The Weight of the Past: Exploring Lagged Determinants of Political Repression

• Published date: 01 June 1996
• Author: Christian Davenport
• DOI: 10.1177/106591299604900207
• Date: 01 June 1996
• Subject Matter: Articles

The Weight of the Past:
Exploring Lagged Determinants
of Political Repression
CHRISTIAN DAVENPORT, UNIVERSITY OF COLORADO AT BOULDER
The present article investigates contemporaneous and lagged effects of de-
mocracy, coercive capacity, and political conflict on repressive behavior. As
designed, 51 countries from 1948 to 1982 are examined with an Almon
distributed lag model on yearly data (N =1820). From the empirical inves-
tigation, both short- and long-term relationships are found to be signifi-
cantly related to the rate at which censorship and political restrictions are
applied. Past values of democracy are found to affect repression negatively
for five years and past values of dissident behavior are found to affect re-
pression positively for seven years. I conclude that a relatively complex
memory structure, encompassing both short- and long-term explanations,
must be employed when one attempts to understand why states use re-
pressive behavior.
Numerous studies maintain that causal determinants of political repression
I
have contemporaneous effects on the rate at which negative sanctions are ap-
plied (Hibbs 1973; Duvall and Shamir 1980; Ziegenhagen 1986; Davis and
Ward 1990; Poe and Tate 1994). Proponents of this view believe that explana-
tory factors for repression yield statistically significant and noticeable changes
within one iteration of the unit of analysis (usually conceived of as the nation-
year). Other studies maintain that causal determinants of repression have no
substantive impact at all when viewed contemporaneously (Duvall and Stohl
1983; Gurr 1986a; 1986b; Goldstein 1983; Petras 1986; Mitchell and
McCormick 1988). On the contrary, these studies suggest that explanatory
1
Repression is defined as government regulatory action directed against those challeng-
ing existing power relationships. This is similar to Goldstein’s definition (1978, 1983)
where, "(p)olitical repression consists of government action which grossly discrimi-
nates against persons or organizations viewed as presenting a fundamental challenge to
existing power relationships or key government policies, because of their perceived
political beliefs (1978: xvi). To decrease redundancy within the text "repression" will be
used interchangeably with "repressive behavior" and "negative sanctions."
377

---

factors exhibit their effects in a delayed fashion and that the use of negative
sanctions is changed slowly and cumulatively over time.
The two positions just identified are extremely important to the literature
on repressive behavior because they reveal very different decision-making
processes and because they influence our understanding of the causal rela-
tionships involved. In the first perspective, the decision-making process is
rather simplistic and only those factors that are immediately apparent to the
regime are considered important. To understand variance in repression here
we
need only survey the current political-economic environment for relevant
influences such as system type, economic development, and so forth. From
the other perspective, the use of memory is invoked and a more complex
decision-making process is indicated. In this case, those who make the deci-
sion to use negative sanctions are not believed to be driven by factors that are
immediately apparent to them, but they are believed to be driven by factors
that are removed from immediate experience (i.e., t-1, t-2,...t-n).
Which relationship and decision-making process is more accurate in cap-
turing the dynamics involved when governments decide to repress? This is
precisely the subject of the present analysis. Observing 51 countries from
1948 to 1982, I examine the impact of both contemporaneous and lagged
relationships on the rate at which negative sanctions are applied. The study
itself is divided into four components.
First, I present the argument for why we should expect there to be differ-
ent effects of certain variables on political repression. The focus here is the
decision-making process involved when negative sanctions are used as well as
the results of previous empirical research. The second component of the analysis
presents an Almon distributed lag model in order to examine the hypoth-
esized causal linkages. After detailing the fundamental components of this
strategy (its basic logic and structure), I discuss the reasons for employing it
as well as the limitations that are confronted when it is used. An empirical
analysis of the proposed hypotheses is conducted within the third component
of the study At this point, statistically significant relationships are identified
as well as support or refutation for the various hypotheses themselves. The
conclusion addresses the implications of the derived findings. This section
directly addresses how the results of this analysis better enable us to under-
stand the use of political repression across both time and space.
EXPLORING DIFFERENT EXPLANATIONS FOR WHY STATES USE
NEGATIVE SANCTIONS
From the literature, it is clear that both contemporaneous and lagged expla-
nations have been used to account for repressive behavior. While this diver-
sity has enhanced our understanding of negative sanctions and why they are
378

---

applied, it has also confused the subject matter because it is not clear which
perspective is more appropriate for representing the dynamics involved. In-
deed, it is not clear whether or not contemporaneous effects, lagged effects, or
both are relevant to explaining variance in political repression.
Fundamentally, I believe that part of the problem has been a lack of atten-
tion paid to the repressive decision-making process. There has been little ef-
fort put forth to understand the way in which those who decide to use political
repression are influenced by different aspects of the political economy This is
especially the case with regard to empirical investigations which tend to ig-
nore the issue altogether.
The use of repressive behavior is generally explained from two distinct
perspectives: rational choice’ (Stohl and Lopez 1983; Lichbach 1987; Karklins
and Petersen 1993) and habituation’ (Gurr 1986b; Ziegenhagen 1986; Hoover
and Kowalewski 1992). From the rational choice view, individuals within the
regime employ a decision calculus to maximize the likelihood of successful
behavioral regulation (i.e., decreasing political conflict). To accomplish this
task, several things within their immediate environment are assessed: (1) vari-
ous components of the political economy (the type of political system, the
preparedness of the coercive apparatus, economic development etc.); (2) the
characteristics of the challengers themselves (their ideological position, their
organizational size, etc.); (3) the challenger’s behavior (the type of activity
used, its frequency, etc.); (4) the availability of different regulatory strategies
(i.e., repression, accommodation); and (5) the preparedness as well as influ-
ence of the organizations affiliated with the different regulatory strategies. The
actual decision calculus itself is conducted in order to identify whether or not
the existing political-economic context is favorable to the use of repression.
Simply, if favorable conditions outweigh unfavorable ones, repression will be
used and if unfavorable conditions exceed the favorable conditions, then re-
pression will not be employed.
2
As Mason and Krane (1989: 178) state, "(t)he usual assumptions and caveats concern-
ing rational choice theory hold here. We simply assume that actors choose from among
available alternatives that course of action (or inaction) which, to the best of their knowl-
edge, most enhances their own well-being, however conceived. In choosing, they dis-
count each alternative in accordance with some subjective estimate of the likelihood of
that outcome and then choose the one with the greatest expected utility. In accordance
with the revealed preference approach, there is no need to presume that (repressive
decision-makers) actually make the calculations posited in the theory, only that they
behave as if they did.
3
By "habituation" I am referring to a situation where decision-makers rely upon previ-
ously established rules to guide their behavior (this could also be labeled "cybernetic").
Here, they are not expected to conduct repeated cost/benefit analyses. Instead, they are
expected to defer to previous experience and standard operating procedures.
379

---

The habituation perspective presents a very different view of the repres-
sive decision-making process. Here, it is maintained that governments are less
able and less willing to conduct a dispassionate cost/benefit analysis in their
decision to employ negative sanctions and that decision-makers are more in-
clined to follow standard operating procedure and emotional responses to
what has taken place before. With regard to the expected causal relationship
to repression, we find that when forces that favor the use of this behavior have
been established over time (institutions, beliefs, etc.) and previous challenges
to the regime have been significant (thus igniting emotional responses to them),
the use of negative sanctions is deemed more likely When these forces have
not been established and previous challenges have not been significant, how-
ever, the use of repressive behavior is less of a possibility
Despite the usefulness of these two perspectives what is perhaps most
problematic about them is that they each exclude consideration of the other
because of the causal relationships they...