Volume 17, Number 2 June, 2005
A Preview of Ongoing Research in the Field of Management - Issues in E-Government Systems Implementation and Evaluation :
There is widespread interest in e-government systems in India with the central and state governments having promoted a large number of such projects over the last few years. The results are not encouraging, though, as the ‘success’ rate of such systems is a meagre 15% according to the World Bank (which has funded many of these projects). The reasons are many and not hard to uncover. The bigger challenge is to find a way to design, implement and evaluate e-government systems such that they are successful and sustainable.
Research in e-government systems is in its infancy, with the published work focusing mainly on technology descriptions or case studies of not much depth, and there is a need for rigorous analysis and theory building. The research reported in this article is an attempt to address some of the issues in egovernment systems analysis, implementation and evaluation. The study, undertaken by Dr. Rahul De’, Hewlett-Packard Chair Professor of Information and Communication Technology for Sustainable Economic Development, IIM Bangalore, was funded in part by a grant from the Ministry of Human Resource Development and by the Centre for Public Policy, IIM Bangalore.
The e-government systems considered for this research are public systems, or government-to-citizen systems, that provide a service or product to citizens at large. These systems are massive public interventions with huge outlays by state governments and with widespread deployment of resources. The focus of this research is on systems that address the urgent needs of the rural population, who have limited access to electronic infrastructure and low knowledge of technology and systems. One such is the Bhoomi system of Karnataka, India, that has been deployed all over the state and provides farmers with access to land records at a nominal price. The system has been deployed in 177 taluks of the state and contains about 7 million records in its database. Bhoomi was rolled out in 2001 and by late 2004 the system had sold over 23 million land records to citizens, making it one of the most widely used e-government systems in India.
Owing to the complex and dynamic nature of the object of study, the case study research methodology, which enables the researcher to examine the phenomenon in its natural setting, employing multiple methods of data collection to gather information from one or a few entities (people, groups, or organisations), was deemed most appropriate. More importantly, the case study method is useful when the boundaries of the research are not clear at the outset, and no experimental control or manipulation is possible. The ongoing data collection for this research was begun in early 2003. The sources include interviews with the Project Champion (a senior IAS officer in the Karnataka state government) and five other district officers involved in the implementation of Bhoomi, interviews with farmers, a survey of 63 users of the kiosks from three different districts of Karnataka, a survey of Bhoomi kiosk operators, over 900 pages of Bhoomi internal reports, 36 media reports on Bhoomi and other e-government systems in India, reports from development and funding agencies, and student reports on field visits. Given below are some important findings of the research, including both theoretical conclusions and observations and insights gained from the data.
Stakeholders
In the field of Information Systems a stakeholder is defined as a person or group that is able to have an impact on the eventual system in a practical sense. By this definition one has Issues in E-Government Systems Implementation and Evaluation A Preview of Ongoing Research in the Field of Management 14 On the Anvil to include all parties who can affect a system, whether their traditional roles and responsibilities are enhanced by the system or depreciated. It is useful to view the stakeholders that impact the eventual success of an e-government system as belonging either to the demand-side (those who will consume the services of the system), or to the supply-side (those who fund, design, implement and maintain the system). Individuals, groups and organisations belong to either stakeholder group, or both, according to their relationship to the system. In developing countries, most e-government systems implementations are driven by the supply-side, who control the resources and design the services, processes and the architecture of the system without consulting any demand-side stakeholders, and who are largely informed by their own ideological commitments or the technological imperatives of their commercial partners .
Demand-side stakeholders consume the services of the egovernment system and, on occasion, provide the revenues that sustain the systems. There are instances where demandside stakeholders such as citizens’ groups and civil society groups have demanded that they be included in the implementation process but this is rarely achieved. They influence the eventual success of the system through use or non-use and are directly impacted by the service efficiencies achieved.
The theory of stakeholder groups outlined above enables a nuanced understanding of the conflicts that arise with the implementation of e-government systems. One example of such conflict is the exclusion of the traditional village accountants from having sole control over managing the land records, which was meant to prevent corruption and provide the state government better and updated information on cropping patterns and land issues. However, for many farmers this meant the loss of their access to a power base. Similarly, the introduction of the district-level official, the Tehsildar, in the land record mutation process aggregated control and power at the district-level, and reduced the power of revenue inspectors, village accountants and Shireshtedars on the supply-side. However, it may also impact the demand-side adversely by increasing the official review time and, in some cases, the corruption levels.
First- and Second-Order Effects
When new technologies diffuse into the economy, the effects are spread out over time, and are referred to as first-order, second-order, and third-order effects. These effects may be examined in the context of e-government systems which are also complex interventions in the economy of a region and in the social environment. First-order effects have to do with the immediate changes in the volume and speed of the service delivery. Second-order effects are subtle, and include the effects on institutions and groups that used to rely on the government service. Second-order effects enable possible reduction in transactions costs for these agencies that reflect in their productivity or performance
First-order Effects: Some of the important first-order effects of the Bhoomi system, the immediate effects of computerisation of a manual system, are described below. The results are from a published report by the Public Affairs Centre in Bangalore (PAC) and from an independent study conducted for this project (IR survey).
- Of the users who had used the earlier, manual system and also the Bhoomi system, 78% found Bhoomi to be simpler to use (PAC). In the IR survey 90% found the system easy to use.
- n the manual system about 60% of the respondents had to meet 2-4 officials for their work, whereas with the Bhoomi system 79% could complete their work by interacting with kiosk staff only (PAC).
- Of the users of Bhoomi, 74% stated that their documents were error-free (PAC), whereas 63% said so for the manual system. (81% stated documents to be error-free in IR survey).
- The costs associated with travelling to the taluk headquarters was an average of Rs 47.10 from the IR survey. The PAC Report shows that 28% had to make more than one trip to the kiosk. The average time to visit a kiosk (one way) is 1 hour, for an average distance of 15 kilometers (IR survey). There was no significant cost associated with travel in the manual system, as the village accountants lived in the same hobli (cluster of villages).
- The average time spent in the queue is 27 minutes in the PAC report, and about 75 minutes in the IR survey.
- With the new system, 3% of users had to pay a bribe (PAC), as opposed to 66% in the manual system.
Second-Order Effects: A large proportion of the users of the Bhoomi system (40% in the PAC report, 45% in the IR survey) sought RTC (a record of Rights, Tenancy and Crop) certificates; since its inception, Bhoomi users have purchased about 700,000 RTC certificates a month from the kiosks state-wide. One reason for this is that the certificates are used to apply for loans from commercial and other lending IIMB Management Review, June 2005 15 institutions. These loans may be used for agricultural inputs, which in turn may have an impact on agricultural yields, incomes, consumption and loan repayments. With this model, two hypotheses about second order effects are: 1) since the inception of Bhoomi there is a significant increase in the rural, institutional lending in Karnataka; and 2) since the inception of Bhoomi there is a significant increase in (some) agricultural inputs in Karnataka. If these two hypotheses are accepted, then it will confirm that there has indeed been a secondorder impact of Bhoomi. Secondary data collected from published sources revealed the following:
- The data on advances (or loans) from commercial banks for the state of Karnataka shows a gradual increase in rural lending from 1992 till 2003. A regression line fitted to the data shows that in 2002 and 2003, post Bhoomi implementation years, the advances were more than the regressed line, but these increases are not significant (at 95% confidence level). In other words, Bhoomi’s impact on bank loans has not been significant, assuming all other factors that affect bank loans have remained the same.
- Data on per capita fertiliser consumption in Karnataka shows a sharp increase in 2003 after a decline from 2000 to 2002. Our background research indicates that this increase is not due to promotional campaigns by fertiliser companies but instead can be attributed to the increased loan availability. However, there are not enough data points to justify a strong conclusion.
Though the data is not conclusive, there are some positive second-order effects, particularly the sharp increase in fertiliser uptake in 2003. However, issuing of loans by rural banks is a highly bureaucratic and slow process requiring several documents – ‘no dues’ certificates from all banks in the taluk, encumbrance certificates for the last 13 years, and surety from a fellow farmer. Thus, despite the relatively easy availability of Bhoomi certificates, loan processing remains slow.
Conclusions
The results show that the theory of demand- and supply-side stakeholders enables a nuanced reading of e-government systems design and implementation success. It allows the designer to explicitly consider conflict in and resistance to the system, as also the impacts on the different stakeholder groups. The analysis of first- and second-order effects allows one to view the impact of the system in a larger and longer duration context than would otherwise be possible (especially from the ‘snapshot’ or ‘report card’ studies available in the literature).
Future work will consider the following issues: 1) The ideas and force of ‘technological determinism’ that over-determines the role and scope of technology for addressing governance needs; 2) The discourse of de-politicisation and reification that is used by the supply-side stakeholders to implement egovernment systems and their impact on the public; 3) An examination of the hypothesis that e-government systems reduce corruption.
Reprint No