Supporting more productive farming in low-income countries

The need

Fragmented farms – where an owner has many small plots, spread over a wide area – are common in less developed countries.

The fragmented farms arise naturally over time from multiple causes, such as the passing of land between generations, limited adjacent land for sale for farmers ready to expand, and because splitting land can act as insurance from weather shocks when suitable insurance markets are missing.

Fragmentation is costly and inefficient, however.

“Farmers are walking multiple kilometres a day between their plots,” says Professor Tom Wilkening from the Department of Economics at the University of Melbourne.

“And they can't use any sort of mechanisation, like tractors, because it isn’t efficient in terms of scale in a fragmented environment.”

Professor Wilkening researches mechanism design, experimental economics and organisations.

He explains that buying land nearby to achieve larger plots and help transition to more modern farming can be complicated by cultural constraints (for example where land has been inherited) and the difficulty in finding owners close by willing to give up farming.

The question then becomes: “How can we help farmers put the land back together that is fair, simple, and respects cultural practices and preferences?”

Developing a solution

“Putting land ‘back together’ is not simple,” Professor Wilkening says, “and requires lots of movement and change.

“This can be economically difficult for farmers who are often credit-constrained. It’s hard to buy land in the first place in order to then sell it, and when they start selling that land, they may be at a disadvantage trying to buy replacement land closer to their other farms in a subsequent trade. As a result, nobody is willing and able to make the first move.”

Drawing on his background in market design (specifically building auctions and exchanges), Professor Wilkening began working with Associate Professor Gharad Bryan (London School of Economics) and Associate Professor Jonathan de Quidt (Queen Mary University of London), whose backgrounds are in development, on a system to help people trade land in a way that benefits all parties.

The market design approach began by first trying to quantify and understand the constraints that prevent efficient exchange, then designing a system tailored to users that can mitigate them.

The team initially ran simple trading experiments in Uganda to better understand land trade frictions.

“These experiments were a bit like a large game of monopoly where we gave villagers toy property titles and currency and allowed them to trade with one another over a week.  We then paid individuals based on their final allocations so that there was a monetary incentive to consolidate land and make efficient trades.”

Consolidating land, they found, was hard for people to do on their own. While farmers took the game seriously and tried to coordinate trade in a variety of ways, land holdings at the end typically remained fragmented and not efficiently reallocated.

Completing sequences of trades to achieve efficient holdings was another problem that emerged from the experimentation.

With the help of Dr Nitin Yadav (University of Melbourne), the researchers then built a prototype exchange – or platform – where people input potential trades and the system searches for possible land sets that can be put together.

“It’s a bit like realestate.com for an agricultural context,” Professor Wilkening says, “but based on trades and leases rather than sale, which helps avoid challenges around cultural constraints.”

“We've basically built a system that allows for conditional trades or swaps. So, a platform user indicates they would sell (or lease) a piece of land, conditional on access to another piece, and willing to pay $100, or need to be paid $100. Then the computer searches for mutually beneficial trades that allow the land to be consolidated efficiently while ensuring people aren’t left in tricky situations, unable to complete a trade. That's the main idea of the research.”

The outcome

Initial experiments with the platform using ‘toy’ property titles have been promising and the team is now expanding the project in multiple directions, preparing to deploy the system in the field.

Using satellite imagery, cadastral maps of a set of villages in Uganda where fragmentation is high have been linked to ownership titles to better understand spatial characteristics.

Small scale swap experiments, and a subset of leases, have also begun.

Together, the experiments have generated crucial insights into how contracts are enforced and the value farmers place on consolidating land.

“While too early to know whether the exchange will work, there is clear need and appetite. A midline survey on a pilot village, for example, showed those who swapped leases were very happy with the trades and likely to renew the contracts when they expired.

“They've asked when we're coming back to do more trading with them. So, we've had a very good reception.”

The team is also close to building the full computerised exchange.

Then, it’s about scalability.

Because while the cost of producing the platform is high – in time and money – once scaled up, the impact could be quite huge for a relatively simple, low-cost system.

“We’re rebuilding the prototype to be rolled out and tested in 50 to 100 villages. We want to create a scalable exchange that can be used in several different contexts,” Professor Wilkening says.

“Ultimately, we’ll be making farmers’ lives better since consolidated land helps reduce travel times, improves labour outcomes and allows for mechanisation.”

Partners

University of Melbourne’s Professor Wilkening is collaborating with Associate Professor Gharad Bryan (London School of Economics) and Associate Professor Jonathan de Quidt (Queen Mary University of London).

Dr Nitin Yadav (University of Melbourne) worked on building the exchange and Mariajose Silva Vargas (J-PAL Europe) oversaw the first field project in Uganda. Tumusiime Apollo (International Growth Research and Evaluation Center) and Dominik Biesalski (IIES) oversaw the second field project in Uganda.

Further Information

Visit tomwilkening.com for working papers and recent presentations.