6
Curve, named after S. Kuznets’ proposed inverted U-shaped pattern between income inequality and
economic growth (Kuznets, 1955, 1965, 1966), has been the subject of substantial debate and scru-
tiny. Yet, research is still far from reaching consensus over its validity. An inverted U-shaped rela-
tionship between economic development and the environment was for example found in cross-
country studies of air pollution, such as CO
2
, NO
x
, and SO
2
, as well as of energy use, clean water,
urban sanitation, nitrates, suspended particulate matter, waste, and deforestation (Shafik and Ban-
dyopadhyay, 1992; Cole Rayner and Bates, 1997; Galeotti and Lanza, 1999; Panayotou Sachs and
Peterson, 1999; Bhattarai and Hammig, 2000; Kallbekken, 2000; Ehrhardt-Martinez et al., 2002). At
the same time, a number of studies have demonstrated an N-shaped pattern for the relationship
between income and CO
2
, NO
x
, SO
2
, and smoke (Grossman and Krueger, 1993, 1995; Selden and
Song, 1994). Moreover, scholars have discovered a linear logarithmic pattern, implying that an in-
crease in emissions is strongly correlated with income, but that further improvements in environ-
mental quality does not necessarily depend on further economic growth (Bruyn, Bergh and Op-
schoor, 1998). Other scholars instead address the issue of reverse causality, assuming that it is envi-
ronmental degradation that causes income to decrease (Stern et al., 1996), since economic activities
depend on environmental resources and that their unsustainable use “reduce[s] the capacity of gen-
erating material production in the future” (Arrow et al., 1995).
Taken together, the effects of democracy on the environment, as well as the effects of economic devel-
opment on the environment, or even the effects of democracy on economic development and vice versa, are obviously
subjects of considerable controversy and disagreement. In an attempt to contribute to these research fields, we set out to
perform a more fine-grained empirical analysis, including levels of economic development and democracy in a joint
analysis. Taking a departure in the discussion on causal mechanisms by Li and Reuveny (2006), we
argue that there are reasons to believe that these mechanisms function differently depending on
surrounding institutions and especially levels of economic development. The five mechanisms of
positive impact of democracy might be more functioning when economic development is high.
Conversely, the four negative effects of democracy might very well be better functioning when
economic development is low. In order to develop this argument, we are theoretically informed by the
well-established – yet in this context partly overlooked – literature on modernization and democratic consolidation.
This literature holds that low-income settings per definition lack institutions stimulating economic
development, and that in the absence of such institutions democracy might be a less effective way
to govern. In short, if not preceded by a constitution, rule of law or secure property rights generat-
ing economic development, democracy does not necessarily function as an instrument of collective
action, but rather risks being used as an instrument of patronage and clientelism (Leftwich, 1993;
7
Zakaria, 2003; Keefer, 2007; Walker, 1999, Diamond 2008, 2007). Moreover, in low-income set-
tings, democracy is often imposed from abroad, lacking legitimacy and correspondence between
informal and formal institutions, which induces leaders to act for their short-term survival rather
than engaging in the provisioning of long-term public goods such as protection of the environment.
This in turn makes the legitimacy of the system decrease even more. In addition, without welfare
improvements, citizens tend to distrust the democratic system and risk ending up engaging in pat-
ronage and clientelism themselves (Collier, 2009; Kapstein and Converse, 2008; Keefer 2007). In
sum, this literature holds that the workings of democracy differ significantly depending on levels of economic in-
come, and if democracy does not deliver, its positive effects as an instrument for collective action will
hence be absent. We thus have reason to believe that the causal mechanisms discussed by Li and
Reuveny (2006) are in fact conditioned by democratic consolidation, and more specifically, the
institutional arrangements prevailing at different levels of economic development.
The case of marine resources
A focus on marine resources when investigating the effect of democracy during different stages of
economic development is appropriate in many respects. For example, being a fungible natural re-
source, it accentuates many of the governance challenges associated with common pool resources
(Ostrom, 1990). Fisheries are in fact often used as textbook illustrations of common pool resource
problems and the importance of collective action mechanisms such as democracy or other govern-
ance arrangements. Yet, empirical studies on the effect of democracy on the marine environment
are scant, and demonstrate conflicting results (Jagers and Sjöstedt, 2011). Similarly, the effect of
economic development on marine resources is far from settled empirically. In a study on the rela-
tionship between income and marine resource exploitation in Turkey over time, Kamanlioglu
(2011) finds an inverted N-shaped relationship between the deterioration of marine environmental
quality and economic growth. However, the author points out that such a pattern is shaped by
country-specific factors. Sabah (2011), on the other hand, finds an N-shaped relationship between
economic development and coral reef bleaching. Clausen and York (2008a, 2008b) report, however,
that the rise of per capita income leads to the decline of the marine trophic level, without further
improvement of the indicator at higher income levels.
In the next section, we further specify how we proceed in testing the impact of democra-
cy and economic development on the marine environment.