An international report on land administration in India states, The records in the sub registrars' office are not preserved properly. Only tobacco leaves and naphthalene balls are used to protect the documents from insects and pests. It says this of land records in Delhi, The system of maintenance of land records in Delhi is a classical (sic) example of ambiguity, confusion, chaos.

Over the past few months, land issues are surfacing everywhere: the brouhaha over relocating industries out of Delhi, the re-use of mill land in Mumbai, the land needs of knowledge industries in our newer cities. This spotlight is a blessing. Because if it stays, it could help address one of the most chronic impediments to India's growth: our incomplete land reforms processes, and our inadequate land record management systems.

Land is murky territory in India, often associated with the mafia, cash transactions, and manipulation of documents. We are invariably reduced to being 'price-takers' in this process, taking care of our specific interests, holding our nose as we do this. Unfortunately, land is an almost ubiquitous element in most issues that affect us, across rural and urban areas. These issues broadly fall into two categories:

1. Policy related: A key one is land reforms for re-distribution, the focus of India's land-related activities since Independence. However, there are other policy areas related to land: protecting tribal land, preserving environmental resources, gender-based ownership, urban planning etc.

2. Land records management: the maintenance and updating of accurate and detailed land records under various laws.

While sound policies are critical, they cannot be implemented without accurate land records. Besides, poor land records management also results in enormous corruption and partronage, and inefficient delivery systems, where even the simplest projects get delayed, resulting in cost- and time-overruns.

Land records management in India is linked to the land revenues that the British extracted. Different systems were established, from the zamindari system in the north, to the ryotwari system in the south. The quality of the land records maintained in these areas in many ways mirrored the needs of the revenue system: the more the revenue formula reflected specific land parcel details, the greater the efficiency in land records management.

Unfortunately, this relationship between revenues and records became somewhat tenuous after Independence. With the sweeping land reforms movement, and relaxation of land revenue obligations, the share of land revenue in state taxes dropped steadily. Between 1957 and 1990, land revenues dropped from over 20% of states' own taxes to barely 2%. This diminishing dependence on land revenues was accompanied by a degradation in land records management.

Over the years, several studies have documented the gaps in India's land records management: incomplete settlement and survey plans, rudimentary valuation systems, weak administrative skills, too many laws and departments, lack of consultation with affected people, no guaranteed land title system, and so on. Many state governments have periodically announced various initiatives to 'fix' the land records problem, but there has been little real change on the ground. Unfortunately, reform doesn't come in a predictable, linear fashion. None of these complex reforms will happen because they need to occur; they need a catalyst, a discontinuous shift that can trigger change. The challenge is to find the catalyt, the driving force of change.

The good news is that there is now a candidate. Just as land revenues drove the first wave of efficient land records, an emerging land-based revenue could possibly drive the second generation of land records management. This candidate is 'Registration Charges', the fees that a state collects on the sale/purchase of land. Consider the statistics for Karnataka: between 1999 and 2004, registration charges tripled from Rs 550 crores to Rs 1600 crores. This year, Karnataka expects to garner Rs 2,100 crores, making it the most buoyant revenue item, quietly growing from 8.6% of state's own revenues to 12% in four years. Soon, registration charges will be equal to excise duties.

This trend is not true only for Karnataka: as the table shows, total registration fees for the states profiled grew 71% in four years. Driven by economic forces, this is occuring despite the current dysfunctional land records process. Beneath our flyovers and factories, gleaming offices and apartment complexes is quicksand, a land management system that cannot match this torrid growth. The cracks are already showing, but we are all ignoring them.

In the seams of conflict can be sown the seeds of change. Clearly, a new force has arrived. This force can be harnessed to establish a new logic for land records management, just as the previous era did. Harnessing this force requires focus: identification of a few critical ingredients for superior land records management, that can trigger larger change. A few suggestions:

• Lower registration fees to incentivise more officially-recorded transactions. Current levels in many states are around 8% - 10%, sometimes higher with surcharges and duties. Like most taxes, lower rates could result in greater compliance and increased revenues.

• Establish highly efficient land data-banks linking multiple players, much like the National Security Depository Ltd (NSDL) did for share registries. Possibly even get them to do this.

• Move towards a system of guaranteed land title, much like most of the developed world.

These changes could create liquid markets, improve transaction efficiency, minimise legal disputes, release land as an asset for credit, and form the platform for larger land reforms.

Taking on such a gigantic task across the country would be virtually impossible. Fortunately, land is a state subject, which allows for a marketplace of changes, driven by a variety of agents. Irrespective of which state moves - or is made to move - towards land records reform, a critical mass of stakeholders outside government needs to seek change relentlessly.

Critics may argue that land records are only one aspect of a larger tapestry of land reforms that need to be woven. This is true, but the question is, "How will we implement policies without a well-functioning land records system?" If we get this platform in place, we can enable all those interested in reform policy - be it poverty alleviation or environmental protection, tribal rights or urban planning - with the tools to ensure that their policy dreams get translated into ground realities.

Otherwise, these dreams will be like our land records: crumbling away, mothballed and tobacco-leaved.