nities outside the frontier will attract labor away from forest-clearing activities
in the uplands, as illustrated by a Philippine irrigation study by Shively and
Martinez (2001). Labor-saving technologies, on the other hand, foster greater
migration to the frontier. Ruf (2001) reports that green revolution technolo-
gies (e.g., mechanization) in Sulawesi were labor saving and spurred the con-
version of forests to cocoa holdings in the uplands.
A comparison of rubber agroforestry in selected sites in Borneo and Suma-
tra in Indonesia illustrates the critical role of the labor market in determining
the forest outcome. Many analysts have blamed the introduction of rubber
into shifting cultivation systems in Southeast Asia for provoking large-scale
forest conversion. The ASB Programme has conducted extensive research in
Sumatra on rubber agroforestry. Although rubber agroforestry has many
attractive features, it has not stopped conversion of primary forest to small-
holder rubber holdings for exactly the reasons explained earlier: high prof-
itability in combination with large in-migration of labor (Angelsen 1995;
Tomich et al. 2001). The situation is a typical example of the second case in
Table 5.1. This result is contrasted with a study by de Jong (2001), who found
that rubber agroforestry contributed little to encroachment into primary for-
est in areas of West Kalimantan (Indonesia) and Sarawak (Malaysia). De Jong’s
study found that incorporating rubber gardens (or rubber-enriched fallows)
and additional tree cover in lands previously used for slash-and-burn agricul-
ture produced both economic and ecological benefits. Several factors explain
this difference, such as better forest management in the Borneo cases. An
important difference was the remoteness of and low in-migration to the Bor-
neo sites. In Sumatra the adoption of rubber was accompanied by substantial
in-migration from Java and other parts of Sumatra, so labor shortages did not
dampen the conversion of forests to rubber agroforestry. The situation was
very similar to that of cocoa agroforestry in West Africa, where expansion was
possible because of in-migration from the savanna zone into the rainforest (see
Chapter 6, this volume). Moreover, the expansion had limited impact on the
world market price of rubber.
Land Tenure
So far we have not dealt much with issues of property rights and land tenure.
Insecure land rights and open access situations often are noted as key under-
lying causes of deforestation and act as a disincentive for investments in land,
including agroforestry (Wachter 1992). However, the empirical evidence is
more complex than the simple theory suggests. The forest impact of techno-
logical change generally, and agroforestry specifically, depends critically on the
existing property regime. Generally, open access situations might encourage
investments by clearing more forests, whereas contexts with well-defined
- Is Agroforestry Likely to Reduce Deforestation? 101