Hvg krause 47K. qxd

Politics, Policy, and Organizations

distinguish policy stability in presidential systems from policy stability in

parliamentary systems. Instead, differences in policy stability between two

different kinds of systems depend on the interaction between the number

of veto points and the distribution of preferences (what may be called the

preference pro

file) of the elected officials populating the veto points in the

two kinds of systems. In fact, it was demonstrated that for some prefer-

ence pro

files presidential and parliamentary systems should be expected

to select similar policies and exhibit similar patterns of policy change, de-

spite the systems’ institutional differences.

1

In this essay, we utilize these two variables—the number of institu-

tional veto points in a system and the extent of policy disagreements

among the elected of

ficials in these veto institutions—to develop an an-

swer to a central question: Are different kinds of democratic political

systems necessarily characterized by different degrees of bureaucratic

autonomy?

Policy Equilibria and Bureaucratic Autonomy

To answer this question, we begin by observing that multiple veto points

in a system may enable the bureaucracy to adopt new policies unilaterally.

The reason is that the multiple veto points may create a set of policies that

are in equilibrium; policies in equilibrium cannot be upset by any possible

decisive coalition of the elected of

ficials (given their individual policy

preferences).

2

The existence of a set of equilibrium policies means that the

bureaucracy could adopt any one of the equilibrium policies, and change

from one equilibrium policy to another, without fear that its chosen pol-

icy will be upset by any decisive coalition of elected of

ficials. If the set of

equilibrium policies is large, then the bureaucracy will have substantial

room for unilateral policy change (and thus might be considered relatively

autonomous), whereas if the set of equilibrium policies is small the bu-

reaucracy will have little room for unilateral policy change (and thus the

bureaucracy might be considered to have relatively little autonomy).

These arguments lead to a clearer and more speci

fic definition of bu-

reaucratic autonomy. Earlier de

finitions referred to the general ability of a

bureaucracy to do what it wants, but the de

finitions did not embed the

bureaucracy in any particular political context. This left it unclear as to

whether the bureaucracy could adopt any policy it wanted or just some

policies, and if just some policies were feasible the de

finition did noth-

ing to specify what particular policies were feasible and why.

76

---

The de

finition advanced here links bureaucratic autonomy to the pol-

icy preferences of the elected leaders and resulting equilibrium policies.

This de

finition of bureaucratic autonomy makes it clear that a bureau-

cracy can be more or less autonomous, depending on the size of the set

of equilibrium policies. Moreover, by relating the extent of bureaucratic

autonomy to the size of a set of equilibrium policies, there is always a

boundary to the set. This boundary sets limits on what the bureaucracy

can and cannot do: it can move from policy to policy within this equi-

librium set, but it cannot sustain a policy that lies outside this equilib-

rium set. While this de

finition makes it clear that the preferences of the

elected of

ficials will always collectively constrain the range of bureau-

cratic choices, it also suggests that as long as the bureaucracy selects some

new policy from inside the boundary the disagreements among the

politicians will keep them from upsetting the bureaucracy’s choice and

imposing some other policy.

The possibility that presidential and parliamentary systems may have

sets of equilibrium policies that differ in size has been examined by

Hammond and Butler (2003). However, while an increase in the num-

ber of veto points can increase the size of the set of equilibrium policies,

it remains to be determined whether we should necessarily expect to 

find

systematic differences in the size of the set, and thus in the extent of bu-

reaucratic autonomy, across democratic systems. The problem is that the

preference pro

file will almost invariably differ from country to country,

even for countries with the same number of veto points. Determining

what, if anything, should thus be expected from the relationships among

veto points, preference pro

files, equilibrium policies, and bureaucratic au-

tonomy is the general problem we must consider in the rest of this essay.

As we will demonstrate, variation in both key variables makes it dif

ficult

to develop logically valid expectations, for hypothesis-testing purposes,

about how much bureaucratic autonomy we should 

find in the different

kinds of systems.

Policy Equilibria in Democratic Systems

To assess the size of the sets of equilibrium policies, and thus the extent

of bureaucratic autonomy, in different kinds of democratic political sys-

tems, we construct several unidimensional spatial models of policy-

making. A political system is de

fined here as a set of rules used to aggre-

gate the preferences of the individuals in the system into the choice of

Veto Points in Democratic Systems

77

---