9
Using this common definition for Repressor
it
to identify “bad” states, we consider
the extent to which these states belong to the human rights legal regime, ratifying the
CCPR or the CAT into national law. We accordingly estimate a simple model:
(1)
Number of Treaties Committed
it
= α + β
1
Repressor
it
+ β
2
Polity
it
+ β
3
Regime Durability
it
+ β
4
GDPpc
it
+ β
5
Trade
it
+ β
6
Civil War
it
+ β
7
War
it
+ β
8
Population
it
+ β
9
North America
it
+ β
10
Europe
it
+ β
11
Africa
it
+ β
12
Middle East
it
+ β
13
Asia
it
+
δ
i
+
µ
it
Our dependent variable, Number of Treaties Committed
it
, is an ordinal variable
coded 0 if a state i in year t has made no formal commitment to either the CCPR or CAT
by ratifying, acceding, or succeeding; 1 if that state has committed to one treaty but not
the other; and 2 if that state has committed to both treaties. Our sole aim in this model is
to understand whether states that are observably repressive are any more or less likely to
belong to human rights laws than those that are reasonably protective, having already
institutionalized the norms to which they make commitments. We pursue this aim as
systematically as possible and control for a variety of other motivations considered
relevant in state decisions on treaty membership (Cole 2005; Hathaway 2005; Moravcsik
1995).
Polity
it
measures domestic regime characteristics, coded by Jaggers and Gurr. The
well-known variable, which takes on values ranging from 10 (most democratic) to -10
(most autocratic), is measured as an index of five primary institutional features: the
competitiveness of chief executive selection; the openness of that process to social
groups; the level of institutional constraints placed on the chief executive’s decision-
making authority; the competitiveness of political participation; and the degree to which
Convention