first version of the model by DeLone and McLean (1992) was a taxonomy of information systems success consisting of six major dimensions or categories: System quality, Information quality, Use, User satisfaction, Individual impact and Organizational impact. These categories were found in over 180 conceptual and empirical studies on the subject, found in seven different journals on Information Systems research (DeLone and McLean, 1992:63). It is a matter of course that there are discrepancies or overlap between articles and the main categories, however there was a 90% interrater agreement from independent article classification. From these studies the taxonomy was identified and a number of publications selected as the main idea for the category. The authors also provided listings of all used success measures in the articles.
Measures of System Quality were not surprisingly, fairly straightforward measures that focus on the systems’ engineering-oriented performance characteristics (DeLone and McLean, 1992:64). Srinivasan (1985) performed an empirical study on computer-based modelling among 29 organizations. The following measures to determine the success of information systems were found: response time, system reliability and systems accessibility. In a study on emergency management by Belardo, Karwan and Wallace (1982) reliability, response time, ease of use and ease of learning were used to measure success of IS. Bailey and Pearson (1983) used convenience of access, flexibility of system, integration of system and response time to measure the overall success of IS.
The second category of the information system success taxonomy that was identified is Information
Quality, focusing on the system’s quality of output information (DeLone and McLean, 1992:65). The
information production of a system has been measured with the use of different quality attributes. Bailey and Pearson (1983) used the accuracy, precision, currency, timeliness, reliability, completeness, conciseness, format and relevance of information as indicators for its quality. In other research King and Epstein (1983) have used similar indicators: currency, sufficiency, understandability, freedom from bias, timeliness, reliability, relevance to decision, comparability and quantitativeness. In other research focused on overall IS Miller and Doyle (1987) assessed information quality by focusing on completeness of information, relevance of reports and timeliness of reports. The Information Use category focuses on quality measures that determine IS success by looking at different methods of information usage. DeLone and McLean further note that this category is the most objective and easiest to quantify conceptually. Several studies (Baroudi, Olson and Ives, 1986; Culnan, 1983; Ein-Dor, Segev and Steinfeld, 1982; Fuerst and Cheney, 1982; Hogue, 1987; Kim and Lee, 1986; Raymond, 1985; Srinivasan, 1985) focus on the frequency that information is used with to determine the success of an information system. Other studies focus on the nature, voluntariness and use time of the output information to determine IS success.
The studies that focus on User Satisfaction to determine IS success often use this criterion on its own, due to several reasons (DeLone and McLean, 1992: 68). First, success of a IS is hard to refute when users say they like a system, derived from a high degree of face validity. Furthermore, user satisfaction is used more often because of the conceptual weakness of other success measures. Most research found in this category focuses on user (information) satisfaction (Bailey and Pearson, 1983; Bruwer, 1984; Doll and Ahmed, 1985; King and Epstein, 1983; Raymond, 1987) and overall satisfaction (Alavi and Henderson, 1981; Ginzberg, 1981; Mahmood, 1987; Rushinek and Rushinek, 1985, 1986). The Individual Impact of information systems as a measure of its success represents a step past the inward assessment measures described earlier. The results from research based on these measures allows for the assessment of the individual contribution of an IS to the organization. Goslar, Green and Hughes (1986) used the number of alternatives considered, time to decision, confidence in decision and ability to identify solution as individual impact variables to determine the IS success. These measures are also found in other studies, although sometimes defined differently; for instance quality of analysis, or alternatives/data consideration.
Finally, the Organizational Impact is concerned with measures that focus on the organizational effectiveness, i.e. profit or cost reduction as an expression of the IS success. Cron and Sobol (1983) and Jenster (1987) provide the most extensive research on information systems’ organizational impact success factors. These authors respectively identify pre-tax return assets, return on net worth, pre-tax profits and average 5-years sales growth, and economic performance, marketing achievements, productivity in production, innovations, and product and management quality.
DeLone and McLean (1992: 87) developed the first model presented in figure 2.1 from the main categories that are distinguished in the taxonomy. Here, the individual dimensions of IS success categories are incorporated to produce a coherent overall model of IS success (DeLone and McLean, 1992: 83). From the research a temporal dimension and interdependency dimension was identified. The first dimension focuses on the fact that organizational effectiveness is a continuous process rather than an end state. The interdependency dimension can be found in the relationship between user satisfaction and the use of a system. According to DeLone and McLean (1992: 83) “system quality
and information quality singularly and jointly affect both use and user satisfaction – positively and negatively – as well as the reverse being true. Use and user satisfaction are direct antecedents of individual impact; and, lastly, this impact on individual performance should eventually have some organizational impact”.
A useful model must include findings from other research as well as easily applicable in such a manner that it does not get caught up in contextual complexity. In figure 2.1 the authors demonstrate a logic interaction between the quality dimensions that were found form the literature study. Here six interdependent constructs are shown and the model both has an explanatory and predictive function.
2.2 The updated D&M model of Information System Success
The D&M model of IS success has been often applied in information research, which also lead to the review of the theory after ten years (DeLone and McLean 2002; DeLone, 2003; Petter, DeLone and McLean, 2012). From research contributions based on the original success model the updated D&M IS success model was drawn up (DeLone and McLean, 2002; DeLone, 2003), shown in figure 2.2. The figure shows that there are three major quality dimensions: System, Information and Service, which all have an impact on the (intention to) use and the user satisfaction. User satisfaction must be preceded by the use of the information system, which must in its turn be preceded by the intention to use. The combination of the use and user satisfaction will create certain net benefits; the outcome will create feedback to the intention to use and user satisfaction. Positive net benefits will lead to a reinforcement of the use and user satisfaction, while negative net benefits will lead to re-evaluation and most likely discontinuation of the IS.
DeLone and McLean (2002: 25) state that the net benefits measure might be the most important one, however cannot be analysed and understood without the underlying success measures. The six categories of measures identified are not independent but exhibit an interrelationship to each other. To clarify the major categories of success measures, definitions and examples of measures are provided in table 2.1 below.
IS Success Variable Definition Examples of measures
System Quality Desirable characteristics of an IS.
Ease of use, system flexibility, system reliability, and ease of learning, as well as intuitiveness, sophistication, flexibility, response time.
Information Quality Desirable characteristics of the system outputs (content, reports, dashboards).
Relevance, understandability, accuracy, conciseness, completeness, currency, timeliness, usability
Service Quality Quality of the service or support that system users receive from the IS organization and IT support personnel in general or for a specific IS.
Responsiveness, accuracy, reliability, technical competence, empathy of the personnel staff.
System Use Degree and manner in which staff and customers utilize the capabilities of an IS.
Amount of use, frequency of use, nature of use, appropriateness of use, extent of use, purpose of use
User Satisfaction Users’ level of satisfaction with the IS.
Single item to measure user satisfaction, semantic differential scales to assess attitudes and satisfaction with the system, multiattribute scales to measure user information satisfaction.
Net Benefits Extent to which IS are contributing to the success of individuals, groups, organizations, industries, and nations.
Improved decision making, improved productivity, increased sales, cost reductions, improved profits, market efficiency, consumer welfare, creation of jobs, economic development.
3. Public Private Partnerships
This chapter will focus on the question: “What are Public Private Partnerships and what are the
distinguishing characteristics of this project form?”. The first step is to conceptualize Public Private
Partnerships and its different forms that appear in practice and theory. Due to its contiguity part of this conceptualization will address PPPs in relation to traditional (IT) outsourcing. Additionally, insight in the implementation of PPPs for information systems projects will be provided, supplemented with an example of the application of PPP-DBFMO in an IS project abroad.
3.1 Public Private Partnerships in the DBFMO form
A Public Private Partnership (PPP) is a form of long-term cooperation between government and a private contractor or a consortium of actors. Additionally, in this manner it is possible for government to implement efficient solutions from the private sector for a reasonable price. For large housing and infrastructural projects in the Netherlands it has been shown that PPPs deliver better results than traditional procurement (Kenniscentrum, 2005).
The idea for Public Private Partnership (PPP) is deducted from the New Public Management reform philosophy that emphasizes private involvement resulting in efficiency and purposefulness, and the idea derived from governance literature that stakeholder involvement leads to more synergy and better products (Klijn and van Twist, 2007). A Public Private Partnership (PPP) is a long-term collaborative effort between a public and private actor, with the purpose of introducing private solutions in public settings (Klijn and Teisman, 2000). The goal of this cooperation is to produce better quality products at lower costs, by optimally dividing tasks and risks3. The public actor can be all kinds of public
organizations, however in most cases this is the central government. The private actor usually has the form of a consortium composed of a number of organizations and investment funds. Capital gains for government are created by private creativity and innovativeness.
There are several forms of PPPs (for more information see: HKEU, 2008), however due to the scope of this thesis the research by Eversdijk and Korsten (2009) is guiding in discussing various forms of PPPs. The authors have investigated this phenomenon in the Netherlands and found two main forms of PPPs; the concession and alliance models. The latter model is, according to scholars of the field public administration, the most interesting due to its partnership characteristics. Because of the genuine cooperation that is possible due to the equivalence in the working relationship, instead of a hierarchical/traditional division of labour that characterizes traditional outsourcing. An earlier thesis by Kleijn (2011) focused on the possible added value of PPP on governmental ICT projects. To this end the author has discussed the concessional and alliance models of PPP in his thesis. The research provides a discussion on the success and failure factors as well as risks that can be attributed to both modes of PPP. The conclusion of this thesis provides an overview of the success and failure factors for both modes of PPP. In research by the authors (Eversdijk and Korsten, 2008) the alliance form of PPP is considered a myth within infrastructural projects in the Netherlands. Furthermore, Eversdijk and Korsten (2008, 2009) indicate that the concessional model of PPP, also called a Design-Build-Finance-Maintain(-Operate) agreement, is the prevalent mode of PPP. The concessional model of PPP is a form of procurement where the consortium also obtains the right to operate the product or service it created.
3 ‘Publiek Private Samenwerking bij het Rijk’. Retrieved from:
http://www.rijksoverheid.nl/onderwerpen/publiek-private-samenwerking-pps-bij-het-rijk on August 20th, 2014.
3.1.1 PPP-DBFMO and (IT) Outsourcing
This section discusses the specific features PPP-DBFMO has in comparison to traditional outsourcing. As was shown in the previous section, the main features of PPP-DBFMO are the integrality and innovativeness of the partnership. We will look at the characteristic general properties of this type of collaboration with respect to traditional outsourcing.
There are some similarities between PPP-DBFMO and outsourcing, the first example is the relationship between public and private that is still hierarchical. Because there is a principal and a contractor there is still a principal-agent problem and an unequal partnership. In addition, there is a
classical division of labour between public and private, meaning that the public party retains the
directing role and the private party remains focused on the execution. The last similarity Eversdijk and Korsten (2009) found was that in both instances the choice of contractor takes place in competition. According to a statement by the Minister of Finance (TK, 2004: 5) ‘the tendering of a concession PPP must take place in competition’4.
When we turn to the properties that distinguish PPP-DBFMO from traditional outsourcing, literature provides us with several characteristics. First we see that different components are granted in one contract to a private consortium, thus ensuring better coordination of the components. Instead of this form of contractual integration, this would in traditional outsourcing have resulted in separate contracts with various private parties. PPP-DBFMO is not aimed at providing an object or product, but agreements are made on the provision of service. For instance with infrastructural projects the provision of a service pertains to the availability of a section of road that has been contracted to the private party. According to Eversdijk and Korsten (2009: 3) this form of payment for availability creates more complex financing structures. A consequence of this focus on service provision is that
functional requirements rather than product requirements should be met. However, in order to achieve
certain functional requirements, which allow for meeting the provision responsibility, product requirements can be assumed. Furthermore, Ruding (2008: 20) states that there is private financing
and public funding, in traditional outsourcing financing and funding were both done publicly.
Financing refers to the source of funds, while funding refers to who ultimately carries the cost of investment. There is an important role for project financers or banks in this type of partnerships between the public and private sphere. Furthermore, risk transfer from public principal to private contractor is an important property of PPP-DBFMO. Lastly, the duration of a PPP-DBFMO cooperation is greater in comparison to traditional outsourcing. In the Netherlands the duration varies between 15 to 30 years (Kenniscentrum, 2005: 4).5
These distinguishing characteristics of PPP-DBFMO are useful in a number of ways, according to Eversdijk and Korsten (2009). First of all due to life-cycle optimisations it is possible to acquire gains in efficiency, by the integration of several components that are necessary for the complete project. Due to the combination of all components in one contract the private service provider is more capable of an optimal coordination of these. This provides the private party with the room to manoeuvre the investment, realisation and long-term maintenance, using its creativity and innovativeness. Secondly, the purchase of service provision is bundled in one contract with a consortium, thus eliminating the
4 Kamerstukken II 2004-05, 28753, nr. 4.
5‘Verschil PPS en traditionele samenwerking’. Retrieved from:
http://www.rijksoverheid.nl/onderwerpen/publiek-private-samenwerking-pps-bij-het-rijk/verschil-pps-en-traditionele-aanbesteding on August 20th, 2014.
purchase risks of fragmented contracts. This will eliminate the risk of project components being neglected, because the controllability of the project coordination responsibility is shifted to the service provider. Third, the involvement of banks is expected to have a positive effect on the cost control of a project, by stricter compliance with the contract. The reason for this is the expectation that project financers together with the private parties in the consortium will make the most optimal balance between the investment and maintenance costs. It is expected that banks will supervise an optimal risk allocation and the revenue of the contract due to the repayment of the loan. Fourth, the risk allocation will be more optimal, in the sense that the risk is with the party that can control it. According to research an optimal allocation of risks to the party that is able to manage the risks will lead to cost saving, which may lead to a decrease in project costs. Lastly, PPP-DBFMO distinguishes itself by the fact that infrastructural projects have been realized in a shorter construction time. According to Eversdijk and Korsten (2009: 7) this can be expected due to the financial incentive to have a short building period, which leads to the start of the income – in the form of an availability fee – and thus the repayment of the loan.
On the other hand, Eversdijk and Korsten (2009) found that PPP-DBFMO also has shortcomings; one is that by the hierarchical and classical division of labour knowledge, experience and creativity are restricted. This is especially due to the mainly executing role of the private consortium. Another difficulty is the involvement of banks, because financing interest is aimed at avoiding risks. This is inconsistent with the need for interaction; flexibility and conscious risk taking that are needed in order to create capital gains in PPP-DBFMO (Kenniscentrum PPS, 2003). Thirdly, there are greater preparation expenses for a project with respect to traditional outsourcing. Fourth, Kerr (2006: 3) sees an adverse effect in the longer contract period, which may lead to an increase in operation costs because of the deployment of a company’s tender team. Lastly, as a result of the long duration of the contract inflexibility may arise, while the context may require changes to the contract (Eversdijk and Korsten, 2009).
3.2 Public Private Partnership and ICT projects
As has been mentioned before, PPP-DBFMO has mainly been used in infrastructural projects by the Dutch government. Research by Fife and Hosman (2007) focuses on PPPs that introduce ICT projects in developing countries. The case studies that were used to derive important success factors from in this research are based in Vietnam, where it is attempted to bring the benefits of technological efficiencies to the developing world. Since the focus is on PPPs in the ICT sector that are used to define and address a development challenge not all success factors that are identified are useful to this thesis. However, some of the success factors that Fife and Hosman (2007: 64) derived from their research can contribute to the goal of this thesis. Some of the factors are in a way applicable to PPP-DBFMO IS/IT projects in general.
First, the authors recommend to start out small, these modest efforts or smaller projects are better capable of adaptations. If the project appears successful and it inspires the confidence of the parties in the cooperation, it can always be chosen to scale the project. This is related to the second success factor; finding the appropriate partners for the partnership. Since all involved partners must be interested in the scalability of the project and in the long-term sustainability. Here the presence of a third party intermediary was found to be very useful by the authors, this actor brought potential partners together. The focus of this third actor in the PPP is not on profit, but on the formation of a sustainable partnership. Finally, it should be recognized by the stakeholders in a PPP that they are not the only two actors that are involved, in other words they should be aware of their context. As Fife and Hosman (2007: 65) state ‘buy-in by citizens is much more likely when a service demonstrates an
improvement over the status quo’. Beside these success factors there were also factors that were focused more on the successful integration of ICT projects in developing worlds.
One often cited ICT PPP project is the HERKULES project, commissioned by the German Federal Armed Forces to modernize and manage its non-military ICT.6 In 2006, the company ‘BWI
Informationstechnik GmbH’ (BWI IT) was founded, of which Siemens and IBM own 50,1% (split equally 25,05% / 25,05%) and the German Federal Government owns 49,9%. In addition, the ‘Bundeswehr’ (the collective name of the German army) is present on the board of directors and the military, German Ministry of Defence and General Accounting Office can audit the consortium. This modernization project has a duration of 10 years and the bid of the consortium was €7.1 billion (Portugall and Richter, 2010). According to the authors, the public sector used this form of collaboration to obtain private capital for the output of public service. Secondly, private management skills were integrated into the ICT project of the public sector. Third and finally, specific to this project was that private parties allowed for the use of their ICT expertise and competencies, and private partners buying power. Additionally, this project would standardise hard- and software, and it allowed the ‘Bundeswehr’ to focus on its core competencies. From the viewpoint of the private sector this collaboration could first of all provide them with access to a new economical sector, hence broadening its activities. In addition, the partnership offers the companies a lucrative investment of the company’s own capital.
As was discussed earlier the principal-agent theory applies to this classical or hierarchical division of labour between the German Federal Government and the consortium. Parties on both sides should, according to Portugall and Richter (2010: 25), obtain benefits through opportunistic behaviour. This behaviour might give rise to problems between the partners; first it could lead to hidden characteristics. Since the performance and quality of service of the agent only becomes known to the principal during the duration of the contract, which could lead to problems. Secondly, an agent’s hidden intentions are mostly unknown and only discovered during or after the cooperation. Lastly, a moral hazard can occur as a result of hidden actions by the agent, which are not observable by the principal.
In December 2013 the ‘Wissenschaftliche Dienste’7 of the ‘Deutscher Bundestag’ published a paper on the state of affairs concerning the Herkules project (Wissenschafliche Dienste, 2013). Here several initial start-up problems are described, which were solved after the expiry of the migration phase (2006-2007) and the integration phase (2007-2012). Since 2012 the Herkules project is in its operational phase and according to Veit (2013) the main goals of the project have already been reached. The success of this project is also endorsed Glaß (2013), stating that the project has passed its practical test. At this time, the German ‘Bundeswehr’ is considering its options for when the Herkules project expires, options are to re-enter a PPP or backsource the activities. The German federal government has until 2017 before the partnerships expires. In 2010 Portugall and Richter (2010) described a difficult start of the project due to its scale and complexity. However, as shown by the Wissenschaftlische Dienste (2013) the partnership has resulted in the achievement of the main objectives. Furthermore, the German federal government is at this moment looking at his options and also considers to renew the partnership.
6 ‘IBM & Siemens in Germany take up the labours of Herkules’. Retrieved from:
http://www.defenseindustrydaily.com/ibm-siemens-in-germany-take-up-the-labours-of-herkules-02925/ on August 20th, 2014.
3.3 Managing information system success in IT project sourcing
Before turning to the assessment of PPP-DBFMO, using the dimensions introduced in the model by DeLone and McLean (2002) that were described in chapter 2, we will shortly focus on one of the most characteristic features of PPP-DBFMO. The aggregation of operations important to a project is one of the biggest assets PPP-DBFMO holds over traditional outsourcing. This design provides an overarching structure that allows for one party (or consortium) to manage all facets of a project. This greatly increases the involvement of the providing party, as well as it ensures a fair distribution of risks. For this it is important for an outsourcing organisation to have sourcing knowledge in terms of the sourceable unit. According to Delen (2011: 45-46) sourcing knowledge is the ability to distinguish core competencies form non-core competencies. Furthermore, it relates to a certain level of technical knowledge concerning the information architecture of the organisation. Research also showed that experience with previous sourcing is decisive for the degree of success. Finally, Delen (2011: 46) states that hiring external advisers – in the case of insufficient sourcing knowledge – can have a positive contribution to the sourcing knowledge. The updated D&M model for IS success also exudes such integrality of the different success dimensions (DeLone and McLean, 2003: 23). Furthermore, the authors also stressed the connectedness and interdependence of the dimensions that are used to assess the success of an IS project (DeLone and McLean, 2002: 238).
3.3.1 Challenges to PPP-DBFMO in terms of IS projects
Finally, it is relevant to the scope of this thesis to also discuss some difficulties with PPP-DBFMO that may be important in the case of IS projects. As it was shown, the involvement of banks in PPP-DBFMO would have a positive effect on cost control and stricter contract compliance. However, according to Eversdijk and Korsten (2009) this also means that risks will be avoided and there will thus be less room for creativity and the conscious risk taking needed for create capital gains (Kenniscentrum PPS, 2003). One of the characteristics of information systems and technology is that rapid rate of succession with which they change. This is also known as Moore’s law, which states that processing power will double every two years.8 Furthermore, Korsten and Eversdijk (2009) stressed that the division of labour in PPP-DBFMO, which is still classical/hierarchical, can restrict the use of knowledge, experience and creativity from the private sector. An executing role for the consortium means that the actual innovation achieved from the partnership may be disappointing. However, as was also shown by the practical example of the Herkules project in Germany in paragraph 3.2, interests in the partnership can be divided in such a way that the government had a minority shareholding. Here it should however be noted that this may give rise to questions of accountability and liability of the government or public commissioning authority.
4. The effect of PPP-DBFMO on the success of IS projects
This chapter will use the D&M model presented earlier and the distinguishing characteristics of PPP identified in the preceding chapter to provide an answer to the question: “How do the distinguishing
characteristics of Public Private Partnerships affect success in an IS/IT project?”. First there will be a
brief review of these two components before assessing PPP-DBFMO with the model provided by DeLone and McLean. This exercise will result in a model that determines the effect of PPP-DBFMO on the success of IS projects. Subsequently, these findings will be validated by interviews and additional information acquired from the interviews will also be discussed.
4.1 D&M model and determining IS success
As was illustrated in the second chapter of this thesis, the DeLone and McLean model for IS success is a model with both explanatory and predictive function. The model will be used here in order to identify the implications that distinguishing characteristics of PPP-DBFMO have on the success of an IS project. The dimensions that were discussed in chapter 2 will be used to assess the characteristics and indicate its effect on IS project success. These dimensions are discussed in table 2.1 and will be guiding in the assessment of the effect of the distinguishing characteristics of PPP-DBFMO.
DeLone and McLean (2008: 238-239) describe seven different dimensions whose individual assessment provide insight in the success of an IS project. The relationship between these dimensions has been illustrated in figure 2.2, which depicts the updated D&M model for IS success. Furthermore, DeLone and McLean (2003: 23) state that “Each [the three quality dimensions] should be measured –
or controlled for – separately, because singularly or jointly, they will affect subsequent use and user satisfaction”. The dimensions that will be used to assess the distinctive features are: system quality,
information quality, service quality, intention to use, use, user satisfaction and net benefits.
4.2 Distinguishing characteristics of PPP-DBFMO
The work of Ruding (2008) and Eversdijk and Korsten (2009) has provided us with research on the distinguishing characteristics of PPP-DBFMO and the use they will have for a PPP-DBFMO project. Paragraph 3.1.1 provided an overview of the characteristics and added value of the PPP-DBFMO project form, which shall be listed here:
- Service provision obligation based on functional requirements - Project components integrated in one contract
- Private financing and public funding - Involvement of banks or private financers
- Longer contract duration allows for life cycle optimisations - Optimal risk allocation by risk transfer
These characteristics have been discussed more elaborately in the third chapter and will be used in assessing its influence on the success of an IS project.
4.3 The effect of PPP-DBFMO on IS project success
The following sub paragraphs will give clarity on the expected influence of PPP-DBFMO on the success of an IS project. All dimensions of the success model will be discussed and the characteristics that have shortly been discussed above will be assessed. In this way, it will become apparent what the
important criteria are when managing for the success of a PPP-DBFMO IS project. To limit the analysis of the effects that distinguishing characteristics of PPP-DBFMO have on IS projects we will now briefly determine what is meant by effects. The effects that will be discussed in this analysis will be the direct effects of the features on the IS project success dimensions. In other words, these are the consequences that emanate directly from a specific action or distinguishing feature or characteristics of some phenomena on a certain success dimension. The choice is made to exclude indirect effects from this analysis since the direct effect preceding these occurs on another dimension. This limitation to the scope allows us to describe the effects of the features of PPP-DBFMO more clearly. Furthermore, the connecting lines in the D&M model for success already represent these interrelationships – i.e. the indirect effects – between the different dimensions.
4.3.1 Quality Dimensions
The quality dimensions in the D&M model for IS success refer to the desirable characteristics of the system, information and service quality. In terms of outsourcing the main dimensions are the system and service quality dimensions, since the information quality is not or hardly manageable (DeLone and McLean, 2003: 22, Smithson and Hirschheim, 1998). This is due to information involved in an outsourced project usually belongs to the outsourcing party. However, for the sake of the success of a PPP-DBFMO IS project it is advisable, when information and its quality are at issue, to include clear agreements on the information requirements in SLAs.
The effect that can be expected from the shift to service provision based on functional requirements on the quality dimensions is that the product or service becomes part of the final result. In other words, the attention is shifted from providing the product or service itself to its delivery for the contract duration. An effect of this will be that the consortium will focus on the system or service quality in the design and build phases of the project. In view of the service provision obligation, it is in the interest of the consortium to evaluate different approaches to the project and determine best practices. This will provide the consortium with the context and guidelines for the design and build of the PPP-DBFMO IS project. Another effect regarding the system and service quality of an IS project will be that there is more focus on the operational service delivery capabilities, since this is what will ensure the public funding. Thus, system and service quality will meet service level requirements that are agreed upon. Therefore both the sourcing knowledge of a client as well as the expertise of the consortium will affect the system and service quality dimensions of the IS project. It can therefore be expected that service level agreements and service level requirements will be important in PPP-DBFMO. For both arrangements it is important that the client has sourcing knowledge, or a comprehensive knowledge of the business processes involved, in order to identify and set requirements correctly. It is also in the interest of the consortium – in order to prevent disputes – to pay attention to the arrangements and requirements at this stage. Furthermore, greater attention to the functionality aspect of the system and service quality can be expected, because the provision of the service will allow for the repayment of the private loan.
The effect of integrating the different project components in one contract on the quality dimensions will be that because of the experience and expertise of the consortium certain elements of an IS project will not be overlooked. Due to the consortium being responsible for providing the system or service, it is in their own best interest to assure the system, information and service quality. Thus, it can be expected that the consortium is committed to achieving a high level of quality of the IS project and will therefore assure that the contracts and arrangements are comprehensive.
Private financing and public funding will affect the quality dimensions in the sense that the system,
information and service quality must meet the operational requirements set in the SLAs and requirements. Since the private loan was contracted to the consortium and the consortium is dependent on the service delivery – or the quality of the usage dimension – for receiving public funding to repay the loan. Due to the dependency of the usage dimensions on the quality dimensions, the private financing will ensure that the consortium uses its expertise and experience to deliver high system, information and service quality. Another effect that can be expected from privately financing a project is that the innovativeness may be dampened. As was discussed earlier, banks and private financers will try to avoid unnecessary risks, which is at odds with the flexibility and conscious risk taking that innovativeness requires.
There is a slightly different effect that can be expected from the involvement of banks or private
financers, since these may encourage the consortium to consider new methods of more efficient
service provision. Examples are new technologies that allow for a more efficient or secure service delivery or offer a qualitative improvement to the system or service’s quality. However, as was noted earlier, the involvement of banks or private financers can also have a dampening effect on the implementation of innovativeness in an IS project. However, this may not have a large effect on the quality dimensions, this will affect the innovativeness of an IS project.
The longer contract duration of PPP-DBFMO will cause the consortium to have a more forward-looking perspective regarding the different quality dimensions. Thus, this will have the effect of a focus on the quality dimension in order to be able to comply with the service provision obligation. Furthermore, the effect of life cycle optimisations will primarily be present in the usage dimensions and net benefits dimension, which will be discussed in the following sections.
The effect of the optimal risk allocation on the quality dimensions of the IS project will mainly be indirect; most effects that have been described in this section are a result of the effect of this characteristics on PPP-DBFMO. In this project form, the consortium is responsible for the provision of the system or service and thus also for its quality. This way of allocating the risk is more optimal and will ensure that the consortium maximizes the quality dimensions of an IS project, because of the relationship of these to the usage dimensions. A direct effect of this risk allocation can also be distinguished, since the consortium is directly responsible for the design and build of the system. In other words, the effect of the risk allocation on the quality dimension is that the consortium is responsible for achieving the agreed and required quality of the system or service.
Quality Dimensions – Desirable characteristics of the
system, service and information quality
Service provision obligation based on functional requirements
QD01 - Determine best practices for IS project in the design and build phase
QD02 - Shift to operational service delivery capabilities in design and build phase and greater attention to the functional aspect or quality of the system or service delivery in the maintain and operation phase
QD03 - Sourcing knowledge required to determine service level agreement, requirements and determination of involved business processes (potential role for external advisors)
Project components integrated in one contract
QD04 – The consortium’s knowledge ensures the contractual comprehensiveness of the system or service, thus the expertise and experience of the consortium will ensure that project specific elements are included
Private financing and public funding
QD05 - Dependency of the usage dimension on quality success and the fact that the consortium is dependent on successful usage (i.e. service delivery) for public funding, is an incentive to deliver the required quality
QD06 - Private financing can limit the flexibility and conscious risk taking in a project, which are labelled as a requirement to innovation (Kenniscentrum PPS, 2003)
Involvement of banks or private financers
QD07 - Can provide incentive to create a more efficient or effective service delivery by putting forward new techniques or methods
Longer contract duration allows for life cycle optimisations
QD08 - More forward-looking perspective regarding the quality dimension; e.g. by investing in techniques that offer a longer service life or have improved maintainability
Optimal risk allocation by risk transfer
QD09 - An indirect effect is that the consortium is responsible for the service delivery and thus also for providing the system and service quality that allow for the delivery
QD10 - A direct effect is that the consortium is responsible for achieving the system or service quality that was agreed in service level agreements or requirements
Table 4.1 Effects of distinguishing characteristics of PPP-DBFMO on quality dimensions
4.3.2 Usage Dimensions
The preceding section already shortly focused on the usage dimensions stressing its dependence on the quality dimensions. We will now turn to assessing the distinguishing characteristics of PPP-DBFMO on the (intention to) use and user satisfaction dimensions. Success in the usage dimensions is important to a PPP-DBFMO IS project, since successful use of the project will ensure public funding. The effect of the service provision based on functional requirements of PPP-DBFMO on the usage dimension of an IS project is that the consortium is responsible for the alignment of the (intention to) use and user satisfaction. In addition the usage dimensions will influence and be influenced by the net benefits dimensions, which will be discussed in the following section. The success of the use dimension of the IS project is determined by the provision of the service and thus dependent on the quality dimensions. Furthermore, since the focus in PPP-DBFMO has shifted to the delivery of the service, it is imperative that the (intention to) use dimension is successful in order to receive public funding.
Integrating the different project components in one contract will affect the responsibility of the
consortium regarding the service delivery. In a PPP-DBFMO IS project the consortium must ensure that the system or service delivery contracts are conclusive and service delivery is successful. This means that the consortium will be involved with the service or system users, in order to assure that there is successful alignment of the (intention to) use and user satisfaction. It is important to be
