topic; we measure it using the number of words included per topic. Our concern in
2009 was to relate these concepts to the endurance of the constitution; here we
simply report differences across regions in level of scope and detail.Figure 2. 1
presents the results in two panels. Our general finding is that Asian constitutions do
not have significantly different levels of scope from those from other regions. They
tend, it seems, to be less detailed than constitutions from Latin America or South
Asia, but are more detailed than those in the Middle East.
Executive and legislative power
A central feature of any constitution is its establishment of institutions through
which power is exercised. This attribute is true even of constitutions in
dictatorships or without rights. We evaluate the institutional structures using two
different measures,legislative powerandexecutive power.
Legislative power captures the amount of power formally assigned to the legisla-
ture. This includes powers that are distinct to the legislature, as well as the relative
power. The legislative power index draws on the survey by Fish and Kroenig,^34 and
consists of a set of thirty-two binary variables drawn from our CCP survey that
match the de facto items in their additive index.
For executive power, we use a measure we created on our own. We drew on
other data on the percentage of executive proposals that ultimately became legisla-
tion, to try to determine which of eight institutional features of executive power
“matter” most.
35
We also use a measure of maximum executive tenure developed
Table 2. 2 Relative rank of East Asia among eight regions.
Endurance 3
Rights 6
Political 5
Economic 5
Criminal 7
Executive power 6
Legislative power 2
Scope 5
Detail 5
(^34) M. Steven Fish and Mathew Kroenig,The Handbook of National Legislatures: A Global
Survey(New York: Cambridge University Press, 2009 ).
(^35) The measure of executive power is described in Zachary Elkins, Tom Ginsburg, and
James Melton, “Constitutional constraints on executive lawmaking,” paper presented at
American Political Science Association annual meeting, 2011 , Seattle, WA. It aggregates
the sub-components of the measure using weights derived inductively from a regression
model. Here, we adopt an unweighted version of that measure, because the sample in this
chapter is drastically different from that elaborated in Elkins, Ginsburg, and Melton.