AUTHOR
Barrie Thompson and Helen Edwards

ABSTRACT

It is widely recognised that Information Society Technologies (IST) are changing the world. For example, in the documentation relating to the priority thematic areas for research within the European Union Framework 6 initiative, it is stated that:

“Information society technologies (IST) are transforming the economy and society. Not only are they creating new ways of working and new types of business, but provide solutions to major societal challenges such as healthcare, environment, safety, mobility and have far reaching implications on our everyday life. The IST sector is now of one of the most important of the economy, with an annual turnover of EUR 2000 billion, providing employment for more than 12 million people in Europe.” (FP6, 2002)

However, as is highlighted in the call for papers for Ethicomp 2004 such technologies raise significant social and ethical risks for individuals, organisations, and society at large. In particular, there are major challenges and responsibilities for the Information System and Software Engineering professionals who must design, develop and support IST based systems. It is therefore imperative that in the academic programs that educate such future professionals there is an appropriate treatment of the relevant ethical, professional, and legal issues that they are likely to encounter in their careers.

At the University of Sunderland we offer a wide range of taught programmes at Masters level in computing. In the majority of these programmes, the students undertake a module entitled Research, Ethical, Professional and Legal Issues (REPLI). The purpose of the module is to provide the students with not only an ability to undertake postgraduate research but also an appreciation of relevant ethical, professional, and legal issues they are likely to meet in the future. In this current academic year the module will be undertaken by more than 300 full and part time graduates on campus who are undertaking six different Masters programmes ranging from a general MSc. in Computer Based Information Systems to a very specialised MSc. in Software Engineering. It will also be taken by a further 300 students at some ten Distance Learning Centres that are based across the world form Hong Kong to Nairobi. Details on the initial development of the REPLI module for two particular Masters programmes, including the influence of early Ethicomp conferences on its content, was included within a paper presented at the 1999 Australian Institute of Computer Ethics Conference (Thompson and Edwards, 1999). However, within the proposed paper for Ethicomp 2004 the following will be covered:

1. How the module has further developed to cater for students from many different national and educational backgrounds.
2. How the teaching and learning strategies has been adapted to meet the needs of a widening range of programmes.
3. How ethical, professional, and issues are addressed within the module including the use of case studies and role-play to emphasise particular aspects.
4. Conclusions regarding the effectiveness of the approaches adopted and what could be done to further improve matters.

Special attention in the paper will be paid to:

The attempts that have been made to involve the students in more practical activities. A major instance of such has been the use of a professional Code of Ethics (such as the Software Engineering Code of Ethics and Professional Practice produced under the auspices of the IEEE-CS and the ACM) and a fictitious case study “The Case of the Killer Robot”. Full details of the case study can be found in the text by Epstein (1997), however, we have used an abridged version which is freely available on the Web (e.g. at http:/ricis.cl.uh.edu/FASE/Killer-Robot.html). In the Killer Robot study it first appears that a fatality caused by an assembly line robot was due to errors, caused by a programmer, in the software used to control the robot. However, as the case study progresses it becomes more and more obvious that there are many other people who could be held fully or partly responsible. In the practical activities undertaken by the students they are divided into teams, one member of each team must present the case for their character being held responsible and another team member must present the opposing view. After each presentation there is a short time for questions from the class. Finally, the class votes on the relative guilt or innocence of the character. In the paper the various profiles of votes produced by students from different programmes and years will be presented and discussed in detail.

REFERENCES

FP6 (2002), Sixth Framework Programme of the European Community, for Research, Technological Development and Demonstration Activities, The priority areas are detailed at: ftp://ftp.cordis.lu/pub/fp6/eoi-instruments/docs/eoi_annex1.pdf

Software Engineering Code of Ethics and Professional Practice. Details of this are available at: http://computer.org/tab/seprof/code.htm

Epstein R.A.G. (1997), “The Case of the Killer Robot, New York”, John Wiley and Sons, New York.

Thompson J. B. and Edwards H. M. (1999), “Ensuring that Ethical and Related Issues are Addressed in Two Masters Level Computing Courses: a Report on Developments and Experiences Over a five Year Period” Proceedings of the Australian Institute of Computer Ethics Conference AICEC99, 14-16 July, Lilydale, Melbourne, pp 365-375.

AUTHOR
Stanislaw Szejko

ABSTRACT

Computer ethics is about principles related to behaviour and decisions made by computer professionals and users, including software engineers, operators, managers, policy makers, as well as educators and students. This means all these people involved should be supported with some “policies for ethical conduct” i.e. policies that guide their actions and increase adequacy of the decisions they make [4].

Three ways of reaching these can be typically met (in spite of everyday education and personality development):

• publishing and promoting codes of ethical conduct like ACM/IEEE Software Engineering Code of Ethics and Professional Practice [9];
• assuring appropriate approaches and techniques in software engineering, like risk reducing or ethical impact analysis ones [6];
• systematic and ordered case analysis that introduces and teaches “good ethical judgement and decision making” [1]. This approach is seen as a kind of pattern recognition as its core is to support detecting ethical situations, see what is right and what is wrong and make the decisions.

However, why shouldn’t we take a step further and establish a set (or sets) of ethical behaviour patterns, lessons drawn from actual situations and good practices, that could be identified and applied throughout a software lifecycle?

Software patterns are a recent software engineering problem-solving discipline that emerged from object- orientation and formal specifications. A typical definition of a pattern names it “a proven solution to a problem in a context” while Coplien proposes[3] “a recurring structural configuration (i.e. a rule to configure) that solves a problem in a context, contributing to the wholeness of some whole, or system, that reflects some aesthetic or cultural value”. It is typical to define patterns within a framework

• Pattern name – an identifier tha characterises the problem
• Problem and its context targeted aims
• Context of the, possibly repeating, situation.
• Forces (tradeoffs) that introduce circumstances and limitations
• Pattern proposed solution
• Results and consequences of applying the pattern proposed solution
• Known pattern applications – this increases confidence in its usage

and classify them into several categories [4] [5] [10] : analysis patterns, design patterns, architecture patterns and application frameworks. In the sequel, the advantages of representing practical knowledge and experience as codified patterns make the technique be applied not only to software development but also to organisational and managerial problems thus making organisational and project management patterns (by the way, isn’t this idea very close to the concept of holons in SSM [2] ?) Two examples (taken from an impressive list in [3]) are SizeTheSchedule developed to prevent from overly ambitious or too general schedules, and TeamPerTask intended against big diversions of project teams that threaten to disrupt the ongoing work and temporarily halt progress.

We can easily think of some patterns of ethical behaviour and decision making that describe the suggested behaviour in software engineering – as in project management process, for instance

• AssessFS could be aimed at taking decision of evaluating objectivity of a feasibility report which is delivered by an external company,
• PerformExtendedAnalysis intended to identifying if the ethical impact analysis should be performed,
• InformClient devoted to answering questions if and when throughtout the life-cycle a customer should be informed about the project progress and problems,as in systems usage and maintenance, for example
• AssisstInE-mailDilemma could be aimed at supporting decisions of Rogerson’s computing officer having supervisor privileges while a crisis arises [8].

Thus a pattern language for project management problems (which is understood as an outline of ways that the patterns can be put together [3]) could be meaningfully extended with the ethical behaviour patterns. The more, typical situations in a project lifecycle where patterns apply could be identified as well.

The paper presents the idea of software and organisational patterns in more details. It introduces a concept of ethical behaviour and decision making patterns, brings examples of their specification and puts a proposal to enrich the project management and organisational pattern language with a set of ethical patterns that can be applied along system and project lifecycle. The paper ends with stressing possible concerns about codifying and applying “ethical solutions” and relates the idea to seeking ethical dimensions within previously set patterns in software engineering and project management.

REFERENCES

[1] Bynum T.W., Rogerson S.: Computer Ethics and Professional Responsibility. Blackwell Publ., 2003

[2] Checkland P., Scholes J.: Soft Systems Methodology in Action. J. Wiley & Sons, 1990

[3] Coplien J.O.: Software Patterns. Bell Laboratories. http://hillside.net/patterns

[4] Flower M. Analysis Patterns, Addison-Wesley Publ. Comp, 1997.

[5] Gamma E., Helm R., Johnson R., Vlissides J.: Design Patterns. Addison-Wesley Publ. Comp., 1995.

[6] Gotterbarn D.: Reducing Software failures: Addressing the Ethical Risks of the Software Development Lifecycle. Proc. of the 5th International Conference on The Social and Ethical Impacts of ICT ETHICOMP 2001, Gdansk, June 2001, vol. 2, pp. 10 – 19.

[7] Moor J.H.: What Is Computer Ethics? http://www.southernct.edu /organizations/rccs/.

[8] Rogerson S.: What Is Computer Ethics; Computing the Social Cost. Slides. CCSR, De Montfort University, Leicester, UK

[9] Software Engineering Code of Ethics and Professional Practice. ACM/IEEE-CS Joint Task Force on Software Engineering Ethics and Professional Practices. http://www.acm.org/serving/se/code.htm

[10] Szejko S. (ed): Software Development Methods (in Polish). MIKOM, Warsaw 2002.

AUTHOR
Bernd Carsten Stahl, Chris Wood and Richard Howley

ABSTRACT

The importance of ethical issues in computing does not have to be emphasised at a conference such as ETHICOMP. It will also be not be contentious to state that one important part of addressing these issues is through the education of future computing professionals. How this is to be done is a rather different matter. In this paper we want to develop a new approach to teaching professional issues in computing by allowing students to develop their own codes of ethics. The purpose of this paper is therefore to develop a sound theoretical foundation for this endeavour which we intend to put into practice in our own teaching.

The student target group for whom we design this module are higher national diploma students in the UK. This means that we are looking at a population of students who do not aspire to the highest academic standards but who are instead interested in a reliable education that will promote their professional outlook. While this sort of student may be susceptible to the idea that ethical matters are possibly of importance to them in their future career as computer professionals they generally have little formal training or experience in dealing with such problems. Furthermore, in a course such as this one there are severe time constraints with regards to time and resources available. While there is a professional issues module in the course this will largely be taken up by practical skills such as project management.

Given this general situation we tried to identify those aspects that might influence the success of the course. At the same time we were looking for theoretical considerations that would allow us to conceptualise the course in a promising manner. We did this by considering two fields of theory: On the one hand we looked at different theories of teaching and learning, particularly as they pertained to the use of computing in learning. This is justified by the fact large parts of the course will take place in computer laboratories where e-teaching and e-learning will be a possibility. On the other hand we looked at ethical theory and how philosophical ethics could help us in the design of the course. Here we put our emphasis on that part of the ethical literature that is concerned with questions of computing or information. Of course this is quite an ambitious undertaking and we cannot hope to do justice to even one of these theoretical fields in a short paper such as this. Our choice of topics within those two fields is therefore necessarily somewhat eclectic but we hope that the readers will still be able to follow our approach.

The most important distinction we extracted from the theory of e-teaching and e-learning was that between constructivist and positivist teaching models. Positivist teaching models are based on the assumption that there is a given reality independent of human perception. True statements can be made by describing this reality the way it is. Teaching thus consists of transferring as many of such true statements and observations from the (knowing) teacher to the (not yet knowing) learner. In the constructivist paradigm knowledge is seen a something that is constructed by the learner through the process of learning whereby the learner’s constructions represent the relevant reality. While the distinction between positivism and constructivism raises many difficult philosophical problems it seemed to us that for our purposes only a constructivist approach could be justified. Ethical problems and ethical rules are entities that belong into the sphere of social constructions rather than natural occurrences.

A look at the ethical theory informed us that there is no generally recognised approach to teaching ethical issues in computing just as there is no generally accepted ethical theory. One possible way of proceeding with our task was thus to introduce the students to philosophical ethics by briefly discussing the most important ethical theories. However, this approach ran into problems because of the time restrictions as well as the lack of prior knowledge on the part of the students. Another approach to computing ethics that is often put forward as a practical solution but also as a way of teaching the subject is the use of codes of ethics. This seemed more practical to us, however, the use of a given code such as the ACM or the BCS code was problematic because it collided with our above decision of using a constructivist approach. Simply giving students a prefabricated code of ethics and indoctrinating them with it (or even critically analysing it) struck us as typically positivist.

The solution that we chose was therefore to have the students develop their own code of ethics for computer use. This combined several advantages: One of the most important arguments for this approach was that it will allow students to use their prior knowledge and experience to identify the ethical problems that computing use can raise. This will allow them to raise their level of awareness and should at the same time prove interesting to them. Second, by using this approach we are not committed to a particular ethical theory but instead rely on the students’ moral intuition. This should create contradicting views and interpretations which in turn should trigger ethical discussions which will allow the students to examine their own convictions. Third, the development of the students’ code of ethics should produce a “living” document which should allow the students to reflect on their own practices without the constant threat of sanctions. Fourth, students will, when confronted with other codes during their future careers, be able to better understand and critically reflect these codes.

A central problem of this approach is that it largely neglects considerations of traditional ethical theories. We realise the importance of teleology, deontology, virtue, justice etc. for these matters but we also had to keep in mind the limitations of the module and the students. They will generally have not background in philosophy and little aptitude for reading complex texts. Our fear is that exposing them to Kant, Mill, Rawls et al. will put them off ethics rather than explicate their moral reasoning structures. Also, we hope that the process of producing their own code of conduct and applying it to case studies and contrasting it with others will encompass will lead them to consider these reasoning skills. We thus see this module as an exercise in raising awareness which, if successful, should be followed up on by a more philosophically minded module.

In writing this paper we will try to lay the theoretical foundations for a module on professional standards for Higher National Diploma students. Our purpose in submitting this to the ETHICOMP conference is firstly to get some feedback that will hopefully allow us to develop these ideas. Also, we will review the theoretical considerations in the light of the teaching reality which will by then have taken place and will hopefully allow us a useful critique of our approach.

AUTHOR
Paul Simpkins

ABSTRACT

A analysis of the tensions in delivering e-government and joined up working in the public sector without infringing the rights of the individual to privacy in respect of their personal data.

The existing legal framework which governs public bodies’ powers to collect, use and share personal data is far flung and often difficult to track down.

There is an interaction with the Data Protection Act 1998, the Human Rights Act 1998, the common law tort of breach of confidence and other relevant legal provisions.

Essentially Public bodies cannot share data (or indeed buy paperclips) unless there is a specific power established by statute that they can do so.

In some cases specific statutory powers have been taken to ensure that proposed data-sharing is lawful. An example of a lawful gateway can be seen in section 122 of the Social Security Administration Act 1992. This provision empowers the Inland Revenue and the Commissioners of Customs and Excise to supply the Secretary of State for Work and Pensions with information for use in connection with the prevention of social security offences or for checking the accuracy of social security information.

Another example is section 115 of the Crime and Disorder Act 1998 which authorises any person to provide information to the police, local authorities, probation authorities, health authorities in connection with the purposes of that Act.

Many other acts of parliament create gateways for the passing of data from one body to another. One of the main gateways is strangely the Data Protection Act 1998 which permits disclosure of personal data where it is required by any enactment, by any rule of law or by any order of the court.

However there is no specific power for public bodies to buy paperclips.

This came to a head with the debate about re-use of council tax data for other purposes such as debt recovery and the verification of entitlement to benefits and concessions. After taking a leading Counsels opinion the Data Protection Registrar in the late 90s opined that it was not lawful to use council Tax data for any other purpose other than Council Tax. Since then a series of equally eminent barristers have interpreted the legislation in a different way based on general powers hidden in ancient acts such as section 111 (1) of the Local Government Act 1972 that provides that a local authority ‘shall have power to do any thing ? which is calculated to facilitate, or is conducive or incidental to, the discharge of any of their statutory functions’.

Similarly Section 2 (1) of the Local Government Act 2000 that provides that a local authority shall ‘have power to do anything which they consider is likely to achieve any one or more of the following objects – (a) the promotion or improvement of the economic well-being of their area; (b) the promotion or improvement of the social well-being of their area; (c) the promotion or improvement of the environmental well-being of their area’.

The issue is complicated by the recent arrival in UK courts of the Human Rights Act. Previously reserved for mainland europe for the last 3 years individuals have been taking action in the courts to protect their image, their home, their children from intrusive photography or journalism. As well as famous people flattering their egos some ordinary human beings have taken exercised article 8 and their right to respect for their home their family life etc.

Unfortunately Article 8 is a qualified right and although many claim it not all are granted.

Down at the chalk face however there is a general assumption that if an individual consents to their data being used by one public sector body that it will be shared by others. The sectors of Health, Education, Police and The Council are considered by many to be actively sharing data. Those most in need of public services don’t have the time to debate their qualified right to privacy or the powers of one public body to share data with consent – they just want their service delivering on time.

Leading commentators feel the rush to share data even with consent does not sit comfortably with the rights of the individual.

This paper will consider how these tensions can be resolved.

AUTHOR
Philip Seltsikas and Helle Zinner Henriksen

ABSTRACT

Introduction and presentation of the research question

The diffusion of information communication technologies (ICT) have impacted collaboration patterns at various levels of society. Most institutions and businesses throughout the world have experienced a number of changes in the way collaboration can be performed due to the introduction of new means of communicating. Especially the “re-birth” of the Internet has fuelled new ways of collaboration. Without going into details of the history of the Internet it can be argued that the media has experienced a re-birth in the mid-90s when the first browsers were distributed to the public and thereby made the, by that time already aged, Internet accessible to businesses and individuals (for a detailed account of the developments from the ARPPANET to the Internet see for example (Kalakota and Whinston 1996). The rate of diffusion of the Internet in Western countries is by far faster than the diffusion of other previously introduced communication technologies. This has created a number of opportunities for businesses, public institutions, and individuals. Leading to phenomena such as e-business, e-government, and e-learning. In the first generation of e-business, e-government, and e-learning emphasis was on exploiting the opportunities at a national level. However, acknowledging the global reach of the Internet focus has shifted towards trans-national collaboration in the present generation of e-business, e-government, and e-learning.

Businesses were among the first players to observe and explore the opportunities created by the Internet in the form we know it today. The collaboration among businesses at a trans-national level has been researched intensively within the realm of e-business for quite a while. Research has mostly focused on issues related to the traditional dimensions of firms’ internationalisation. These dimensions are among others related to 1) the internationalization patterns of firms (products and value chain activities, operation methods and control, markets and geographical spread) and 2) internationalisation capacity of firms (technological resources and production economies, human resources and expertise, financial resources)(Petersen and Welch 2003). Common for the list of activities represented in the dimensions is that they aim at improving the profitability of the firms involved in the arrangements.

Another type of institutions, which have taken the opportunity to deploy the new means of collaboration due to “re-birth” of the Internet are universities. The adoption of the Internet among universities has contrary to the private businesses been driven less by profitability. In the early days when the Internet was still known as the ARPANET universities and other research institutions benefited from the network in their scientific activities (Kalakota and Whinston 1996). However, another central task of universities, which has prospered from the massive diffusion of the Internet is teaching. Consequently, e-learning came on the agenda after the Internet had diffused into households in the Western part of the world. Some universities have offered full or partly education over the Internet thereby creating new opportunities for those hindered in participating in the Socratarian way of knowledge acquisition (Dreyfus 2001). Other universities have found other ways of utilizing the media by creating curriculum where the foundation for teaching and participation has been the Internet. An example of such a curriculum is the GEM-programme (the Global eManagement, Executive MBA). The GEM-programme facilitates virtual teaming for distance learning and meeting technologies so participants can interact electronically (www.gem.cbs.dk). Given that universities and business schools from six European countries and Mexico participates the GEM-programme represents an illustrative example of trans-national collaboration fuelled by the Internet.

Common for trans-national business and teaching is the necessity for mutual confidence and commitment from those parties involved. The objective of this paper is to present how a platform for trans-national collaboration takes shape and how the involved parties prepare the implementation among the players involved a recently launched project, which focuses on establishing trans-national collaboration.

The empirical setting and research method

In June 2003 four countries, Austria, Croatia, Italy, and Slovenia, inaugurated a trans-national collaboration project aiming at creating a single electronic market in the region. The Internet is to be used as the communication infrastructure. The project aims at strengthening e-collaboration between the countries and to develop single electronic markets in the region. The notion of single electronic markets should be understood in the broadest possible sense given that it involves a number of sectors and institutions from the countries involved. One of the activities within the collaboration between the four countries is the ALADIN project (the ALpe ADria university INitiative). The ALADIN project focuses on the strengthening of the collaboration on e-commerce teaching between universities in the region. The network is historical in the sense that the trans-national collaboration is based on the Internet as a means for communication. By using the Internet the project reaps the benefits of having an established infrastructure as a means for communication and collaboration from day one. External researchers, who will monitor the developments in the project, will follow the project closely. The objective of the research is to investigate how new ways of trans-national collaboration are being established in the region. The investigation will consider which sectors and institutions are first-movers and particularly what their motivation for joining the project is. The research will provide a critical analysis of the implementation processes associated with the establishment of the trans-national collaboration. The results will potentially contribute to an understanding of the qualitative characteristics of the collaboration. The results should be used to manage the collaboration structure toward a sustainable cooperative relationship. Other regions and disciplines hoping to institute similar cooperation networks can learn from these results and potentially replicate the successful aspects of the processes.

The main data collection technique applied in the study of the trans-national e-collaboration project was active participation in project meetings and events organized on behalf of the project. The data was analysed using qualitative research instruments.

REFERENCES

Dreyfus, H. L. (2001). On the Internet, Routledge, London.

Kalakota, R. and Whinston, A. B. (1996). Frontiers of Electronic Commerce, Addison-Wesley Publishing Company, Inc..

Petersen, B. and Welch, L. S. (2003). “International Business Development and the Internet, Post-hype.” Journal of International Business, 1(43), 7-29.

AUTHOR
Roderick Mendoza Rupido

ABSTRACT

This paper explores the dynamics of information technology (IT) towards achieving a socially responsible economy and their roles and implications to growth and development. It presents evidences to show that economic growth is strongly related to the extent of an economy’s initial social development.

Society is changing rapidly under the impact of information technologies although we are not sure how if it is for better. Increase in information and innovation brings about benefits coupled with big challenges. The major challenge is how to manage this valuable resource in a responsible manner. The information revolution is changing the nature of information from a physical and tangible entity to its digital counterpart. Networks have expedited the movement of information so that users can be informed more quickly and efficiently. Control of information resources is a critical source of competitive opportunity. However, an even larger challenge arises in managing and implementing technological changes towards social transformation.

The Internet, for example is beginning to transform the way we do things. It seems to engulf the whole world and has certainly transformed our lives and dramatically reshaped the way we live. Its influence has been pervasive and profound. But this rapid diffusion has also created some notable ethical and social challenges.

With the increased access to information and the redefinition of information as a commodity, the potential for misappropriation and/or misuse of information may likewise pose some threats. Thus, while some challenges deal with the management of information, others deal with the development of technologies that create, process, store and retrieve such information.

In tandem with access and speed, advanced hardware and software systems enable users to enhance the quality and relevance of information. Through sophisticated search engines, research can be conducted in a relational manner, delving deeper into topics and associated themes. However, the qualities of information acquired are crucial and are deemed instrumental to the individual’s social and moral development.

The nature and focus of information technology needs to map balanced views for monetary and non-monetary benefits. Taking this seriously will provide key to understanding and learning social development through proper utilization of information and communication technologies.

The nature of IT has to change in order to reflect its new roles to meet the need for a more holistic approach towards social development. Evaluation needs to be more wide-ranging to incorporate the richness of the intangible and unexpected costs, benefits and risks of modern IT projects. Sensible evaluation decisions should be made by the careful analysis and fair judgment, with social responsibility as the core basis.

Information Technology transformation should be regarded as a process and as a continuous opportunity for learning. This will promote a deeper understanding of its content, context and process. It is through this richness of understanding that situations will be dealt with more accurately. Evaluation research and practice therefore, needs to be used to meet the demands of the dynamic changes since new information and communication technologies have the potential to make significant impacts.

Advances in information technology have generated some unique and difficult problems. The increased availability, mobility, and value of information create a special responsibility to maximize benefits and prevent its misuse. Specific concerns should be geared towards morally upright utilization of information and communications facilities, which calls for setting up of ethical and social management systems to benefit the society.

The study attempts to answer the following specific questions: What are the effects of IT transformations to an economy? What is the role of IT in achieving socially responsible economy? Assuming that the forces that encourage IT transformation will remain and become much stronger in the future, how can the government manage their significant effects without necessarily impeding progress of economic growth?

IT requires sound management for sustained and continued growth, improving the social capabilities and upholding the ethical and moral standards of the society. Its development has to be managed effectively to achieve a socially responsible economy. Policies, processes and systems must be developed, implemented, maintained and updated to ensure that information acquired and/or used really contribute to the demands of the society.

Institutions must strive to ensure that IT utilization is judiciously made in decision-making and priorities are set to develop a morally endowed economy. This challenge must be addressed not only by the individuals, but also by the entire global community.

Note:
This paper is conducted at Graduate School of Economics, Kyoto University, Japan under the International Doctoral Program in Economics funded by the Japan International Cooperation Agency (JICA). The author acknowledges the invaluable support of Professor Takashi Hikino of Department of Economics, Kyoto University for his support, comments and suggestions.