India's e-government project

Land registration offices throughout Andhra Pradesh now operate computerized counters to help citizens to complete registration requirements within an hour instead of several days, as was necessary under the earlier system. The lack of transparency in property valuation under the old system resulted in a flourishing business of brokers and middlemen leading to corruption. Antiquated procedures such as manual copying and indexing of documents, and storage in paper forms in ill-maintained backrooms have all been replaced. This case illustrates some of the key implementation issues faced by state and national governments in their efforts to use IT to improve citizen-government interfaces.

Application Context                                   

Registration to document changes in ownership and transactions involving immovable property is governed by the Indian Stamp Act of 1899. Deeds of various kinds are required by law to be written on stamp paper of prescribed value. Certain transactions require a fixed duty. For others the ad valorem method is used, whereby the stamp duty is a percentage of the property value or loan that is the subject of the instrument. The ad valorem method ensures that inflation will not erode the value of stamp revenues. This method accounts for over 90% of the total revenue from stamp duty.

Registration is carried out at the office of the Sub-Registrar of Assurances. In Andhra Pradesh (AP) there are 387 Sub-Registrar offices that register
approximately 1.2 million documents per year. The work of the Sub-Registrar is supervised by a hierarchy of District Registrars (28), Deputy Inspectors (6), and the Inspector General (1). The traditional 11-step registration procedure is complex and time consuming, beyond the comprehension of most citizens.

step 1
The value of the property is determined.

step 2
Stamp duty, transfer duty, registration fee and other fees are calculated

step 3
Citizen must purchase stamp paper.

step 4
The legal registration document and certificates to be enclosed with the document must be prepared.

step 5
These documents are presented to the Sub-Registrar of the jurisdiction.

step 6
The Sub-Registrar scrutinizes the documents, reviewing the valuation of the property, calculation of stamp duty, transfer duty, registration fees and miscellaneous fees.

step 7
Payment of deficit stamp duty, if any, is required.

step 8
Final document certified by the citizen before the Sub-Registrar and two witnesses.

step 9
The document is copied into the register books.

step 10
Copies are posted to 2 indexes (by name and property), and accounts and reports.

step 11
The document is returned to the citizen.

Below is a brief account of various actors involved in the conventional registration process.

Stamp Vendors
Stamps are sold to the public through private stamp vendors (licensed by the Registration and Stamps Department) and at stamp counters located at the offices of the Sub-Registrars. The private stamp vendors have commonly charged an illegal premium on the face value of the stamps when there is scarcity of stamps of a particular denomination. They have also resorted to the sale of fake stamps and post-dated stamps for an additional charge. There are about 2,300 licensed stamp vendors and 221 departmental stamp counters in AP.

Document Writers
The document writers have been given official recognition in several states of India through a system of licensing (there are 3,908 licensed writers in AP). In AP, when a document is not written by a licensed document writer an additional fee (approximately $5 or Rs.215) is levied at the time of registration. Document writers prepare the maps and location sketches to describe the property, fill in various forms, and assist citizens in procuring certificates from various authorities. For their comprehensive services they demand a fee higher than that prescribed by law.

Registration Agents
These are self-employed individuals and firms who, for a lump sum payment, get a document registered, covering the whole range of services.
This manual registration system generated a number of important drawbacks. Most importantly:
Lack of transparency in valuation: Since the stamp duty is linked to property values, valuation procedures are vital. A system of market value guidelines was introduced in 1975, whereby the rate per unit of rural/urban lands is assessed for all villages/towns and incorporated in a register for public guidance. However, the basic value registers usually are not accessible to the public, and even if they were, it is difficult for a common citizen to read them and calculate the amount of stamp duty, transfer duty, registration fee and miscellaneous fee. All this creates an impression that the valuation of property is "flexible" and "negotiable," prompting a host of corrupt practices and a flourishing business of brokers and middlemen who exploit the confusion surrounding the registration process.
Tedious back office functions: Conventional manual methods of copying, indexing and retrieving documents are laborious, time consuming, and prone to errors and manipulations. Thus, a premium is often paid for speedy delivery of services.
Difficulties in preserving documents: The registers occupy a lot of physical space, usually in ill-maintained backrooms. They also deteriorate with age and repeated handling.

A New Approach                                                 

The Computer aided Administration of Registration Department (CARD) is designed to eliminate the maladies affecting the conventional registration system by introducing electronic delivery of all registration services. CARD was initiated to meet the following key objectives:

demystify the registration process

bring speed, efficiency, consistency and reliability

substantially improve the citizen interface

These goals were to be achieved by:
introducing a transparent system of valuation of properties, easily accessible to citizens
replacing the manual system of copying and filing of documents with a sophisticated document management system using imaging technology
replacing the manual system of indexing, accounting and reporting through the introduction of electronic document writing
Since 60% of the documents, Encumbrance Certificates (ECs) and certified copies relate to agricultural properties, the success of the CARD project would greatly benefit the rural farming community. Agriculturists would also benefit from a possible link-up of the CARD network with the rural bank network, which would enhance the efficiencies of the rural credit services by eliminating the need for paper-based procedures.

Implementation Challenges                                                
Implementation of an IT project involving over 200 locations state-wide was a formidable challenge. The project was divided into 9 major tasks and 64 sub-tasks. Approximately 2,000 hardware items and software packages were procured within a span of about five months through the agency of AP Technology Services. The project had to be implemented rapidly so that the technology (both hardware and software) would not become obsolete prior to the project launch. Implementation required considerable re-engineering. First, the national Registration Act of 1908 did not contemplate the use of computers to handle registration procedures. The Registration Act therefore had to be amended, a process that took over a year. The Act, in its application to the state of AP, has been amended to provide for the following:
Document registration and copying may be completed with the aid of electronic devices like computers, scanners and CDs; and copies may be preserved and retrieved with the same tools.
Copies of documents registered and stored electronically, retrieved, printed and certified by the sub-registrar shall be received as legal documents.
The registration software shall be prescribed by the Inspector General.
Second, to use these new technologies effectively, a large and well-designed training program was carried out by a private sector company at a cost of $262,000 (9% of the project cost). A training programme of 1-3 weeks was organized for different categories of officers. Seventy-five Data Processing Officers (DPOs) were trained for six months and 1,200 data entry operators were provided 2 weeks of training. Extensive system reforms cannot be brought about without adequate motivation within the organization. The following decisions were taken to motivate employees:
A cross-section of the field personnel was closely associated with the design and development of the software, and especially in the task of business process re-engineering.
No external technical personnel were recruited.
The head of the department undertook extensive tours throughout the state and conducted workshops, presentations, and special training camps involving all departmental employees. The officials who managed the two pilot sites were closely associated with this effort.
Senior functionaries of the government such as the Principal Secretary and Minister of the Revenue Department were closely associated with, and supportive of, the project.
A third implementation challenge was the tremendous data backlog. The CARD masters (state level) could be built without much difficulty, as the data is both limited and readily available. However, the project encountered major challenges in building up basic value data and the EC data for the last 15 years. The basic value data consisted of about 50,000 records at each Sub-Registrar Office (SRO). These data were entered into the systems by the trained staff in 6 to 8 weeks. The task of entering EC data, which has a more complex size and structure - about 1.2 million records of 2 KB size each - was out-sourced to five agencies in March 1998. Fourth, installation of CARD application software in 212 locations was considered a major challenge. Seven versions of the software had to be developed, tested, and deployed in a period of 4 months to achieve the desired functionality across the counter. This task was made possible by the relentless efforts of the DPOs who were groomed in preparation for this task. One significant strategy adopted to 'de-bottleneck' this process was to enable the DPOs to contact the head of the department and a core of technical personnel at the headquarters at any time to solve problems encountered in installation. The CARD project was launched on 4 November 1998. Political figures from each region inaugurated the new centers on the same day, thus helping to solidify political support for the project. An appropriate media campaign was also undertaken to educate the public and thereby bring about the elimination of middlemen and brokers in the registration process.

Benefits and Costs                                                             
Six months following the launch of the CARD project, about 80% of all land registration transactions in AP were carried out electronically. Some transactions are still being handled manually at a few locations due to hardware and software related problems. The time required for services such as valuation of property, sale of stamp paper, and provision of certified copies of registered documents now takes 10 minutes instead of a few days as under the earlier system. ECs are now issued to citizens in a span of 5 minutes, using a system that searches through more than 15 years of records from over 50 offices. Land registration can be completed in a few hours, whereas earlier it took 7-15 days.After factoring out the natural upward trend in nominal revenues, the CARD system has generated a modest increase in revenue. AP Land Registration Revenue, 1995-2000       Year       
    

  Target      
Gross Revenue
Net Revenue
% Growth (in net rev.)1995-96
570.00
460.27
325.62
3.831996-97
570.00
605.87
435.16
33.641997-98
639.00
635.91
449.61
3.321998-99
675.00
710.03
493.51
9.761999-00
778.00
786.00
591.00
19.75

Still another benefit of the CARD program is that it has prompted the public to pressure government for similar changes in other areas.

Concerning the CARD program itself, the following improvements are contemplated: introduction of a Telegu version of the software creation of a CARD service centre to provide all registration-related services under one roof (except registration of deeds relating to any property in the twin cities) networking all the servers/PCs at 214 centres using the AP State Wide Area Network (APSWAN) so that all registration services (except registration of deeds) can be accessed at any of the 214 offices irrespective of location of property provide registration information services on the Internet development of a property title database, which would be the precursor for introducing the Torrens System of registration (whereby registration of a sale deed guarantees title to a property) linking the databases of all land-related departments, such as land revenue, municipal administration, and irrigation linking the EC database with the banking network to facilitate speedier processing of applications for rural credit The cost of the CARD project was funded entirely by the AP government. The 1996 pilot project to computerise two Sub-Registrar Offices cost about $55,000. The original outlay for the full CARD project was about US$3 million; and this figure is likely to grow to $4.3 million. (This cost includes hardware, software, training, site preparation, data entry, air-conditioners, furniture, stationery and storage media, and other miscellaneous expenses.) Although employees were not fired, Post-Spot Inspections reveal that the new system has not found favor with employees due to a loss in enforcement power. If an under-valuation is discovered, a separate notice must be issued to collect it, and this collection may go into litigation. There is also a need to encourage payment by check or electronic means instead of the current practice of accepting cash.

Key Lessons                                                                         

Government should have a clear, coherent rationale and plan for choosing a particular E-Government application.
The AP government prioritised and selected for attention a service that generates high tax revenues, has a large citizen interface, and some prior involvement with IT. The land deeds registration service is one such area.

Is there a clear motive and understanding of the benefits emerging from using IT?
Often the value-added of technological re-engineering of services is not clearly understood or targeted at the outset, and the outcomes are therefore disappointing. In this application IT solutions were used for the specific goal of reducing the time it took for citizens to register their deeds. Reducing corruption was never the stated goal of the CARD project, nor has it been eliminated to a significant degree. Any government that sets out to eliminate corruption as an explicit objective is likely to encounter greater resistance from the employees who stand to loose. Anticorruption is a fringe benefit of the reform, achieved here mainly through the elimination of intermediaries.
Effective change management is essential.
Of all the factors that contributed to the success of CARD, this clearly emerges as the most important. In fact, when asked about the manner in which he had to distribute his time and effort, the manager of the CARD project attributed 45% to change management, 35% to the re-engineering of processes, a mere 15-20% to software, and 5% to other factors. To circumvent predictable and formidable opposition from the intermediaries who stood to lose from these changes, the project did not confront them directly, but chose instead to co-exist with the old system, allowing the market to eliminate gradually the demand for these intermediaries. Care was also taken not to antagonise the lower rungs of bureaucracy. The government announced at the outset that no downsizing would result from the introduction of this technology, and is now trying to transfer excess capacity into previously neglected activities (e.g., fieldwork). The AP Government has decided that such projects in the future will be led only by public administrators who have been trained to understand technology, rather than by technical specialists trained to manage. In a huge and costly demonstration of commitment to this ideal, the government has set aside funds to interview, select, and train promising public administrators for future projects. This experience underscores that e-governance projects can perhaps be managed best by public servants.
Appropriate physical telecom infrastructure is absolutely necessary for the application of IT solutions, but an insufficient condition to achieve successful E-Government reforms.
The Andhra Pradesh Technology Service, a different kind of "infrastrutural" element, was instrumental in the progress of Andhra's overall IT and e-governance agenda. This government-owned company functions as a sort of in-house consulting group for government projects. It is not bound by the government's pay structure and is capable of attracting and retaining specialists from the private sector as well.
The choice of software and technology is often secondary to other factors such as the proper change management.
Certainly, the choice of software was important. However, the CARD experience suggests that its importance for success was secondary. Wisely, this e-government application was designed to be flexible and scalable to accommodate new services, statutory changes in registration procedure and new computing environments.
Making a successful transition from a manual to an electronic process demanded changes to a number of established work procedures.
Process re-engineering was needed to realise the promised benefits and deal with the new challenges of the new medium. Related elements, such as legislation, also had to be updated.

Why has tax revenue not increased more significantly since the CARD was implemented?
The department has no way to measure acurately the revenue loss in the existing system. Surrogate indicators such as the number of court cases protesting valuations can be used to suggest whether valuation is lenient or strict. Changing the basis of valuation from the current reliance on historical records (which reflect depressed prices to evade stamp duty) would increase revenue. However, while using historical prices does not optimize revenue collection, it is considered relatively transparent. A system based on market intelligence could use current prices, but would be considered arbitrary unless very well specified. Such specification would require building a GIS database of all properties and collecting market prices of new buildings, with an explicit depreciation rule, etc. A system of that kind will require a large one-time effort, as well as continuous monitoring. One reason why the implementation of CARD succeeded is because it skirted the contentious issue of pricing.

What of the prospects for charging citizens for transformed services?
Every e-government project will require new investments. If manpower cannot be reduced, then operational costs are likely to increase, as well. In the long run the CARD application may generate more tax revenues. However, in the short term it has been a net fiscal loss for the government. The government needs to consider charging a transaction fee to offset the costs. This is likely to be met with resistance. But examples elsewhere have shown that even very poor rural citizens are often willing to pay reasonable fees for legitimate and useful improvements in services.