Comparing Institutional Forms 79
activities—problem perception and defi nition involve the comparison of some
pieces of information with others, while choice involves the comparison of one
option with another—and the argument in this essay has been that institutional
rules creating hierarchies have a substantial impact on the nature of these com-
parisons. (1993, 143)Comparing Institutional Forms
Th e association between the institutional structures or designs of institutions and
the policy and administrative outcomes of those institutions is an important and
long-standing subject in political science. Some of the scholarly work on this sub-
ject is based on the study of the national state as a political jurisdiction, work
that is generally well known (Lijphart 1984; Weaver and Rockman 1993). Less
well known, but in many ways more empirically and theoretically signifi cant, has
been the study of the relationship between institutional structures and designs
and the policy outcomes in American cities. At the level of the nation-state, insti-
tutional and constitutional designs diff er in how they unify or divide government.
Governments can be divided in various ways, including the formal separation
of powers, such as we see in the US federal government as well as in each of the
fi ft y states; split partisan control of the executive and the legislative branches;
and split partisan control between the legislative chambers. Partisan gridlock is
the contemporary description of divided government. As a general descriptor,
presidential-form government is divided government with checks and balances.
Parliamentary-form national government is unifi ed.
In the 1950s, political scientists were generally of the view that parliamentary
structures were more unifi ed than presidential structures and, therefore, improved
the prospects for both political party eff ectiveness and generalized public policy
effi ciency (American Political Science Association 1950; Ranney 1954). In more re-
cent times, scholars have considered divided government (Jacobson 1990; Fiorina
1996) and the implications of divided government (Mayhew 1991). Th is research
draws the general conclusion that divided government, particularly split partisan
control, is bad for public policy because it structurally allows for veto, empowers
interest groups, frustrates eff ective policy development, and diminishes the pros-
pects for eff ective policy implementation (Frederickson 1997a). David McKay,
basing his fi ndings on his studies of the American national government, writes
that “DG,” or divided government, “is almost universally perceived as a bad thing;
among other sins, it allegedly undermines coherent and cohesive policymaking
by removing the vital institutional connective tissue provided by common party
control. DG has been invoked, therefore, as the cause of a number of problems, in-
cluding the budget defi cit, diffi culties associated with the presidential appointment
and treaty-making powers, and a general inability to produce eff ective domestic
policy” (1994, 525).