A day before the end of the monsoon session of Parliament, the government tabled its long awaited Land Acquisition Bill to replace the archaic Land Acquisition Act of 1894. But it struggles to define public purpose, shield multi-cropped and irrigated land and impose rehabilitation and resettlement obligations on both governments and private companies. It leaves the kind of loopholes through which a law's intent often escapes into a world ruled by money and power. On the central question - If the State must seize private property for social need, at what price should such transfer take place? - the bill displays little evidence of economic logic or fresh thinking.
The 1894 Act states it must be at the historical market price for land in the region, as recorded in recent sale deeds. The new bill just multiplies this by four for rural areas and two for urban areas. Where did these numbers come from? Why not three or five instead?
What is wrong with paying compensation in accordance with the market price? Well, plenty. Land transactions in rural India are infrequent, records are spotty, a big part is paid in black money and distress sales occur too often. But the main reason is that the price from voluntary transactions will always underestimate how much owners value their assets in the case of forced acquisitions. Owners are owners for a reason; they find the market price too low to sell.
Compensation at market prices could still work because land, unlike time, can be bought back, in principle. The problem is that rural land markets in India are highly imperfect. Lack of coordination, information and financing - functions usually carried out by organised brokers - means many potentially useful trades go unrealised. Moreover, the setting up of an industry or a housing estate in some area will usually send land prices through the roof, which means compensation at historical market rates will not be enough to buy land in the future.
Meanwhile, landowners vary a lot in how much they value land and at what price they are truly willing to part with it. Marginal farmers depend on it not only to grow crops but also take loans against it as collateral, employ family labour, achieve food security and maintain some social status. An absentee landlord has few of these benefits, and will probably be happy to get rid of the land at a relatively lower price. A good arrangement is one which compensates landowners according to how much they value it, ensure that land is taken from those who value it least, stand in for missing land markets and minimise coercion as far as possible.
The bill does none of this. Like Oscar Wilde's cynic, it only understands price, not value. A better way of doing it is through a land auction. The government should buy up some land at the project site as well as its neighbourhood by conducting an auction. It can then compensate farmers who own land on the site, but who haven't sold their land in the auction, by giving them equal-sized plots outside, instead of cash.
Let's take the case of Singur where 1,000 acres were needed for a car factory. Include another 1,000 acres in a surrounding belt, and suppose all landowners in the combined area of 2,000 acres are asked to submit tender bids to the government for sale of their land. The 1,000 acres on which bids are the lowest will be acquired against cash compensation. However, the compensation rate will be uniform and set equal to the lowest losing bid (the bid on the 1001st acre when arranged in ascending order of bids).
Some land in the project site will remain unsold in the auction. Owners of these plots can be given equal-sized plots in the surrounding belt that have been procured through the auction. In the end, a farmer whose land falls in the project zone is given compensation either in land or cash, and he is effectively able to choose between the two. He is also able to name his own price - the minimum amount he will accept in cash. The government may set a reserve price to ensure that the cost doesn't get out of hand. If the auction price exceeds the reserve price, the acquisition effort must be suspended and land sought elsewhere.
This method is far less coercive. It takes care of two common complaints - that farmers have received inadequate compensation, and many have been deprived of an asset that is central to their lives. If the farmer doesn't want to give up cultivation, he can name an astronomical price and make sure he is given a similar plot of land in the surrounding belt. Those who are forced to give up land often feel cheated because the owners of neighbouring plots, lying outside the project zone, can sell later, fully appropriating the higher real estate prices that come in the wake of commercial development. Under the auction, everybody is equal and owners can factor in anticipated future land price increases into their bids. The competitive aspect of an auction will also keep farmers' demands honest, and reallocate the remaining farmland to those who value it most.
If the affected farmers are guaranteed a higher standard of living, does it matter whether a car factory is built on their farms or an army camp? If the industry is willing to bear a greater compensation burden for multi-cropped land, to reap the advantage of strategic location, are not the social benefits worth the social costs? The numerous caveats and exceptions stacked on to the Land Acquisition, Rehabilitation and Resettlement Bill, 2011, betray a lack of faith in its ability to meet its main goal - finding a fair price. The drafters have taken a shot in the dark, hoping it will hit the mark.
The country deserves better.
Maitreesh Ghatak is professor of Economics at the London School of Economics and editor-in-chief of the Journal of Development Economics.
Parikshit Ghosh is reader, Department of Economics, Delhi School of Economics
The views expressed by the authors are personal