By Ben Balmford
30 March 2021
Ben Balmford is a SWEEP PhD student. His research looks at how to design mechanisms to incentivise positive environmental outcomes.
In the UK, as across the world, there is growing interest in the use of schemes which pay landowners to change their behaviour to benefit the wider world. Prominent examples include paying for land management in the Catskills to safeguard the water quality of New York’s drinking water, and the REDD+ programme paying for forest conservation to reduce carbon emissions. Yet a persistent question remains: how to set the price paid to scheme participants?
One solution long favoured by economists has been to use auctions. Under an auction approach, landowners offer to carry out a particular land management for a particular price. The funder then selects those with the lowest bids which receive funding to go ahead and carry out the particular activities. This is where we, and EnTrade, come in. EnTrade act as a broker designing and running auctions for specific funders, which are normally water companies looking to purchase land management practices that would reduce surface run-off and hence improve water quality and lower filtration costs. At the Land, Economics, Environment and Policy Institute, part of the University of Exeter, where I am based, we had been offering guidance on the design of various funding schemes for farmers to improve environmental outcomes on their land, as part of the NERC funded SWEEP programme.
EnTrade successfully set-up a market in Poole Harbour, and since 2016, farmers have been funded by Wessex Water to plant cover crops to reduce nitrogen leaching in the catchment. Yet, EnTrade wondered whether a different auction design may improve the auction outcomes, and hence reached out to us in the Mechanism Design at LEEP Institute at the University of Exeter to ask for some advice. We worked with them to explore the potential to implement a “uniform price” auction rather than a “Pay-as-Bid” design. In a Pay-as-Bid auction, farmers would submit bids, with those submitting the lowest bids getting contracts and being paid the amount specified in their bids. A Uniform Price on the other hand accepts the lowest bids, but pays the amount specified in the highest accepted bid to all accepted bids. The key point is that the best bid changes. In a Pay-as-Bid auction, bidders must bid above cost, working out how much to ask for based on what their costs are and how much extra profit they can ask for and still be successful; quite a complex calculation. In the Uniform Price auction, instead, a farmers best bid is state their actual cost. Their profit instead coming from the fact the amount they will be paid if they are accepted is basically bound to be more than they bid.
Diagram 1: bidding behaviour in the Pay-as-Bid (left) and Uniform Price (right) auctions. In both cases, bids are ordered from the least cost (bottom left; price indicated on the vertical axis) through to highest cost (top right). The horizontal axis is the cumulative supply. The blue curves in each case are the same and represent some hypothesised supply curve – the true cost curve. The green vertical line is the quantity constraint, to the left of which all bids would be accepted, with the peach area representing total spend. The key difference is two-fold. First, in the Pay-as-Bid auction accepted bids are paid the amount stated in the bid, hence the curved top line for the peach area. This contrasts with paying each bid the amount in the last accepted bid, as in the Uniform Price auction (and hence the horizontal top line for the peach area). Second, bids in the Pay-as-Bid auction must overstate cost, resulting in bids being above cost. This contrasts with the Uniform Price auction in which bids are at cost where bids are at cost – profit for bidders coming from the gap between one’s own bid and the highest accepted bid.
For us, as an exercise in academic-driven impact, it was all pretty exciting as the uniform price rule had not previously been experimented with in any payment for ecosystem service scheme in the world. But it was important too in generating on the ground change now. While we couldn’t be sure how the results would go, we were fairly confident that bids at least must be lower, and that it would reduce the burden on farmers in the bidding process. Thankfully, the results bear this out.
The switch led to lower bids and a reduced burden of participation for farmers. Of course, paying farmers above their bids – as the uniform price does – we didn’t know whether the Pay-as-Bid rule that had previously been applied would have allowed the funder to buy more or less water quality improvement than they could under the Uniform Price. Yet here too there was good news. For the same budget, Wessex Water were actually able to purchase far more improvement in water quality, and have far greater acreage in the scheme under the Uniform Price. The result then was a brilliant win-win-win: more land under cover crops for Wessex Water, lower costs of participating for the farmers, and greater water quality improvement for all those in the Poole Harbour catchment.