Content uploaded by Daniel Auer
Author content
All content in this area was uploaded by Daniel Auer on Jan 04, 2017
Content may be subject to copyright.
1
United Nations Interventions
and Post-Conflict Democratization
Daniel Auer*
February 2016
Acknowledgments: I am thankful to Livia Schubiger and Stefanie Walter for their constant
guidance throughout a long writing process and to Johannes Kunz for his important feedback on
the methods applied. I would also like to thank my colleagues in Lausanne, Flavia Fossati, Fabienne
Liechti, Delia Pisoni, and Philipp Trein, as well as the participants of the 2016 Conference of the
Swiss Political Science Association for their very helpful comments on previous versions of this
paper.
*IDHEAP, University of Lausanne
daniel.auer@unil.ch
2
Abstract
Throughout the last decades, the United Nations has been continuously involved in attempts to
settle civil wars and to provide external assistance for post-conflict development. However, the
effectiveness of these engagements has been questioned within the public sphere as well as in the
scientific literature. The purpose of this paper is to shed new light on the effect of UN interventions
on a country's post-conflict democratization. The theoretical argument acknowledges that external
intervention allows war-torn societies to resolve commitment problems and to initiate a
democratization process with external support. The consent of the main parties involved thus plays
a crucial role for an intervention to demonstrate its `democratization capacity'. Using weighted least
squares regression, the results show a large positive effect of consent-based UN interventions on
post-conflict democratization. UN enforcement and missions that lack a UN mandate show no
effects or even negative effects. Furthermore, the positive effect of UN interventions on post-
conflict democratization decreases with the intervention's level of intrusion.
Keywords: civil war; third-party interventions; post-conflict democratization
3
Introduction
The promotion of democracy has become increasingly pivotal for international efforts in the
aftermath of intrastate conflict, with the United Nations (UN) at its forefront. Since 1948, the UN
has sent soldiers, diplomats, observers, and other specialists into various centres of conflict to
prevent belligerents from fighting, rebuild infrastructure, and establish state authorities (United
Nations, 2014). Their deployment is legitimized by the promise of post-conflict democratization
as part of the UN's grand strategy:
`We recommit ourselves to actively protecting and promoting all human rights, the rule of law and
democracy and recognize that they are interlinked and mutually reinforcing and that they belong
to the universal and indivisible core values and principles of the UN […]' (General Assembly of
the United Nations, 2005: 27, §119).
In fact, democratization has consistently been at the centre of UN peace missions (Fortna and
Howard, 2008). From a normative perspective, the UN considers a democratic regime that can be
integrated into the structure of a `global democracy' desirable (Archibugi et al., 2012).
However, the history books are studded with cases of bitter failure, such as the disaster in Rwanda
in 1994. The question remains whether the UN can live up to its own expectations of raising a
country's level of democratization in the aftermath of a conflict.
This paper elaborates on a theoretical argument that regards the presence of legitimate mediating
forces as conducive for post-civil war democratization. Even if war-torn societies have an interest
in striving for democracy, sustainable settlement and the readiness to (re-)distribute state power
non-violently through elections are often constrained by commitment problems and uncertainty.
For (former) belligerents, it is almost impossible to credibly convey their good intentions of abiding
by an agreement without exposing themselves.
4
The findings suggest that a UN mission can exert its `democratization capacity' if it receives the
consent of the main parties involved. Under such circumstances, the UN's presence is conducive
for a war-torn country's post-civil war democratization. However, a higher level of intrusion or
enforcement on behalf of the UN operation can interfere with what is always a fragile process of
shifting from conflict to sustainable development and democratization.
The remainder of this study proceeds as follows. The next chapter briefly reviews relevant treatises
on UN peace missions to which this study contributes. Chapter 3 discusses the theoretical
foundation of this paper and elaborates an argument stating that external interventions require the
consent of the main parties involved to successfully provide support for post-conflict
democratization. Chapter 4 presents the data and the estimation strategy. The results are presented
in Chapter 5, and Chapter 6 concludes.
Previous Findings
Much of the existing literature discusses the nexus between ending a war and post-conflict
development (e.g., Abbott and Snidal, 1998; Fearon, 2004; Fortna, 2004a; or Werner and Yuen,
2005). The majority of these investigations agree that democratization is a key element for
sustainable peace. The main obstacles are summarized in the security dilemma terminology, a
situation in which parties are unable to agree on democratization or even on ending violence.
Although the findings have been questioned, they tend to credit UN involvement with a positive
impact on settling civil wars peacefully (e.g., Gurses and Mason, 2006; Jervis, 1978; Metternich,
2011). A prerequisite is the UN's capability to both enforce negotiated settlements and to provide
a transitional authority (e.g., Joshi, 2010).
5
A growing number of studies has investigated the role of external interventions with regard to
actively promoting democratization (e.g., Bueno de Mesquita and Downs, 2006; Doyle, 2005;
Fortna and Huang, 2009; Gundel, 2002; Heldt, 2011; Jung, 2008; or Zartman, 1995). For instance,
Joshi (2010) and Wantchekon (2004) argue that the UN plays a central role in supervising joint
disarmament and mediating state building. Relatedly, Sambanis (2008: 9) stresses the positive
contribution of peacekeeping missions to the ‘quality of peace’, that is, to securing more than the
mere absence of war. However, he holds that the UN cannot extend the duration of peace as the
impact of interventions dissipates over time. When troops start leaving the country or warring
parties adapt to aggravated conditions for violent actions, commitment problems may once more
pose a threat to stability if democratic institutions have not yet become effective.
Pickering and Peceny (2006: 555) critically argue that the UN's higher `success rate' in terms of
post-conflict democratization compared to other third parties might be explained by the timing of
deployment. The likelihood of succeeding in sustaining peace could be higher when a third party
intervenes after a long and costly war than when it attempts to prevent two enthusiastic groups
from clashing. This hypothesis is consistent with findings by Doyle and Sambanis (2000), who
initially postulate a positive correlation between multidimensional UN peacekeeping and
democratization after civil war. However, they limit their statement insofar as the absence of a
peace treaty can reduce the effectiveness of a UN intervention in assisting institutional and political
reforms.
Although democracy promotion has become more prominent in post-Cold War UN interventions,
investigations by Paris (2004) and Richmond and Franks (2009) suggest that the UN might still be
unable to achieve more than negative peace, or the mere absence of war. Similarly, Metternich
(2011) finds that only those interventions sent by international organizations and with a clear
democratization mandate are capable of shortening conflict and implementing power-sharing
arrangements. This finding is in line with a recent study by Steinert and Grimm (2015), who found
6
a positive effect of UN peacebuilding missions that ‘include democracy promotion in their
mandates’ (Steinert and Grimm, 2015: 513). The authors distinguish UN missions with a low level
of intrusion (observer missions), those with an enforcement element, and traditional peacekeeping
missions from multidimensional efforts that explicitly promote democratization. Their findings
suggest that a country has a higher likelihood of increasing its level of democratization after a war
if a multidimensional mission is present.
According to Gleditsch et al. (2007), only democratic countries that intervene as third parties are
capable of promoting post-war democracy. However, these authors stress that interventions are
successful in the short run only. One possible explanation is provided by Werner and Yuen (2005),
who postulate that third-party interventions are confronted with a sensitive framework of
competing interests. If the (electorally) imposed allocation of state power does not reflect groups'
particular military power, the country runs the risk of returning to war as soon as the intervener
leaves. Similarly, Greig and Diehl (2005) stress that an intervention by the UN (or other third
parties) can ‘disrupt more natural resolution processes [or] create perverse incentives, especially
when involved in military deployments’ (in Beardsley, 2012: 335).
Contemporary research frequently emphasizes the need to distinguish between third-party
interventions according to their strength or their mandated goals. Further, the historical context
(e.g., post-colonial or Cold War period) as well as the region in which a civil war is located should
be considered in addition to the intervention troops' affiliation with a nation or institution. In
general, most studies highlight the superiority of the UN with regard to credible commitment and
institution-building expertise (e.g., Paris and Sisk, 2007).
This paper links the consent of involved parties with regard to the deployment of an external
intervention with the success of such missions in terms of post-conflict democratization. The paper
investigates the effect of different types of interventions: those mandated by the UN and non-UN
missions. Hultman et al. (2013) argue that previous studies differed in their results because they did
7
not capture the size of missions or the various types of UN personnel deployed. Similarly, Beardsley
(2012) finds ambiguous effects of different intervention types on their respective ability to manage
crises.
In addition, the majority of studies investigate post-conflict democratization at one point in the
aftermath of a civil war. For instance, Steinert and Grimm (2015) measure changes in a country's
level of democratization 5 years after the end of a civil war, whereas Bueno de Mesquita and Downs
(2006) base their inference on a country's democracy level after 10 post-war years. This paper
applies an alternative estimation strategy and measures the average effects of external interventions
throughout each year of a 10-year observation period, hence establishing a more complete picture
of short-run effects and longer-term trends.
An intervener's `democratization capacity'
Efforts to settle a civil war peacefully and to foster democratization are more often than not
doomed to failure, even if the (former) combatants are, in principle, willing to negotiate a peace
agreement and to re-establish an institutional framework. In fact, Wantchekon and Neeman (2002)
show that societies often have strong incentives to strive for democratization (see also Hartzell and
Hoddie, 2003).
However, conflict settlement and the readiness to (re-)distribute state-power non-violently through
elections is often constrained by commitment problems and uncertainty as groups are primarily
concerned about the adequate representation of their interests (e.g., Hartzell et al., 2001; Jervis,
1978: 167; Joshi, 2010: 828). In the process of demobilizing and disarming during a tense phase of
shifting from battles to negotiations, groups risk losing the ability to take part in establishing the
institutional framework, which allows their opponents to concentrate (military and civil) state
8
power. Thus, both sides must ensure that they will not be excluded from a yet-to-be-established
government. Even when a group is democratically legitimized, it lacks concrete guarantees that an
opposing group will refrain from misusing its authority and threaten the interests of the former
group once the latter group comes into state power.
This security dilemma often prevents belligerents from laying down their arms and entrusting the
state with protective duties. For instance, Fearon (2004) and Powell (2006) stress that commitment
problems are the key factor in a civil war's duration. Many civil wars are prolonged simply because
of a lack of guarantees in terms of protection of life and belongings (Kalyvas, 2007 and 2008) and
in terms of the possibility of participating in a newly formed government. Only if both sides obtain
these guarantees can a settlement be successful.
This argumentation is also supported by empirical evidence. Walter (2002), for instance, finds that
parties seem to be willing to negotiate peace more often than not. However, in several instances
throughout history, belligerents were unable to find common grounds and resumed fighting after
they agreed in principle to settle their violent conflict.
The presence of an external intervener can lead to the desired climate for negotiations in which the
payoffs from executing a settlement's terms exceed the payoffs from deflecting. A key factor lies
in the intervener's capability to minimize risk when parties enter the peace process (e.g., Hartzell
and Hoddie, 2007: 88). Parties can then rely on the intervener's protection and its ability to identify
and prevent any major violations of agreements. As both sides are aware of the gainless outcome
of such a deflection, each side's commitment receives enhanced credibility, which in turn increases
the likelihood of cooperation (Walter, 1999: 135f). In addition, with the support of the intervener's
expertise, the country can initiate elections and begin to establish an institutional framework that
guarantees security and the transparent distribution of state power. `Without those institutions, the
promise of security of property for citizens or political rights for the warring factions will not be
realized, and the choice of democracy will not be validated' (Wantchekon 2004: 22).
9
Hence, an intervener's `democratization capacity' consists, in principle, of its institution-building
capacity, ranging from the capability to initiate and police elections to the provision of interim
authorities, such as local jurisdiction. In addition, an intervention must provide the necessary means
to overcome warring parties' security concerns. Above all, the intervener must establish buffer
zones between factions, monitor demobilization and disarmament, and possibly restructure the
national army and police force. Importantly, the intervener's actions should be transparent. Often,
relevant political actors and (former) belligerents are not only many in number, but they also
interact within a complex system that is shaped by non-transparency (Marks, 1992: 397f) precisely
because the system lacks necessary institutional structures. Third-party involvement minimizes
opponents' perceived risk (Doyle et al., 1997; Stedman, 1997; Walter, 2002).
However, a key prerequisite is that the main parties agree in principle to invite an external mediator.
An agenda that targets democratization can only be effective if an intervention enjoys the consent
of the main parties. In that sense, the argument of Sambanis and Schulhofer-Wohl (2008) can be
adapted such that an intervener's impartiality transitions into the consent of the warring parties
under the assumption that they will agree to the deployment and abide by the rules of an impartial
external authority.
When considering the external interventions under UN mandate, one can broadly distinguish
between peacekeeping missions, ranging from mediation efforts to the deployment of neutral
civilian and military personnel, and peace enforcement operations (e.g., Doyle and Sambanis, 2006).
The key difference is the fact that the latter are usually deployed without the consent of the main
parties involved. Consent-based peacekeeping operations ‘[require] a commitment by the parties
to a political process. Their acceptance of a peacekeeping operation provides the UN with the
necessary freedom of action, both political and physical, to carry out its mandated tasks’ (United
Nations, 2008: 31f). Another difference lies in the general timing of deployment. Although
peacekeeping missions are usually present after fighting has been halted, peace enforcement
10
typically conveys a preceding coercive phase to restore peace, however fragile (United Nations,
2008: 18). However, peace enforcement eventually shifts into the same field of operations as
peacekeeping—that is, to initiate sustainable and self-enforcing peace through democratization
(e.g., Richmond, 2004).
Non-UN interventions involve a complex group consisting of actors with diverging interests and
capabilities. Scholars are often compelled to accredit democracy-promotion capacity to the UN
more than to other third-party interventions. For instance, according to Coicaud (2007), the United
States has never eliminated the tendency to instrumentalizing conflicts for its self-interest when
intervening (see also Ikenberry, 2000). Fortna (2004b) stresses that interventions led by the US are
likely to founder in promoting post-conflict democratization. Similarly, other regional powers
cannot decisively convey their role as legitimate, even-handed referees because of their regionally
bound involvement, which strengthens the suspicion of self-interest-based motivations (e.g.,
Pickering and Peceny, 2006: 540). Consider, for example, the presence of British troops during the
Northern Ireland conflict between 1969 and 2007, which were referred to as an occupation army
(e.g., Tonge, 1998: 94). In contrast to those interventions, the UN seems to obtain the necessary
capabilities that Betts (1994: 28f) describes as `imperial impartiality'. However, all types of external
interventions lack legitimacy as long as the parties do not agree to their deployment.
In a more graduated manner, legitimacy in the form of consent is a key requirement for `norm
diffusion` or `norm localization` (e.g., Acharya, 2004, 2012), which in turn represents a cornerstone
of sustainable democratization (e.g., Hyde, 2011). External actors that attempt to foster
democratization in a war-torn country confront a thin line between acceptance of and resistance
to their promotion efforts. If they are too aggressive in their demands to establish new institutions
and apply new rules, resistance could prevent successful democratization. Intuitively, the more
intrusive an intervention is, the more likely it is to founder when faced with such resistance.
11
Thus, this paper's key hypothesis can be formulated as follows:
H1: An external intervention equipped with the main parties' consent is able to increase a country's
post-conflict level of democratization.
H2: The positive effect of external interventions on a country's post-conflict level of
democratization decreases with the intervention's level of intrusion.
Data
The selection of cases includes every outbreak of intrastate violence, with two key restrictions to
this broad selection:
- Third-party units are usually deployed in conflicts that experience major violence (United
Nations, 2014). Hence, restricting the sample to such cases allows for a comparison of a
relatively homogeneous sample. The main identifying criterion for major civil war in this
data is the threshold of at least 1,000 battle-related deaths per year.
- The sample was restricted to countries that were sovereign states over a period of at least
one year prior to the civil war to obtain a measure of their pre-war democracy levels,
thereby excluding wars of independence.
The core information on intrastate conflicts stems from a comprehensive dataset generated by
Doyle and Sambanis (2000). It originally includes 124 major civil wars from 1944 to 1997 (and to
1999, if ongoing). For the present investigation, if a civil war was coded as ongoing and ended
before 2003, the parameters have been adjusted accordingly. Table IV in the Appendix presents a
full list of major civil wars in the sample according to their respective intervention status.
12
Additional data capturing general country information are provided by the Correlates of War
Project `State System Membership' data (Correlates of War Project, 2011) as well as the UN's online
presence (United Nations, 2013).
External interventions
To recap the research question, countries that experience major civil war should be better off in
terms of their respective post-conflict level of democracy if an external intervention is deployed
with the consent of the main parties.
To measure the impact of different types of interventions on democratization, each type is defined
as a binary treatment variable D={0,1}.
Peace missions mandated by the UN are legitimized under either Chapter VI or VII of the UN
Charter. Chapter VI comprises consent-based missions. Following the usual classification
according to the level of intrusion, this group consists of mediation, observer, and peacekeeping
operations. The latter can be further divided into traditional and multidimensional peacekeeping
missions.
First, mediation describes the least intrusive form of UN interventions. Mediation describes the
UN as an intermediary targeting the negotiation of peace agreements and the promotion of
reconciliation. For this analysis, the mediation category further includes the presence of UN envoys
and the sending of fact-finding missions. All of these require the consent of the (main) parties
involved (United Nations, 2014).
Second, UN observer missions usually ‘monitor a truce and help negotiate a peace through the
presence of military and civilian observers’ (Doyle and Sambanis, 2000: 781). As for mediation
missions, observer mandates require the consent of the main parties.
Third, traditional and multidimensional peacekeeping operations (PKO) are defined as the
deployment of both military and civilian personnel to guarantee demobilization and to guard buffer
13
zones, eventually facilitating conflict settlement. Multidimensional PKOs further entail additional
efforts to ‘build self-sustaining peace’ (Doyle and Sambanis, 2000: 781), such as the (re-
)construction of the economy and institutional transformation. Steinert and Grimm (2015) define
multidimensional PKOs as those that explicitly entail democracy promotion components. Again,
PKOs require the consent of the main parties involved (United Nations, 2014).
In contrast, enforcement missions are authorized under Chapter VII of the UN Charter and
involve robust military force intended to restore public order (e.g., Fortna, 2004b). Notably,
enforcement does not require the consent of the target county's government or of any other major
party to the conflict.
An intervention can be led and executed by a non-UN actor, such as French troops in the Chadian
war (1965 – 1979) or the United Kingdom, Mozambique and other African countries in the context
of the war in Zimbabwe (1972 – 1980).
To test this article's hypothesis, I group these missions into UN consent-based operations
(mediation, observer, traditional and multidimensional peacekeeping), UN operations lacking the
consent of the main parties (enforcement missions), and interventions not mandated by the UN.
1
In addition, I provide an analysis of each of the above-mentioned intervention types separately.
Operationalization of the Democracy Concept
Democratization is operationalized with Unified Democracy Scores' (UDS) point estimates for a
country's average level of democratization for any given year (see Pemstein et al., 2010). UDS
represents any overarching continuous measure of democratization stemming from 10 different
indices.
2
Similar to the most prominent measures of democratization, the synthesized UDS provide
1
If both UN units and another third party have been present, a UN mission is coded (Doyle and Sambanis, 2000).
This aggregation is congruent with Sambanis and Schulhofer-Wohl's (2008) findings that non-UN interventions
complement UN operations in peace-building efforts.
2
UDS incorporates information from Arat (1991), Bowman et al. (2005) (BLM), Bollen (2001), Freedom House (2007),
Hadenius (1992), Przeworski et al. (2000) (PACL), Polity scores by Marshall and Jaggers (2006), Polyarchy scale by
Coppedge and Reinicke (1990), Gasiorowski's (1996) Political Regime Change measure (PRC), and Vanhanen (2003).
14
values for democracy for each country-year between 1946 and 2012 in a continuous format. Basing
an inference on a continuous democracy measure implies that the results presented subsequently
do not capture the thresholds of regime types. That is, such a format allows inferences to be drawn
regarding whether a country managed to progress on a continuous scale of democratization instead
of assessing whether the country shifted, for instance, from autocratic to democratic rule or vice
versa.
Estimation Strategy
I adopt a weighted least squares estimator assuming conditional exogeneity
(Khandker et al. 2010:37, 64f).
A fully efficient estimator of ATT using weighted least squares (Hirano et al., 2003) can be written
as follows:
using alternative weighting based on a logistic estimation of the propensity score of receiving an
intervention with weights of 1 for countries that received an external intervention and weights of
for the control group (i.e., countries without intervention).
captures the change
in country i's level of democratization calculated as the first difference between the 1st and the 10th
post-conflict year t={1;10} and the level of democratization prior to the civil war. This pre-conflict
level of democratization should reflect the baseline from which a country drifted into war. To
minimize possible short-term variation, it is captured by the 10-year average UDS score prior to
the civil war.
3
With regard to the post-conflict period, the levels of democratization are captured
over 10 consecutive years. Thus, the measure captures a country's relative position in the sample
All individual indicators base their understanding of democracy more or less strictly on Dahl's (1971, 1989, 1998)
seminal elaborations.
3
The baseline level of democratization captured by the last year prior to war has been implemented, for instance, by
Ho et al. (2007).
15
with regard to its level of pre-war democratization, with the first differences calculated as
.
A mission deployed towards the end of a civil war might have been successful in fostering (at least
rudimentary) institutional redevelopment within one year after the conflict's relaxation. However,
a certain time frame should be considered necessary before power-sharing institutions can fully
evolve. For example, conducting elections takes careful preparations and time. Additionally,
considering distinctive ATTs over 10 post-conflict periods should guarantee the minimization of
possible short-term effects of third-party missions, such as a war party lying low as long as the
external actor is present and the conflict receives international attention.
captures the presence (1) or absence (0) of an external intervention, where the type of
intervention is adapted to each set of estimations accordingly.
Finally, describes the vector of covariates included in the model, which can be divided into four
groups: country characteristics, the pre-war socio-economic situation, war-related factors, and
general parameters.
Aspects such as the region in which a country is located as well as the terrain have been found to
be predictors of different post-war outcomes. To take these important characteristics into account,
I control for [i] the region, [ii] the logarithmic function of the extent of mountainous terrain within
the country, [iii]} the log of the country's size, and [iv] the country’s number of borders with
neighbouring states.
In addition, the socio-economic situation of a country prior to the outbreak of civil war influences
not only the conflict itself but also the events in its aftermath. Many authors regard the economic
environment prior to a war as the decisive element for post-conflict developments (e.g.,
Przeworski, 2005). Hence, I control for [v] the log of the pre-war GDP per capita in current US$
and [vi] the log of life expectancy in the years prior to conflict. Because democratic legacies may
influence post-conflict developments, I further control for [vii] the 10-year average level of
16
democratization prior to the war. Thus, I also consider that a certain type of intervention could
have been deployed only in countries with an exceptionally high or low pre-war level of
democratization.
War-related factors are captured, including [viii] the type of war, namely, whether it was an
ethnic/religious/identity war or an ideological/revolutionary conflict.
4
Further, I control for [ix]
the duration of the war in years, [x] the size of the government army in thousand soldiers, [xi] the
outcome of the war (i.e., ongoing/government victory/rebel victory/informal truce/peace treaty),
and [xii] whether a major power was involved in the war. Major powers include the five permanent
member of the UN Security Council, as long as they did not participate under a UN-mandated
operation (see Doyle and Sambanis, 2000).
General parameters control for [xiii] the country's ethnic composition through Vanhanen's index
of ethnic heterogeneity (Vanhanen, 1999, 2003) and [xiv] whether the country experienced another
war within 10 years prior to the conflict of interest.
5
To be able to control for systemic constraints,
both [xv] a decade variable and [xvi]} a dummy variable indicating whether the conflict took place
during or after the Cold War have been included. The changing approach of international actors
deploying external interventions (e.g., Paris and Sisk, 2007) should be controlled for in the same
manner as the global dynamics of democratization. Most scholars identify the end of the Cold War
as a watershed for United Nations interventions, attributing a broader, more proactive role and a
greater legitimacy to post-Cold War missions (e.g., Paris, 2004; Richmond, 2006).
Weights are based on a logistic regression estimating the probability of receiving an external
intervention. The logit estimation is performed using the same set of covariates as described for
the WLS estimator, excluding the duration of the war (ix), the size of the government army (x) and
the outcome of the war (xi). Including these parameters could result in a post-treatment bias (i.e.,
whether an intervention was deployed is likely to affect war outcomes and their duration) as well
4
For a more detailed description, see Doyle and Sambanis (2000) as well as their work's data notes.
5
Hegre and Sambanis (2006) identify the time since a previous war as a key variable to explain civil war onsets.
17
as the government army's size, assuming that demobilization is one key goal of external
interventions.
Table I about here
The descriptive statistics in Table I above show that countries that receive different types of
interventions or no intervention at all are, on the whole, rather similar. As an exception, non-UN
interventions have been deployed in conflicts that lasted considerably longer, although they
showed, on average, the lowest level of democratization while the conflict was ongoing. Of the
three cases that had a UN enforcement operation, only one took place during the Cold War, all of
them had a UN Security Council member involved, and the government army's size was, on
average, the smallest. Interestingly, countries that did not receive an external intervention had the
lowest pre-war GDP per capita, whereas consent-based UN missions were predominantly deployed
when the war ended in an informal truce or a peace treaty, which is rather unsurprising.
Figure 1 about here
Figure 1 plots the average level of democratization by intervention status 10 years prior and after
the major civil war. At first glance, the difference in democracy levels between conflicts with an
intervention and those without (independently of its type) seems to be substantial. Although both
pre-war averages circulate slightly below -0.5 UDS, countries with an external intervention are able
to lift their level of democratization to approximately -0.25 and increasing, whereas those without
any external intervention remain at approximately -0.5.
18
Results
The following paragraphs present the average treatment effects on the treated (ATT) for different
intervention types and for each post-conflict year. Interpreting the results translates
straightforwardly into `on average, how would a country that received an intervention of type X
have fared if it had not had this intervention?'
Restating the paper's hypothesis, one would expect a positive effect ($ATT\geq0$) for consent-
based missions; that is, countries in which the warring parties accepted external support should
have effectively profited from an intervention's `democratization capacity' and increased their post-
conflict democracy level relative to their pre-war level. The logistic regression of receiving an
external intervention, determining the weights for each country, is shown in the Appendix, Table
Iv. The accompanying propensities are presented in the Appendix, Figure 4.
Figure 2 about here
First, Figure 2 compares the ATT of consent-based UN missions with UN enforcement
operations. The difference is unambiguous. Whereas consent-based UN missions are capable of
raising democracy levels significantly throughout the entire observation period, enforced UN
efforts may even show a negative sign despite being statistically insignificant. Recalling the range
of pre-war democracy levels in Table I, one can infer that a treatment effect between +0.2 and
+0.4 is exceptionally high. This means that a country that experienced a major civil war and whose
warring parties agreed to a UN peacekeeping mission achieved a substantially higher level of
democratization than if the UN mission had been absent in the same country. On the contrary, in
cases in which the UN enforced its presence, the same country would not have been worse off, at
best, if the UN mission had been absent.
19
Table II about here
Table III about here
The full regression results are presented in Tables II and III above. As shown in Figure 2, the effect
of consent-based UN interventions is large in magnitude and statistically significant throughout the
entire observation period (Table II). In contrast, UN enforcement operations are associated with
a negative effect in almost all periods, but the coefficients are statistically insignificant, except the
first and second year after the major civil war has ended (Table III).
Both models (Tables II and III) show that conflicts located in the Middle East and North Africa
decrease the difference between the pre- and the post-conflict democracy level (line 3). Similarly,
Cold War conflicts (line 6) and previous wars within 10 years (line 7) have a negative influence. In
contrast, a higher life expectancy (line 13) as a proxy for pre-conflict development increases
democratization after the civil war. Interestingly, the same holds for the duration of the civil war
(line 16), although the magnitude of the effect is limited. If a UN Security Council member was
involved in the conflict outside a UN mandate (line 17) and if the war ended with a government
victory (line 19), the post-conflict level relative to the pre-war level of departure decreases. For
both models and all observation periods, a previously higher level of democratization (line 23) is
associated with higher levels after the civil war has ended.
In summary, one can infer that UN interventions are, on average, capable of increasing a country's
post-conflict level of democratization if the mission enjoys the consent of the main parties
involved. Therefore, Hypothesis 1 cannot be rejected.
Figure 3 about here
20
Figure 3 plots the effects of the five different types of UN interventions, classified according to
their level of intrusion (see Doyle and Sambanis, 2000). In addition, the effect of non-UN
interventions has been plotted. An overview of the particular intervention regression coefficients
can be found in the Appendix, Table VI.
For less intrusive UN mediation and observer missions, the effects are positive and large
throughout the entire observation period and statistically significant for the larger part. The effect
sign of multidimensional PKOs remains above the zero line, although the results are statistically
insignificant. However, there is good reason to believe that low levels of statistical significance are
mainly driven by a small number of observations. Hence, it seems that multidimensional UN PKOs
have a positive yet limited effect on post-conflict democratization relative to the countries' pre-war
levels. The picture is even more ambiguous for traditional PKOs, whereas UN enforcement
operations and interventions deployed without UN mandate show a negative effect. Recalling that
the presented average treatment effects on the treated indicate the difference in post-war
democracy levels if an intervention were absent in the same country, the results are particularly
striking despite little statistical significance. Intuitively, democracy promotion might not be the
primary goal for more intrusive UN interventions. However, considering post-conflict
democratization alone, a country might be better off without the UN's support. This result is
sobering for democracy promotion through more intrusive or enforced UN interventions.
Furthermore, the results show that interventions without UN mandates clearly fail with regard to
fostering post-conflict democratization. On the one hand, this might be the case because non-UN
missions lack the commitment and capability to promote democratization. On the other hand,
these missions are likely to lack the parties' consent, such as the United Kingdom intervention in
Northern Ireland, as mentioned above.
21
Assuming that this categorization of external interventions reflects the ordered level of intrusion,
Hypothesis 2, stating that democratization success decreases with an intervention's level of
intrusion, cannot be rejected.
Two robustness checks were performed with similar results. First, ordinary squares regressions
were performed to consider the possible mis-weighting of cases. The results for equally weighted
cases can be found in the Appendix, Table VII.
Second, applying a semi-parametric approach in the form of kernel matching acknowledges the
critique that WLS/OLS standard errors and t-statistics might be biased in analyses with very small
numbers of observations. The matching results are presented in the Appendix, Table VIII.
Interestingly, the decreasing positive effect of UN interventions disappears. In fact,
multidimensional PKO turns out to be one of the more effective mission types. In general, the
results suggest a higher average effect of external interventions, increasing even non-UN
interventions slightly above zero. Although these results support this paper's main analysis, they
might over-estimate the `true effect' of external interventions because the propensity scores used
for assigning matches are based on the probability of receiving a specific type of intervention. For
instance, UN observer missions are compared to cases without any intervention based on the
propensity of receiving a UN observer mission. Although cases that received another type of
intervention are ruled out to function as a control group, this approach might fail to take into
account that the propensity of receiving a specific intervention is, in reality, also dependent on the
possibility of receiving interventions of different types.
22
Concluding remarks
This article has shown substantially positive effects of UN-led interventions on a country's post-
conflict level of democratization if the intervention enjoys the parties' consent and is characterized
by a low level of intrusion. Traditional and multidimensional UN peacekeeping operations seem to
be conducive, but to a lesser extent. UN enforcement and peacekeeping missions without the UN's
mandate have a negative effect on a country's post-conflict democratization.
Of course, it could be the case that the effects found here are driven by an unobserved factor that
renders democratization substantially different in cases of interventions with higher levels of
intrusion. However, because most of the `usual suspects' that shape a conflict were controlled for
in the analysis, it is not plausible that this is the case.
Consequently, the mere enforcement of peace seems to be a less fruitful strategy for democracy
promotion. Warring parties must be `ready' to accept external support. Only then can an external
intervention demonstrate its `democratization capacity'.
Most importantly, the shift to a higher level of democratization is maintained throughout a 10-year
period after the end of civil war. Opposing groups seem to be willing and able to uphold
democratization once they are or have been supported by an external force in overcoming security
dilemma and (re-)establishing institutions. This finding should be especially encouraging for UN
peacekeepers.
This article supports previous findings, stressing that in most cases, the deployment of UN units
should be regarded as positive, if not essential, for a country's post-conflict democratization.
However, the findings restrict the positive influence to consent-based types of interventions.
Success decreases with a higher level of intrusion. In general, expecting a fully-fledged democracy
in the aftermath of a civil war might be too much to expect. However, the UN's commitment to
democratization is much more than cheap talk.
23
References
Abbott KW and Snidal D (1998) Why states act through formal international organizations. Journal
of Conflict Resolution 42(1): 3-32. DOI: 10.1177/0022002798042001001.
Acharya A (2004) How ideas spread: Whose norms matter? Norm localization and institutional
change in Asian regionalism. International Organization 58(2): 239-257. DOI:
10.1017/S0020818304582024.
Acharya A (2012) Local and transitional civil society as agents of norm diffusion. Paper Presented to
the Global Governance Workshop, Oxford, UK, 1-3 June 2012, pp. 1-11. Oxford: University of
Oxford.
Arat Z (1991) Democracy and Human Rights in Developing Countries. Boulder: Lynne Rienner.
Archibugi D, Koenig-Archibugi M and Marchetti R (2012) Introduction: Mapping global
democracy. In: Archibugi D, Koenig-Archibugi M and Marchetti R (eds) Global Democracy -
Normative and Empirical Perspectives. Cambridge: Cambridge University Press, pp. 1-21.
Beardsley K (2012) UN intervention and the duration of international crises. Journal of Peace Research
49(2): 335-349. DOI: 10.1177/0022343311431599.
Betts RK (1994) The delusion of impartial intervention. Foreign Affairs 73(6): 20-33. DOI:
10.2307/20046926.
Bollen K (2001) Cross-National Indicators of Liberal Democracy, 1950-1990. Ann Arbor, MI: Inter-
university Consortium for Political and Social Research.
Bowman K, Lehoucq F and Mahoney J (2005) Measuring political democracy: Case expertise, data
adequacy, and Central America. Comparative Political Studies 38(8): 939-970. DOI:
10.1177/0010414005277083.
De Mesquita BB and Downs GW (2006) Intervention and democracy. International Organization
60(3): 627-649. DOI: 10.1017/S0020818306060206.
Coicaud J.-M (2007) Beyond the National Interest - The Future of UN Peacekeeping and Multilateralism in
an Era of U.S. Primacy. Washington, DC: United States Institute of Peace.
Coppedge M and Reinicke WH (1990) Measuring polyarchy. In: Inkeles A (ed) On Measuring
Democracy: Its Consequences and Concomitants. New Brunswick, NJ: Transaction, pp. 47-68.
Correlates of War Project (2011) State system membership list, v2011. Available at:
http://correlatesofwar.org (accessed 12 January 2014).
Dahl R (1971) Polyarchy - Participation and Opposition. New Haven: Yale University Press.
Dahl R (1989) Democracy and Its Critics. New Haven: Yale University Press.
Dahl R (1998) On Democracy. New Haven: Yale University Press.
24
Doyle M (2005) Three pillars of the liberal peace. American Political Science Review 99(3): 463-466.
Doyle MW and Sambanis N (2000) International peacebuilding: A theoretical and quantitative
analysis. American Political Science Review 94(4): 779-801. DOI: 10.2307/2586208.
Doyle M and Sambanis N (2006) Making War and Building Peace: United Nations Peace Operations.
Princeton: Princeton University Press.
Doyle M, Johnstone I and Orr R (1997) Strategies for peace: Conclusions and lessons. In: Doyle
MW, Johnstone I and Orr RC (eds) Keeping the Peace - Multidimensional un Operations in
Cambodia and el Salvador. Cambridge: Cambridge University Press, pp. 369-392.
Fearon JD (2004) Why do some civil wars last so much longer than others? Journal of Peace Research
41(3): 275-301. DOI: 10.1177/0022343304043770.
Fortna VP (2004a) Does peacekeeping keep peace? International intervention and the duration of
peace after civil war. International Studies Quarterly 48(2): 269-292. DOI: 10.1111/j.0020-
8833.2004.00301.x.
Fortna VP (2004b) Peace Time: Cease-Fire Agreements and the Durability of Peace. Princeton: Princeton
University Press.
Fortna VP and Howard LM (2008) Pitfalls and prospects in the peacekeeping literature. Annual
Review of Political Science 11(1): 283-301. DOI: 10.1146/annurev.polisci.9.041205.103022.
Fortna V and Huang R (2009) Democratization after civil war. Presentation at: The Annual Meeting
of the American Political Science Association, Toronto, 3-6 September 2009.
Freedom House (2007) Freedom in the World. Available at: http://www.freedomhouse.org
(Accessed 20 June 2015).
Gasiorowski MJ (1996) An overview of the political regime change data set. Comparative Political
Studies 29(4): 469-483. DOI: 10.1177/0010414096029004004.
General Assembly of the United Nations (2005) World summit outcome. Resolution adopted by
the General Assembly. 60/1, 24 October. New York: United Nations.
Gleditsch N, Christiansen L and Hegre H (2007) Democratic Jihad? Military Intervention and
Democracy. World Bank Policy Research Working Paper 4242.
Greig J and Diehl P (2005) The Peacekeeping - Peacemaking Dilemma. International Studies Quarterly
49 (4): 621-645.
Gundel J (2002) Peacebuilding as politics: cultivating peace in fragile societies. Journal of Refugee
Studies 15(3): 334-335. DOI: 10.1093/jrs/15.3.334.
Gurses M and Mason T (2006) Democracy out of Anarchy: How Do Features of a Civil War Influence the
Likelihood of Post-Civil War Democracy? Denton, TX: University of North Texas.
Hadenius A (1992) Democracy and Development. Cambridge: Cambridge University Press.
25
Hartzell C and Hoddie M (2003) Institutionalizing peace: Power sharing and post-Civil War conflict
management. American Journal of Political Science 47(2): 318-332. DOI: 10.1111/1540-
5907.00022.
Hartzell C and Hoddie M (2007) Crafting Peace - Power-Sharing Institutions and the Negotiated Settlement
of Civil Wars. University Park, PA: Pennsylvania State University Press.
Hartzell C, Hoddie M and Rothchild D (2001) Stabilizing the peace after civil war: An investigation
of some key variables. International Organization 55(1): 183-208. DOI:
10.1162/002081801551450.
Hegre H and Sambanis N (2006) Sensitivity analysis of empirical results on civil war onset. Journal
of Conflict Resolution 50(4): 508-535. DOI: 10.1177/0022002706289303.
Heldt B (2011) Peacekeeping and transitions to democracy. In: Fjelde H and Höglund K (eds)
Building Peace, Creating Conflict? - Conflictual Dimensions of Local and International Peace-Building.
Lund: Nordic Academic Press, pp. 47-72.
Hirano K, Imbens GW and Ridder G (2003) Efficient estimation of average treatment effects using
the estimated propensity score. Econometrica 71(4): 1161-1189. DOI: 10.1111/1468-
0262.00442.
Ho DE, Imai K, King G et al. (2007) Matching as nonparametric preprocessing for reducing model
dependence in parametric causal inference. Political Analysis 15(3): 199-236. DOI:
10.1093/pan/mpl013.
Hultman L, Kathman J and Shannon M (2013) United Nations Peacekeeping and Civilian
Protection in Civil War. American Journal of Political Science 57 (4): 1-17.
Hyde SD (2011) Catch us if you can: Election monitoring and international norm diffusion.
American Journal of Political Science 55(2): 356-369. DOI: 10.1111/j.1540-5907.2011.00508.x.
Ikenberry G (2000) America's Liberal grand strategy: Democracy and national security in the post-
war era. In: Cox M, Ikenberry G and Inoguchi T (eds) American Democracy Promotion: Impulses,
Strategies, and Impacts. New York: Oxford University Press, pp. 103-126.
Jervis, R. (1978) Cooperation Under the Security Dilemma. World Politics 30 (2): 167-214.
Joshi M (2010) Post-civil war democratization: Promotion of democracy in post-civil war states,
1946-2005. Democratization 17(5): 826-855. DOI: 10.1080/13510347.2010.501173.
Jung J (2008) Mission impossible? Democracy building in post-civil war societies. Annual Meeting of
the Midwest Political Science Association, Chicago, 3-6 April 2008. Ithaca, NY: Cornell
University.
Kalyvas S (2007) Civil Wars. In: Boix C and Stokes SC (eds) The Oxford Handbook of Comparative
Politics. Oxford: Oxford University Press.
26
Kalyvas S (2008) Ethnic Defection in Civil War. Comparative Political Studies 41 (8): 1043-68.
Khandker S, Koolwal G and Samad H (2010) Handbook on Impact Evaluation. Washington DC: The
World Bank.
Marks G (1992) Rational sources of chaos in democratic transition. American Behavioral Scientist 35(4-
5): 397-421. DOI: 10.1177/0002764292035004005.
Marshall MG and Jaggers K (2006). Polity IV Project - Political Regime Characteristics and
Transitions, 1800-2010. Dataset Users' Manual. Vienna, VA: Center for Systemic Peace.
Metternich NW (2011) Expecting elections: Interventions, ethnic support, and the duration of civil
wars. Journal of Conflict Resolution 55(6): 909-937. DOI: 10.1177/0022002711408015.
Paris R (2004) At War's End - Building Peace after Civil Conflict. Cambridge: Cambridge University
Press.
Paris R and Sisk T (2007) Managing Contradictions: The Inherent Dilemmas of Postwar Statebuilding.
Research Partnership on Postwar Statebuilding. New York: International Peace Academy.
Pemstein D, Meserve SA and Melton J (2010) Democratic compromise: A latent variable analysis
of ten measures of regime type. Political Analysis 18(4): 426-449. DOI:
10.1093/pan/mpq020.
Pickering J and Peceny M (2006) Forging democracy at gunpoint. International Studies Quarterly 50(3):
539-559. DOI: 10.1111/j.1468-2478.2006.00413.x.
Powell R (2006) War as a commitment problem. International Organization 60(1): 169-203. DOI:
10.1017/S0020818306060061.
Przeworski A (2005) Democracy as an equilibrium. Public Choice 123(3-4): 253-273. DOI:
10.1007/s11127-005-7163-4.
Przeworski A, Alvarez M, Cheibub J et al. (2000) Democracy and Development: Political Regimes and
Economic Well-Being in the World. Cambridge: Cambridge University Press, pp. 1950-1990.
Richmond OP (2004) UN peace operations and the dilemmas of the peacebuilding consensus.
International Peacekeeping 11(1): 83-101. DOI: 10.1080/1353331042000228403.
Richmond OP (2006) The problem of peace: Understanding the `liberal peace'. Conflict, Security &
Development 6(3): 291-314.
Richmond OP and Franks J (2009) Liberal Peace Transitions: Between Statebuilding and Peacebuilding.
Edinburgh: Edinburgh University Press.
Sambanis N (2008) Short- and Long-term effects of United Nations peace operations. The World
Bank Economic Review 22(1): 9-32. DOI: 10.1093/wber/lhm022.
Sambanis N and Schulhofer-Wohl J (2008) Evaluating multilateral interventions in civil wars: A
comparison of un and non-UN peace operations. In: Bourantonis D, Ifantis K and
27
Tsakonas P (eds) Multilateralism and Security Institutions in an Era of Globalization. New York:
Routledge, pp. 252-287.
Stedman SJ (1997) Spoiler problems in peace processes. International Security 22(2): 5-53. DOI:
10.1162/isec.22.2.5.
Steinert JI and Grimm S (2015) Too good to be true? United Nations peacebuilding and the
democratization of war-torn states. Conflict Management and Peace Science 32(5): 513-535. DOI:
10.1177/0738894214559671.
Tonge J (1998) Northern Ireland: Conflict and Change. New York: Routledge.
United Nations (2008) United Nations peacekeeping operations: Principles and guidelines
(capstone doctrine). United Nations Peacekeeping. Available at:
http://pbpu.unlb.org/pbps/Library/Capstone_Doctrine_ENG.pdf (accessed 7 August
2014).
United Nations (2013) Peacekeeping operations. United Nations peacekeeping. Available at:
http://www.un.org/en/peacekeeping/operations/ (accessed 10 October 2015).
United Nations (2014). Peacekeeping Fact Sheet Archive. Retrieved from United Nations Peacekeeping.
Available at:
http://www.un.org/en/peacekeeping/resources/statistics/factsheet\_archive.shtml
(accessed 5 August 2015).
Vanhanen T (1999) Domestic ethnic conflict and ethnic nepotism: A comparative analysis. Journal
of Peace Research 36(1): 55-73. DOI: 10.1177/0022343399036001004.
Vanhanen T (2003) Democratization: A Comparative Analysis of 170 Countries. New York: Routledge.
Walter BF (1999) Designing transitions from civil war: Demobilization, democratization, and
commitments to peace. International Security 24(1): 127-155. DOI:
10.1162/016228899560077.
Walter BF (2002) Committing to Peace: the Successful Settlement of Civil Wars. Princeton: Princeton
University Press.
Wantchekon L (2004) The paradox of `warlord' democracy: A theoretical investigation. American
Political Science Review 98(1): 17-33.
Wantchekon L and Neeman Z (2002) A theory of post-civil war democratization. Journal of
Theoretical Politics 14(4): 439-464. DOI: 10.1177/095162902774006822.
Werner S and Yuen A (2005) Making and keeping peace. International Organization 59(2): 261-292.
DOI: 10.1017/S0020818305050095.
Zartman W (1995) Putting things back together. In: Zartman W (ed) Collapsed States: the Disintegration
and Restoration of Legitimate Authority. Boulder, CO: Lynne Rienner, pp. 267-273.
28
Tables & Figures
29
30
31
32
33
34
Appendix
35
36
37
