Earlier this week, I wrote about an alternative to permanent land ownership, arguing that permanent land ownership benefited those originally allocated the permanent ownership, at the expense of future generations. The alternative involved allocating land every 17 years to everyone in a hypothetical town called Jubilee.
But how is this allocation performed? Who decides how to divide the land so that everyone receives an equally valuable plot?
Also, Eric Crampton raises an objection that reallocation every 17 years destroys incentives to invest in improving the land for long-term benefit. Can we restore these incentives?
The most obvious way to reinstate some of the incentives is to perform the land reallocation less often, perhaps every 50 years. But this causes two problems. One is that some people will have a land allocation for their entire adult lives, while others, through an accident of birth, will have to wait until they’re nearly 50 years old to have the same opportunity. The other problem is that even if you have the right incentives immediately after you’ve been allocated land, your incentives will deteriorate as your allocation approaches its expiry date. I address both these problems in this article.
First, we can perform long-term land allocations every 17 years, ensuring that every adult has the value of an allocation. We do this by dividing Jubilee into three regions. Every 17 years, everyone (anywhere in Jubilee) is given their land-use rights. These are the “currency” used in allocation auctions; every parcel of land in one of the three regions is auctioned, and the winners of the auctions get to use the land for 51 years. People can choose to use their land-use rights to bid in the auctions, they can save the land-use rights for 17 years until the next region is auctioned, or they can sell the land-use rights to someone wanting to bid for a large or otherwise valuable parcel of land. This last option is a low-fuss alternative to bidding for an allocation and then leasing it to someone more skilled at working the land.
But having gone this far, we can smooth things out a bit. Instead of three regions, we could divide Jubilee into 50 regions; every year everyone would be given their land-use rights; every year there would be auctions to allocate for 50 years the land in one of the regions.
Auctioning the land allocations means that we can ensure that everyone is allocated land-use rights of equal value; we also avoid having to have some fallible authority estimate land values in order to effect this equal allocation. And we’ve addressed one of the problems of making long-term allocations; every adult will have an allocation, no matter what year they were born in.
But we still have the problem that land managers’ incentives to invest in improving the land appear to deteriorate as their 50 years come to an end. Suppose you own a pine plantation, and there are only a few years left until the land it’s planted on is up for auction again. You consider planting more trees to replace ones you’ve recently cut down. But the cost of planting them is outweighed by the risk that someone else will be allocated the land, and you won’t benefit. So you don’t plant them, even though the benefit to society of planting them outweighs the cost.
[EDIT: The rest of this post has largely been superseded by a newer post on a combinatorial auction process for preserving incentives.]
But consider what happens at the auction. Suppose the value to you of the land and trees both together is . If you lose the auction, you can dig up young trees, cut down old ones, and sell them, for a net salvage value of . In the auction, you’re willing to pay up to for the land; any more than that, and you’d be better off keeping the land-use rights and selling the trees.
Your rivals in the auction place a value of on the land without the trees. Naively, we might expect the rivals to bid up to for the land, knowing that you’ll take the trees away if they win; let’s go with that for the moment, but see below for a more sophisticated strategy. You need to bid only enough to beat your rival, so you end up paying approximately for the land, which is worth to you, so your incentives are restored. (If , then the land and trees are more valuable to society separately than together; your rival wins the auction, and society benefits.)
But now suppose your rival places a value of on the land and trees together. They have a good estimate of , and they decide to bid up to for the land. If , then they win the auction, paying approximately your high bid of . Then, they offer to buy the trees from you. Their offer is higher than , so it’s in your interest to accept, but lower than , so the total they’ve paid for trees and land is less than , so they’re better off.
The problem is that you now expect to pay close to if you win the auction, so planting trees shortly before the auction just increases what you expect to pay to keep the land; your incentives are demolished again.
But now, consider this counter-strategy. Suppose that before the auction, you have a good estimate of . You make a public, binding commitment not to sell your trees in-place for less than your estimate of . Knowing this, if your rival intends to buy the trees in-place from you, they will bid no more than for the land, so that the total they pay for land and trees together is no more than .
I may have made a mistake, but as far as I can see, this strategy and counter-strategy form a Nash equilibrium, and all the right incentives are restored for people to invest in improving their land, regardless of how soon it will be auctioned.