The Land Acquisition Bill is a key piece of legislation under consideration in the Indian Parliament. This column argues that the current policy on compensating landowners, as proposed in the Bill, is misguided and could adversely affect the pace and character of future growth in India. It draws lessons from economic theory as well as the failed land acquisition experience in Singur to propose a workable model for determining appropriate compensation for land acquisition.
The Indian Parliament is currently considering a Land Acquisition, Rehabilitation and Resettlement Bill (LARR Bill), as are various state legislatures. These are likely to affect the pace and character of future growth in India in a significant way. Setting up new industries, service establishments and real estate development will inevitably require land to be acquired from rural areas. Hence legal rules and government policies for the process by which land is to be acquired, and how current owners are compensated, will play a critical role. The dramatic developments in Singur and Nandigram in West Bengal a few years ago drew national attention, and have influenced both the outcomes of the 2011 state elections in West Bengal, as well as framed the discussion regarding the Land Acquisition Bills currently being drafted.
The West Bengal fiasco has naturally generated calls for adequate compensation of those displaced. Reinforcing this are accounts of how various Maoist rebellions in other parts of India have been fuelled by the displacement of various tribal populations to make way for private mining concessions. The new Bills are proposing to modify the Eminent Domain Law of 1894 which has traditionally allowed the government to appropriate land from private owners upon paying compensation at market rates. There are three principal modifications. First, compensations will be set at between two to four times the market rate, depending on the nature and location of the land and the specific Bill in question. Second, eminent domain clauses will only be utilised by the government in cases where land is acquired by the public sector, not when land is acquired by private industry. Third, special measures for resettlement and rehabilitation of tribal communities are mandated.
Given the importance of the land acquisition policy to economic growth and development, we need answers to the following questions: What light does economic analysis throw on the problem of designing a land acquisition policy? What can be learnt from the West Bengal experience? To what extent is the newly evolving policy a step in the right direction? Do they have any serious deficiencies? If so what kinds of alternative policy approaches ought to be adopted?
The need for a stronger compensation policy
From the standpoint of economic theory, there are good reasons to insist on compensation of displaced landowners at above-market rates in cases where eminent domain clauses are applied. This is even if arguments of distributive justice or considerations of political sustainability are ignored, and the law is designed only with economic efficiency in mind. The strongest arguments follow from economic first principles – those who do not choose to sell land at the current market price, by definition, value their asset at more than the market price and therefore will naturally be unhappy if compensated at the market price. Also, if land markets are ‘thin’ – that is, have a low volume of transactions – and in general financial markets are not well functioning, then the current price of land is unlikely to reflect the future price of land before an industrial project is set up, and therefore, this creates an additional reason for owners to be resistant to be compensated at the current market price. There are more subtle arguments that we develop in one of our recent papers (Ghatak and Mookherjee 2011).
Compensation policy affects growth and economic efficiency in two significant ways. First, it affects the incentives of state governments to acquire land under eminent domain clauses by altering the cost of acquisition they incur. A stronger compensation requirement will slow down the pace of industrialisation by raising the cost of land acquisition. State governments are usually anxious to raise industrial growth rates for both political and economic reasons – government revenues, employment generation and perhaps the lure of illicit kickbacks from awarding industrial contracts and concessions. There is a natural inclination to overlook the costs imposed by such industrialisation on farmers who stand to lose their livelihoods. Mandating compensation at least at market rates ensures that state governments internalise these costs adequately.
A second set of effects concerns the investments made by farmers and governments in enhancing the agricultural productivity of rural land, which are likely to be larger with a stronger compensation policy. Stronger compensation mandates reduce the likelihood that any given piece of agricultural land will be acquired in the future as the cost of acquisition is higher. This enhances security of tenure of farmers, augmenting their incentives to invest in enhancing farm productivity – such as improving soil quality or developing local irrigation facilities. State governments would also be induced to improve local infrastructure such as supply of roads, water and electricity. There is considerable empirical evidence of the beneficial productivity effects of enhancing security of tenure in the context of Operation Barga in West Bengal in the 1980s (Banerjee, Gertler and Ghatak 2002, Bardhan and Mookherjee 2011). Owing to moral hazard problems - or problems that arise when a farm owner is unable to observe a tenant farmer’s effort – such investments tend to be under-supplied. Hence strengthening compensation policy has the side benefit of inducing additional agricultural investments.
The challenge of determining the value of land
While the argument for compensations to be set at least at market rates is clear on the grounds of both efficient and inclusive growth, the preceding arguments leave open the question of how much above market rates they ought to be set. There are good reasons to believe that many landowners value their land substantially in excess of the market rates. There are many reasons for this. Many view the income from land (or opportunity to consume crops grown) as a form of valuable security against various risks of high inflation or economic recession. The severe economic crises in Russia and Indonesia in the 1990s witnessed a move of a large section of the urban population back to rural areas for precisely this reason. Others may value land as it can serve as collateral for bank loans. It is also a secure form of holding wealth, and provides some insurance value, as well as old-age support to its owners. Farmers and tribal communities have developed special skills in farming, livestock rearing, fishing and hunting in their traditional habitat which are of no use in other occupations. For all these reasons – as well as others related to the role of land as a source of social status, prestige or ancestral identity – the value of land to its owner may well exceed the market value. Indeed, this is precisely the reason that many land owners have held on to their land instead of selling it in the market.
A recent survey of households whose lands were acquired in Singur provides evidence of the heterogeneity of land valuations to owners and the important role this played in opposition to the land acquisition (Ghatak et al 2012). The compensation offered by the West Bengal government was on average equal to the market values reported by the owners. Yet one third of these owners refused the compensation and opposed the land acquisition. This is partly explained by the inability of compensation offers to include information relevant to market values of individual plots, such as irrigation or multi-cropped status, or proximity to public transport facilities. But even after controlling for the extent to which the compensation deviated from market values, some personal characteristics of landowners were related to their decisions about whether to accept the offered compensation. Households for whom agriculture played a larger role in income, or those with a larger fraction of adult members who were workers, were less likely to accept the offer. This points to the role of income security as an important consideration, and the role of complementarity of land with farming skills that are non-transferable. Those who had a stronger financial interest (such as those who had purchased rather than inherited the plot or absentee landlords) were also less likely to accept.
How should appropriate compensation be determined?
Given such heterogeneity in land valuations, the same arguments articulated above require compensation rates to be set at the valuation placed on a piece of land by its owner, or the market value, whichever is higher. This is necessary to ensure that state governments adequately internalise the costs imposed on erstwhile owners when acquiring their lands. But this raises a quandary, for the personal valuations are known privately by owners. There is no documentary or factual method of ascertaining their personal valuations. Setting compensations at whatever personal valuation is claimed by an owner would then generate incentives for the owner to overstate his true valuation. This would choke off all prospects for industrial growth altogether.
Faced with this quandary, the various land acquisition bills have settled on arbitrary ratios for mandated compensations to market values. The problem is that getting this ratio correct is crucial. If it is set too high, the cost of acquiring land will become prohibitively large and industrialisation will slow down too much. If it is set too low, the problems seen in Singur will re-emerge. State politicians will be eager to avoid the West Bengal fiasco, and thus will tend to set the ratio too high to minimise the prospect of a political fallout similar to what the Left Front faced. No wonder that the LARR Bill in the Lok Sabha sets the ratio as high as four times! Combined with the withdrawal of eminent domain use for acquisition of land by private industry, the country faces the prospect of a protracted slowdown of private investment in industry and services.
An auction-based approach might be the solution
Is there a way out of this quandary? The economist’s preferred solution to elicit private asset valuations is to use auctions. This approach has been advocated recently by Maitreesh Ghatak and Parikshit Ghosh (Ghatak and Ghosh 2011). The Singur experience also illustrates some practical problems in paying compensations according to a formula based on market values. The true market value of a piece of land depends on many detailed characteristics such as type of soil, irrigation, elevation, and location that are very difficult for government land records to incorporate. The last cadastral land surveys in India were carried out prior to Independence. The considerable costs imposed on landowners on registering their plots and updates of plot characteristics means that government land records are hopelessly out-of-date. The survey of acquired households in Singur cited above found this to be a significant cause of undercompensation of owners of multi-cropped Sona lands relative to market values. These problems would be entirely avoided if compensations were based on bids submitted by owners in auctions.
The auction-based approach can be extended in various directions. The choice of location of a factory can also be decided by extending the auction to a multi-stage process. At the first stage, the industry in question or government could set a reserve price and minimum quantity of land needed. Next, different communities can be asked to bid for the factory to be located in their respective regions. These bids are set equal to the minimum price at which they can in turn procure the necessary amount of land from landowners within their areas (as elicited by a local auction).
The Indian government has considerable experience with conducting procurement auctions for dealing with private contractors on public projects. Hence the administrative expertise needed to conduct auctions for land acquisition is present in abundance. However, decentralising responsibility to local panchayat bodies in conducting these auctions within their jurisdictions will help minimise the sense of land acquisition being foisted on local communities by state or national governments in a top-down manner. In that case panchayat leaders would have to be trained (or assisted) by bureaucrats to conduct such auctions. But this would be help them acquire skills necessary for panchayats to take a more active role in business development within their respective areas.
Determining compensation policy: other considerations
Two other issues in compensation policy deserves some attention. Households in the cited Singur survey exhibited considerable preference in being compensated in alternative ways that incorporate their concern for financial security, time preference, and pattern of skills. These concerns exhibited considerable diversity, with a corresponding diversity of preferences over alternative forms of non-cash compensation. Hence a menu of alternative compensation packages ought to be offered, to cater to this diversity. Creating a more well-informed and flexible way of compensating displaced landowners can go a long way in ensuring fast industrial growth along with an equitable distribution of its benefits.
Finally, process matters. Consistent with the legal framework for acquisition inherited from colonial times, the process used in Singur was very top down. Local residents of the area repeatedly mentioned their sense of outrage at this. The state government did not consult the local community in choosing the area for the Tata factory. Only after protests snowballed did it offer to negotiate the compensations offered. The politicization of the compensation process followed. Clearly there is much room for a more consultative process, in which local communities are consulted and involved in selecting areas to be acquired, and in the design and implementation of compensations. Creating a quasi-judicial regulatory process to oversee the process would also help reduce the politicization of the process.
- M. Ghatak and D. Mookherjee, "Land Acquisition for Industrialization and Compensation for Displaced Farmers,” Working Paper, October 2011, Institute for Economic Development, Boston University. http://people.bu.edu/dilipm/wkpap/land%20acquisitionJDEmscrpt.pdf
- A. Banerjee, P. Gertler and M. Ghatak, `Empowerment and Efficiency: Tenancy Reform in West Bengal,´ 2002, Journal of Political Economy, 110(2), 239-280.
- P. Bardhan and D. Mookherjee, `Subsidized Farm Input Programs and Agricultural Performance: A Farm-Level Analysis of West Bengal’s Green Revolution,” American Economic Journal: Applied Economics, October 2011, 3, 186-214.
- M. Ghatak, S. Mitra, D. Mookherjee and A. Nath, ` Land Acquisition and Compensation in Singur: What Really Happened? Working Paper, March 2012, Institute for Economic Development, Boston University.
- M. Ghatak and P. Ghosh, “The Land Acquisition Bill: A Critique and a Proposal,” Economic and Political Weekly, October 8 2011, Vol XLVI (41), 65-72.