eGovernment for Development

eTransparency Case Study No.11

Gyandoot: Trying to Improve Government Services for Rural Citizens in India

Case Study Authors

Alok Kumar Sanjay (alok@actionaidindia.org) and Vivek Gupta (vivekgupta@icfaipress.org)

Application

In 2000, the State Government in Madhya Pradesh, India, set up a chain of computer kiosks to help provide better access to government information and services in one of its districts - Dhar District.

Application Description

20 kiosks ("soochanalayas") were initially set up in various rural centres, with each kiosk typically serving a population of 20,000-30,000 villagers. A further 18 kiosks were added later. Each kiosk was run by a trained operator, and can provide a range of services for a nominal service charge, ranging from around US$0.10 for 'ask the expert' through US$0.20 for registrations to US$0.50 for use of the matrimonial site (many villagers subsist on less than US$2 per day). eTransparency-related information and services provided include:

• Land records: farmers need an official map and land record of their land ownership if they are to apply for loans. It was intended that they could apply for these via the kiosk, and then come later to collect the official documents.
• Below poverty line and other lists: government provides special welfare schemes for those families in a village deemed to be living below the poverty line. This list - previously available only to governing bodies in the village - is now available online, so that families can check directly to see whether or not they are listed. Lists are also available relating to public distribution of food, and pension and rural development scheme beneficiaries. There are also details of the grants given by government for use by village committees.
• Grievance registration: if a citizen has a grievance to raise against a public servant, they previously had to travel to district headquarters. Now registration and progress reporting could all be done remotely, through the kiosk.
• Market prices: related to e-transparency in markets rather than government, this is the facility by which prices for agricultural goods from the district wholesale markets are made available to farmers via the kiosk. Having such prices would enable them to wait at home until the right price is reached, rather than having to rely on intermediaries (or make a special journey to market, which might be up to 25 kilometres distant, needing 4-5 hours' travel time with a typical journey cost of around US$0.50 plus loss of income).

Other services set up include: rural email facility; a village auction site; a matrimonial site; an "ask the wiseman" service for children; an "ask the expert" service for farmers; a village newspaper; an e-education site; and employment news (aimed at semi-skilled workers). The kiosks can also be used free of charge by local government officials, e.g. for email or to exchange health/education data with district headquarters.

Role of ICT

Gyandoot's central hardware based at Dhar District government headquarters consisted of a server with a 450 MHz Pentium III processor, 128 MB RAM, 40 GB disk drive, 2 MB graphics card, 15" monitor, and 48x CD-ROM. Kiosk client PCs had a 433 MHz Celeron processor, 32 MB RAM, 4.3 GB disk drive plus floppy, 4 MB graphics card, 14" monitor, and 48x CD-ROM, though specifications for later kiosks were upgraded slightly. Telecommunications was effected using 56kbps modems - one each on the client side, five on the server side - connected as an intranet using a mix of dial-up lines and wireless in local loop connections. Each kiosk had a dot-matrix printer and a five-hour back-up UPS. The server system runs on Windows NT with IIS server; client PCs run Windows 98. MS SQL Server, Visual Basic, Java Development Kit and MS Access were used to develop the applications, plus an Indian language font set. The system uses Internet Explorer as the main front-end.

Whenever information is required, the kiosk operator ("soochak") should dial through from the kiosk to the server at district headquarters. All the software applications are menu-driven and Web browser-based. The operator uses them on behalf of the users who visit the kiosk.

Application Drivers/Purpose

Madhya Pradesh state government was a leader in the promotion of local self-governance (known in India as "panchayati raj") at the village level. Gyandoot was therefore seen as a means to support this philosophy, making information and services more available and more transparent. This was also seen as a means of reaching out to a large potential vote-bank that had, to date, seen no tangible benefits from the "ICT revolution". A spin-off benefit was seen as the creation of new jobs/small enterprises in the form of the Gyandoot kiosk operations. It should be recognised, though, that a major driving force was the chief government official in Dhar District - the District Collector - who was very keen on ICTs and who linked his personal reputation to the scheme.

The formally-stated objectives of the project were:

• Ensuring empowerment of the downtrodden segments of the society and to encourage consensus building and enhance transparency.
• Creating a cost-effective, replicable, economically self-reliant and financially-viable model for taking the benefits of information technology to the rural masses.
• Improving the quality, speed and sensitivity of the state delivery apparatus towards the needs of the local citizen.
• Enhanced participation in community affairs and governance through creative use of information technology.
• Implementing a new grass-root entrepreneurial model with participation of groups of non-traditional entrepreneurs.

Stakeholders

Government officials in Dhar District (especially the Collector) are major stakeholders, as are all the villagers in the District, and the kiosk operators. Other stakeholders include senior officials in the Madhya Pradesh government; the IT vendors, designers and implementers; and those intermediaries who profit by interceding between villagers and government, or between villagers and wholesale markets.

Transparency and the Poor

Gyandoot is one of relatively few e-transparency projects to have made a specific effort at trying to assist disadvantaged groups. This is seen in the choice of location (Dhar is a relatively poor, relatively rural district); in the identity of users targetted by information and service design; and in the identity of the kiosk operators (who were selected from disadvantaged groups). In part, some of its identified problems derive from - and can arguably be offset against - its attempt to reach out to the poor.

Impact: Costs and Benefits

Costs for the initial system were around US$50,000, paid by the local/village government units. This does not include the set-up costs of the kiosks since the first 20 soochaks funded these costs through bank loans; nor does it include the operational costs which soochaks are supposed to meet from the fees they charge. They should also pay 10% of their income to the local government unit. The 18 later kiosks were set up as private enterprises, paying a US$100 annual licence fee to the district government.

Overall, the Gyandoot system has not really worked as intended. It has delivered some infrequent transparency benefits. In the past, citizens often had to pay bribes in order to have public services performed; some groups (from lower castes, or the disabled) faced discrimination and the humiliation of being ignored or their applications rejected; and most faced the prospect of losing a day's wages plus paying transport costs each time they were forced to visit district headquarters. The Gyandoot system has - in a few places, at a few times, for a limited range of services - replaced this with a lower-cost, faster, and more transparent service. Corrupt "under-the-counter" payments to public servants have, occasionally, been replaced by transparent payments to private sector kiosk operators. Transport costs, wage losses and discrimination have been reduced on some occasions - a particular benefit for the poor. The public availability of welfare scheme lists has brought sporadic transparency benefits, enabling a few families to correct their omission from the lists.

A majority of Gyandoot's few users seem positive about it, but - when asked - they see by far the main impact of the project as its improvement of their understanding of ICTs. There is a perception among many users that Gyandoot has reduced corruption, and improved transparency. Yet, at the same time, those users continue to see a need for bribes to be paid.

In part this relates to the fact that some services have not been deliverable as initially intended. For example, it was intended that land records would be deliverable via the kiosk without needing to involve officials. However, since such records lacked an authorised signature, they were not accepted as an official document by banks and other agencies. The system has thus been adjusted. In some cases applicants merely apply via the kiosk and are then emailed with a date when they can travel to a government office to pick up their land record. In other cases, the record can be printed at the kiosk, but must then be signed by a government official. This may bring some benefits of reduced travel, but the intended disintermediation has not occurred. Indeed, there are complaints that adding the kiosk into the equation has merely added to the complexity and cost of the process.

A similar result has prevailed with public grievances - some are dealt with online, but in other cases the applicant must appear in person, in which case Gyandoot just functions as a system of complaint registration and appointment-making. There could be some improvements since speed of response to document requests, grievances, etc. is monitored by the government's management information system. In practice, though, 90% of those submitting grievances via Gyandoot felt their complaint had not been satisfactorily resolved.

There has also been a growth in problems associated with Gyandoot over time, limiting even potential benefits still further. A number of kiosks lie idle for significant periods of time: around one-third appear to be permanently closed; many others are closed for hours or days at a time. Lack of electricity is one cause, as are ongoing telecommunications problems. Limited income and, related, the commitment of kiosk operators is another cause of closure. When kiosks are open, service response times are often poor because hardware capacity is limited, and because several kiosks may attempt to access services at the same time when the electricity comes on.

In some communities, particularly those not close to district headquarters, this has led to a message of unreliability spreading about Gyandoot, further undermining use and sustainability of the service. This vicious circle is strengthened by the fact that some information on the system is not regularly updated. One quarter of users reported sustaining losses in their produce sales because they had based those sales on prices accessed via Gyandoot that turned out to be out-of-date. As a result, usage levels of the kiosks are incredibly low - typical usage levels are just one Gyandoot user every two-three days.

As a result of all these problems, some of the kiosk operators have lost confidence, since they depend on Gyandoot's proper functioning to earn their living. They are concerned about their lack of income and concerned that communities now suspect them of deliberately holding back information or services in order to try to extract bribes. As one soochak commented:

"First of all Gyandoot should try to fulfill the promised services to the people effectively before going in for further additions, as due to the ineffective functioning of the present facilities and utilities the credibility of the project as well as ours [ the soochaks ] are at stake. People feel that we [ soochaks ] deliberately do not avail information in the right time, to earn extra from the community".

This reflects the way in which - when Gyandoot functions - the soochak partly replaces village or district officials as a key gatekeeper in the lives of villagers. This in itself, though, has led to problems. Some villagers have been unwilling to use kiosks where the operator was a woman; in two other cases, the soochak has barred lower-caste villagers from using Gyandoot.

Evaluation: Failure or Success?

Gyandoot was awarded the Stockholm Challenge award for 2000 in the Public Service and Democracy category, and a Computer Society of India/Tata Consultancy Services National IT award for best IT usage in 2000.

These awards should be seen as justifiable recognition for design innovation, relevance and potential. However, longer-term evaluation shows that the project has fallen well short on all of its - admittedly over-ambitious - objectives. It should be described as largely unsuccessful. It has identified a model by which transparency of government information and services could be increased, and corruption, discrimination and other costs reduced. It has also made very occasional progress towards these ends. However, it faces a number of practical difficulties that have constrained the benefits delivered to date.

The details provided in this case are drawn particularly from an evaluation undertaken in 2002 by a team from ActionAid India, which visited the Gyandoot main centre plus three kiosks in outlying villages. Evidence has also been drawn from an even more comprehensive evaluation conducted in 2002 by a team from the Centre for Electronic Government, IIM-Ahmedabad. One main contrast the ActionAid India team found was between views about the project expressed by state officials, and those expressed by villagers.

Enablers/Critical Success Factors

1. Top-level champion . This e-transparency project was championed by the senior official in Dhar District, the District Collector. It therefore received the political, financial and other resources necessary for successful initial implementation.
2. Cost sharing . The investment required by government in this e-transparency project was limited to the US$50,000 invested by local government units. State and central government paid nothing (except the contribution of system design/development by the National Informatics Centre, a central government body). The kiosks were funded by bank loans. Not only did this make the project relatively cheap to finance, it also helped ensure at least the initial ownership and commitment of key stakeholders.
3. Meeting real citizen needs . Even though the system has not worked as intended, Gyandoot has always directly acted to try to save money and increase wages for villagers.

Constraints/Challenges

1. Infrastructure . This is a significant constraint in some parts of the district, particularly those more then 5km from district headquarters. The UPS batteries have a charge-up time between four and eight hours. But in some villages, there is only an hour or two's electricity per day, making Gyandoot operation only sporadic. There have also been connectivity problems with the intranet in some centres, combined with the problem of poor response times. The kiosk can easily get into a spiral of declining use and declining income that undermines its sustainability.
2. Rotation of senior staff . An enthusiastic senior team is a critical success factor for e-transparency projects. However, in government service, staff are often rotated to another job after just a couple of years. This brings in a new team who will have far less ownership of, and thus perhaps less enthusiasm for, the projects started by their predecessors. When the District Collector and other key staff moved on, new incumbents proved less committed to a project they themselves had not initiated.
3. Process delays . The Gyandoot system acts largely as a communication vehicle - it does not automate the core processes of creating documents, dealing with grievances, etc. These processes remain subject to the usual bureaucratic delays, corruption, and inconsistencies. This has also affected the updating of information for the Gyandoot system.
4. 4. Financial sustainability . The average Gyandoot-generated income for the kiosk operators is around US$35 per year. Their calculated break-even point is nearer to US$100 per year. As a result, the soochaks have been driven to attempt other income-generating ideas. In some cases, this has involved closure of their kiosks as they seek their fortunes in other ways.

Recommendations

1. Involve communities . The Gyandoot project was designed and implemented in something of a top-down manner. Thus, although developed with good intentions, the Gyandoot system did not always provide relevant services for villagers and has not always provided services in a relevant way. This could have been avoided with better involvement of the local NGOs and community-based organisations that represent the villagers. Not only would this have assisted design, it would also have helped to raise awareness about Gyandoot. In particular, failure to view matters from a bottom-up perspective has meant that core service processes remain untouched; yet it these processes - getting their land records signed; getting their grievances resolved - that most concern villagers. The same issue arises in connection with evaluation. Some e-transparency evaluations (including those used as the basis for some awards) rely on self-reporting by state officials who initiated and implemented the projects. Too few evaluations follow the approach used here of interacting with the citizen-users. Yet it is obvious that they should be the main focus of any evaluation: "If you want to know the quality of dog biscuits, you need to ask the dog".
2. Focus on sustainability . For e-transparency projects that reach out into poor communities, there must be a clear model for financial sustainability. This could justifiably include ongoing subsidies rather than the self-sufficient model utilised to date by Gyandoot. Alternatively, it could look to make use of cheaper technologies like radio or mobile telephony.

Further Information

Additional material for this case study was derived from:
http://www1.worldbank.org/publicsector/bnpp/Gyandoot.PDF

http://gyandoot.nic.in/index.html , http://icmr.icfai.org/catalogue/IT%20&%20Systems/ITSY022.htm

Case Details

Case Editor : Richard Heeks.
Author Data Sources/Role : Interviews, Observation and Documents; No Direct Role.
Centrality of Transparency : Mixed. Type : Transaction. Audience : External. Content : Mixed. Sector : Multiple. Outcome : Largely Unsuccessful.
Region : South Asia. Start Date : 2000. Submission Date : November 2003.