The Poverty and Environment Network analyzes data gathered from some 8,000 households in 24 developing countries.

BOGOR, Indonesia — Forest resources may be less important as a buffer between harvests and in times of hardship than previously believed, according to a comprehensive global study.

The global study is the product of the Poverty and Environment Network (PEN), a collaborative effort led by the Center for International Forestry Research (CIFOR). The largest quantitative global-comparative research project to date on forests and rural livelihoods, it analyzes data gathered from some 8,000 households in 24 developing countries.

“Safety nets, gap filling and forests: a global-comparative perspective” is one of five initial papers to emerge from the global study; they appear in an upcoming issue of the journal World Development.

How does land tenure affect agricultural productivity?

A systematic review

Much earlier research has indicated that forests can provide a key “natural insurance policy” to poor rural households in developing countries. In the event of floods, fires or pests that reduce cropping areas, for example, studies have shown that smallholders might spontaneously convert more forests to cropland. Most literature, however, focuses on a household’s option to increase extraction of timber and non-timber forest products (NTFPs) in times of crisis.

Challenging conventional wisdom, the PEN research team found that barely one out of 10 households ranks extracting natural forest or other non-cultivated resources as its first response to an emergency. Other strategies, such as seeking off-farm employment, asking for help from neighbors, or reducing consumption, all proved more important.

“People often think about forest products as something households can access to sell or consume easily when they need them most,” says Sven Wunder, the paper’s principal author and head of CIFOR’s regional office in Brazil. “But that’s probably not how we should look at them. There can be exceptions, like commercially valuable timber stands that are harvestable all year, but most extractive products have their own production cycles, and are seasonally or otherwise constrained in supply.”

In addition, the study found little evidence to support the popular view that forest income fills a gap for smallholders between harvests. In most cases, quarterly forest income is positively correlated with both total income and crop income, which effectively excludes it from being a gap filler, Wunder said. Indeed, the data suggest smallholders are more likely to seek temporary employment or other options between harvests than to exploit forests.

Looking for reasons

Why such a discrepancy from the established literature?

One major reason, Wunder says, is that many researchers are probably too focused on the absolute rather than relative value of forests following a crisis. “If you ask, ‘Did you go to the forest to gather more forest products to better cope?’ then many people will say they did,” he said. “But maybe they also did half a dozen other things, like having someone in town send home money, take work outside as a farm hand, or ask neighbors to help out, that were more important in their response strategy.”

Another possible reason is the reliance on case studies, says co-researcher Jan Börner, a CIFOR associate from the Center for Development Research at the University of Bonn. “Depending on how these case study locations were chosen, there could have been a bias towards locations where forests do, in fact, have important safety net functions,” he said.

The paper also identifies other factors that could explain the discrepancy between earlier research and the global study. These include the growing importance of labor markets, the prominence of development cooperation and emergency relief, and the possibility that previously rich forest resources have been depleted.

“We can’t really say to what extent the world has changed, or if many earlier studies were just not asking all the pertinent questions,” Wunder said. “We are deliberately leaving those two interpretative options open, because ours is not a time-panel study, it’s only a snapshot that cannot be extrapolated back into the past.”

Finally, the research team also acknowledges potential shortcomings in the PEN sample coverage. “We didn’t have full-blown calamities events like Hurricane Mitch or catastrophic floods included in our data,” Wunder noted. “We can’t really be sure that people wouldn’t go into the forest much more often if those extreme events were to happen, rather than our more moderate shocks.”

He also points out that the paper does not build on a specialized shock-response questionnaire. “This has the disadvantage that we don’t know much detail about the nature of the shock and the response to it, but it has the advantage that we know a lot about the household’s livelihoods, and thus can contextualize the response very well,” he said. “We can take all the other variables and use them as controls to explain why people might have reacted as they said they did.”

Policy implications

While this paper questions the primary role of forests as safety nets in a crisis and as a seasonal gap filler, another article from the global study found that forests and other wild lands are more important as regular household income contributions than previously thought. “Forests may help people generate more regular income, but may be less useful to help cope with shocks,” Wunder said.

“There is much more one can do with the rich PEN data set,” Börner said. “Here we looked at what people said they did in response to a shock. But our data may also offer some insights as to what people actually did following quarters with a severe income shortfall,” he said.

“I doubt that these follow-up analyses would invalidate the general findings, but we can certainly learn more about how households react to crisis events, and what role forests may have in the bigger picture.”

For more information about the issues in this article, please contact Sven Wunder at s.wunder@cgiar.org.

This study was supported by ESRC-DFID, Danida, USAID (BASIS-CRSP), IFS and CIFOR.