Premium
Los Complejos Vínculos entre la Autoridad y la Biodiversidad
Author(s) -
BARRETT CHRISTOPHER B.,
GIBSON CLARK C.,
HOFFMAN BARAK,
McCUBBINS MATHEW D.
Publication year - 2006
Publication title -
conservation biology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.2
H-Index - 222
eISSN - 1523-1739
pISSN - 0888-8892
DOI - 10.1111/j.1523-1739.2006.00521.x
Subject(s) - language change , explanatory power , natural resource , corporate governance , affect (linguistics) , resource (disambiguation) , biodiversity , public economics , environmental resource management , economics , econometrics , ecology , psychology , computer science , biology , epistemology , art , philosophy , computer network , literature , communication , finance
We argue that two problems weaken the claims of those who link corruption and the exploitation of natural resources. The first is conceptual and the second is methodological. Studies that use national‐level indicators of corruption fail to note that corruption comes in many forms, at multiple levels, that may affect resource use quite differently: negatively, positively, or not at all. Without a clear causal model of the mechanism by which corruption affects resources, one should treat with caution any estimated relationship between corruption and the state of natural resources. Simple, atheoretical models linking corruption measures and natural resource use typically do not account for other important control variables pivotal to the relationship between humans and natural resources. By way of illustration of these two general concerns, we used statistical methods to demonstrate that the findings of a recent, well‐known study that posits a link between corruption and decreases in forests and elephants are not robust to simple conceptual and methodological refinements. In particular, once we controlled for a few plausible anthropogenic and biophysical conditioning factors, estimated the effects in changes rather than levels so as not to confound cross‐sectional and longitudinal variation, and incorporated additional observations from the same data sources, corruption levels no longer had any explanatory power.