agroforestry as a means to reduce deforestation. We start off by reviewing the
agroforestry-deforestation debate. Based on a study from Peru, originally pub-
lished in Sanchez and Benites (1987), a common claim is that 1 ha of new
agroforestry systems saves 5 ha of forest. We demonstrate how this single case
study has been misused and extrapolated to global predictions about how
agroforestry will reduce tropical deforestation. The reasoning ignores the exis-
tence of market repercussions and the fact that farmers are rational and will
take advantage of new opportunities. We then outline three typical situations
of agroforestry adoption and describe how the deforestation impact is likely to
vary between them. Then we discuss the subsistence logic underlying the land
degradation–deforestation hypothesis and question why farmers should not
expand their agricultural land into forests if a new and profitable agroforestry
technology becomes available. In the following sections we consider in more
detail three broad sets of conditioning factors that shape the agroforestry-
deforestation link: the characteristics of the farmer, technology, and market
and land tenure conditions. We use several examples to illustrate our argu-
ments.
This chapter draws on earlier work on how new agricultural technologies
affect the rate of tropical deforestation (Angelsen and Kaimowitz 2001). View-
ing agroforestry as one type of technological change, we can apply the devel-
oped framework and examine some of the case studies to answer the question
raised in the title.
The reader should note that we focus on the impact of agroforestry on the
conversion of natural forests to agroforestry and thus on off-site biodiversity.
This chapter does not discuss the question of on-site biodiversity, which is cov-
ered by several other chapters in this book (see Part III). However, a key issue
is possible trade-offs between on-site and off-site biodiversity, which is only
superficially dealt with in this chapter. Furthermore, we focus on deforestation
and not on the use of forest products and the impact of agroforestry on forest
product dependence (see Murniati et al. 2001 for a case from Sumatra).
Agroforestry Research and the Deforestation
Debate
In a review of agroforestry research and debate over the past couple of decades,
a few observations relevant to the topic of this chapter stand out. First, as Mer-
cer and Miller (1998) note, “biophysical studies continue to dominate agro-
forestry research while other important areas have not received the attention
they deserve.” (177) They found that only 22 percent of the articles published
in Agroforestry Systemsbetween 1982 and 1996 dealt mainly with socioeco-
nomic issues. Among these articles, quantitative economic studies constitute
more than half, with cost-benefit analysis being the most popular method.
Nair (1998) reported that the share of socioeconomic research articles pub-
88 II. The Ecological Economics of Agroforestry