Structural Integration and The Trees: An Analysis of Deforestation in Less-Developed Countries, 1990–2005

Abstract

This study tests a hypothesis derived from an expanded theory of foreign investment dependence. The tested hypothesis states that less-developed countries with higher levels of primary sector foreign investment exhibit greater rates of deforestation. Findings for cross-national analyses of deforestation from 1990 to 2005 for 40 less-developed countries confirm the hypothesis, providing support for the proposed theorization. Additional results indicate that the presence of environmental international nongovernmental organizations is beneficial for natural forest areas, while population growth is a key driver of deforestation in less-developed countries. Besides confirming the hypothesis, this research underscores the importance for sociologists to consider both political–economic forms of integration and human-ecological factors when investigating how humans impact the environment.

ACKNOWLEDGMENTS

The author thanks the editor and anonymous reviewers for helpful comments on an earlier draft of this article.

NOTES

Notes

¹ While the overall forest size of developed countries has increased through reforestation, aforestation, and the partial outsourcing of extraction and production, forest degradation does indeed still take place in developed countries. Moreover, while we emphasize the overall divergent patterns of forest degradation for developed and less-developed countries, notable regional variation in deforestation rates also exist (Rudel 2005).

² Deforestation is an important form of environmental degradation to investigate for multiple reasons. For example, forested regions provide household fuel and other sorts of important material resources for agrarian communities in less-developed countries, many of which are highly impoverished and lack economic resources to access similar resources elsewhere. Moreover, forest degradation contributes to soil erosion, desertification, seriously disrupts regional ecological systems, and threatens biodiversity.

³ Elsewhere, we control for total population size and rural population size in 1990 (point estimates). The effects of both point estimates are nonsignificant and their inclusion does not alter the reported effect of primary sector foreign investment as well as the additional statistical controls.

⁴ To further account for the possible effects of international trade and the extent to which they may suppress the effect of primary sector foreign investment, elsewhere we also control for growth in primary sector exports. The effect of the latter is nonsignificant and its inclusion does not suppress the reported effect of foreign investment on natural forest degradation in less-developed countries. We thank an anonymous reviewer for suggesting these additional analyses as well as those reported in endnote #3.

⁵ Using appropriate diagnostics (e.g., Cook's distance), we determine that the sample analyzed in the current study does not contain any overly influential cases.

⁶ Aforestation usually refers to the artificial establishment of forests by planting or seeding in an area of nonforest land. Reforestation usually refers to the restocking of existing forests and woodlands which have been depleted, with native tree stock. The latter can occur naturally or artificially.

⁷ All variables included in the analyses that are logged are done so for analogous reasons.

⁸ In a series of unreported analyses, we include measures of economic growth, total population change, and rural population change for 1990–2005. The effects of all three predictors are very similar to those reported for the same three measures from 1980–1990. Because of issues of temporality and causality, we consider the latter measures to be more appropriate for the current study.

⁹ We thank Smith and Wiest for sharing these data with us.

¹⁰ Elsewhere we include different measures of democratization and state strength as additional predictors. The effects of the additional predictors are all nonsignificant, and their inclusion does not suppress the reported positive effect of primary sector foreign investment.

¹¹ This VIF is a function of the strong negative correlation between per capita GDP and primary sector activities as percentage of GDP (see Table 2).

¹² Following the suggestions of an anonymous reviewer, elsewhere we include (1) a quadratic for forest stock to examine if total forest area has a nonlinear effect on deforestation, and (2) Rudel and Roper's (1997b) measure of forest fragmentation for the mid-1980s to early 1990s. Forest fragmentation is a ratio of the perimeter to the area in tropical moist forests in a country, and the ratio will be high in fragmented forests and low in more consolidated forests, and the perimeters include all forest-nonforest boundaries (Rudel and Roper 1997b:101). The effect of the forest size quadratic is nonsignificant, while the effect of forest fragmentation is positive and marginally statistically significant (p < 0.10, one-tailed test). More importantly, including both additional controls does not suppress the reported positive effect of primary sector foreign investment dependence. In fact, when controlling for forest fragmentation, the relative magnitude of the effect of primary sector foreign investment increases. We thank the anonymous reviewer for making these very helpful suggestions.

¹³ The impacts of ecologically unequal exchange on deforestation are beyond the scope of the current study (see Jorgenson 2006c).