Random trials in Uganda to show if payments for ecosystems services really work

To assess whether a promising approach to reduce poverty and conserve forests in Uganda really works, we need a randomised controlled trial, says Paul Hatanga.

Paul Hatanga's picture
Guest blog by
10 March 2014

Deforestation and fire ravage a forest area in Uganda (Photo: Paul Hatanga)

One way to promote sustainability is to pay villagers to protect local environmental resources, such as forests, because of the wider benefits this will bring. It's a simple idea – but does it work? That's what our project in Western Uganda aims to find out.

We are working with smallholder farmers in Hoima and Kibaale Districts, where the average household has seven members and a weekly per capita income less than US$4. The farmers chop trees for fuel wood and building materials. They clear the land to grow tobacco, maize and rice, even though this reduces soil fertility.

Across the region, forests are shrinking fast. A recent study led by Wildlife Conservation Society estimates that within 15 years, there will be none left there outside protected areas.

This spells trouble both locally and at larger scales because the forests provide a variety of goods and services, from regulating water flow and the climate to providing food, medicines and other useful products. The forests are also home to many wild species, including a chimpanzee population that will go extinct if nothing is done to avert the trend.

Economic reality

The more trees the villagers fell, the more they risk creating problems for themselves and others. But their economic reality means that they must do this to survive — it's the only way they can afford school fees and emergency health care.

The farmers can only change if they have an alternative. So, in 2010, the Chimpanzee Sanctuary and Wildlife Conservation Trust, International Institute of Environment and Development, Uganda's National Environment Management Authority and the United Nations Environment Programme launched a large-scale project that aims to both improve the situation and assess its own effectiveness.

We want to see if forest cover will increase if we pay villagers who avoid deforestation, reforest land, change agricultural activities and protect watersheds.

Proving it works

It can be difficult though to prove that such 'payments for ecosystem services' (PES) schemes work, for reasons Seema Jayachandran of Northwestern University explains in this blog post.

First, owners who would have conserved their forests anyway are more likely to join a PES scheme than farmers who plan to cut trees down. So comparing those who join the scheme with those who don't will overstate the benefits.

Second, comparing the situations before and after a PES scheme begins could overestimate the benefits if deforestation was falling for other reasons – and would underestimate the gains if deforestation had increased across the area.


To overcome these biases, we randomly assigned 70 villages to be eligible to take part in the payments scheme and assigned another 70 to be ineligible. We visited the eligible villages to inform people there about PES, how the forest cover was changing, and different ways to manage and use forests sustainably.

Forest owners from eligible villages could then choose to participate in the PES scheme or not, with individual farmers negotiating their own contracts. This approach will allow us to assess the actual impact of the PES scheme, over other factors, in terms of the change in forest cover, livelihoods and the status of wildlife populations.

Such 'randomized controlled trials' are central to medical science, for instance in testing new drugs. The development sector is increasingly using this approach to test how effective interventions in education, health and microfinance are. In the environment sector too they can provide the evidence needed to influence policy and attract private sector investment.

Tenure matters

The design phase of the project revealed another challenge. When IIED and CSWCT consulted local communities at the start of the process, we found that few had land documentation such as land titles.

More than 85 per cent of local people own land customarily through inheritance or local purchase agreements. Local leaders witness these processes, which lack modern surveys and land titling documents. This raises the risk of land related conflicts.

To build community-level confidence and ensure payments would go to rightful land owners, the scheme worked with local village leaders and monitors selected from each community. They worked to confirm and document land ownership, mapping the forest boundaries of 413 land owners in the 'treatment group'. Village local leaders supported the forest owners to verify that ownership status through consultation with their neighbours.

Since we began the project, we have seen the enthusiasm grow among these farmers to conserve their forest, improve performance as well as getting extra benefits such as securing land tenure, improved forest governance and accountability, and engaging in forest-based enterprises. Within six months, when our research ends, we will be able to conclude just how effective the PES scheme is for small holder farmers.

Paul Hatanga is a Project Manager with the Chimpanzee Sanctuary and Wildlife Conservation Trust (conservation@ngambaisland.org). He will present the latest findings of this project at an IIED conference on payments for ecosystem services in Edinburgh on 21 March 2014.

Was this page useful to you?