Does War Influence Democratization?

Chapter 2, Mansfield & Snyder • 1

Chapter 2

Does War Influence Democratization?

Edward D. Mansfield

Jack Snyder

Contrarians have been fascinated by the idea that war can lead to progress.[1]In particular, there is a long tradition of viewing war as a midwife of democracy.For Americans raised on the history of the U.S. Revolutionary War, the idea of fighting for freedom seems intuitively plausible.More scholarly analyses note that prosecuting a war requires governments to raise armies and revenue, which places substantial demands on society.In exchange for meeting these demands, citizens in non-democratic countries have sometimes pressed for an expansion of the franchise, stimulating democratic reforms.

An opposing school of thought, however, notes that in order to wage war countries often militarize society, centralize power, and restrict civil liberties, thereby promoting the rise of authoritarianism and the establishment of garrison states.Although scholars have advanced these competing claims for well over half a century, they have not been thoroughly or conclusively tested.[2]The purpose of our chapter is to help fill this important gap in the literature.

We begin by conducting a statistical analysis of the influence of war on democratization spanning the nineteenth and twentieth centuries.We examine the effects of both interstate and other external wars on subsequent democratic transition, using four different measures of democratization.Our results provide no support for the view that war inhibits democratization; instead, they furnish some scattered support for the view that war promotes democratization.However, further historical analysis of the cases drawn from the statistical analysis casts doubt on the claim that war has any strong, systematic impact on regime change.In a number of cases – including Britain and Germany after World War I, Portugal after its colonial wars in Angola, Mozambique, and Guinea Bissau from 1961 to 1975, and Argentina after the Falklands War – war did stimulate democratization.But in many other cases comprising our sample, the association between war and regime change is more coincidental than causal.Taken together, the quantitative and narrative case evidence in this chapter provides little indication that war either stimulates or inhibits democratization.Our results suggest that the most powerful causes of democratization are not war but rather economic development, the character of the international neighborhood around the transitional state, and the legacy of the state’s prior political institutions.

The Links between War and Democratization

Some scholars have argued that war undermines democracy by creating a garrison state, providing excuses for the curtailment of liberties, and offering opportunities to revile the domestic opposition as traitors.[3]Others have argued the opposite.War, they claim, fosters democratization by breaking the power of elites, mobilizing the masses, and creating incentives for the state to bargain with the people it needs to contribute to the war effort.[4]

It may be that both positions are correct but under different circumstances. War might undermine democratic liberties by reinforcing conformity in the short run, but promote democracy in the long run by making elites more dependent on the support of energized mass groups.War might strengthen the absolutist state in continental powers that commandeer men and materiel for war through coercion, whereas it might strengthen democratic institutions in states that finance war through market loans secured by strict legal guarantees.[5]Short wars might bolster dictators seeking prestige through military victories, whereas long, costly wars or extended military rivalries that require the mobilization of vast military resources might stimulate democracy by creating a need for nationalism and popular government.[6]Victories might help incumbent elites, whereas defeats may sweep them from power and lay the foundation for democrats to obtain office.[7]International wars may rally the masses around the ruling elites, whereas civil wars may topple them.[8]

But these claims cannot be evaluated simply by logical argument and apt illustrations.A case could easily be made for the opposite of any of the propositions that we just laid out.Systematic tests supplemented by the tracing of specific causal mechanisms are needed to sort out these claims and counterclaims.

In this chapter, we focus on the general argument that war increases the likelihood of democratization by enhancing popular groups’ freedom to contend for political power, by expanding the scope of political participation, and by strengthening the institutions that make democracy work.[9]Specifically, we will look for the following proposed mechanisms, which have been highlighted in previous research:

(1)War breaks the power of anti-democratic elites, either sweeping them from power, sapping their resources, or discrediting them, and opens the door for more democratic forces.[10]

(2)War leads to the mobilization of mass groups that had previously played little role in politics or had been weakly organized, increasing their capacity for collective action, developing civil society networks, heightening their political consciousness, placing arms or economic resources in their hands, or expanding their repertoires of political action in ways that forge tools for successful democratization.[11]

(3)War induces anti-democratic elites to bargain with mass groups in order to gain their cooperation with the war effort, resulting in broader political inclusion and greater political power for pro-democracy mass groups.[12]

(4)War strengthens institutions and ideologies that facilitate democratic rule, including state administrative capacity, rule of law, economic capacity, national consolidation, or patriotism.[13]

Measuring Democratization

To test the effects of war on democratization, we need to develop a measure of regime change.Consistent with much existing research, we use the Polity IV data as well as indices developed by Ted Robert Gurr and his colleagues for this purpose.[14]This data set includes yearly information on a variety of domestic institutions for a very broad cross-section of countries from 1800 to 2004.The Polity data have been widely used in studies of international relations and comparative politics, and they cover a wider array of countries and a longer time period than any alternative compilation.[15]

Gurr, Keith Jaggers, and Will Moore combine annual measures of the competitiveness of the process through which a country’s chief executive is selected, the openness of this process, the extent to which institutional constraints exist on a chief executive’s decision-making authority, the competitiveness of political participation within a country, and the degree to which binding rules govern political participation to create 11-point indices of each state’s democratic (Democracy) and autocratic (Autocracy) characteristics.[16]The difference between these indices (Regime = Democracy - Autocracy) provides a summary measure of regime type that ranges from -10 to 10.Jaggers and Gurr define democracies as states where Regime > 6 and autocracies as states where Regime < -6.[17]The remaining states (that is, those where -7 < Regime < 7) are coded as mixed, or anocracies.

The relatively few statistical studies of the effects of political-military conflict on democratization have measured regime change using some variant of Jaggers and Gurr’s summary index.[18]However, in addition to this index, which we refer to as the composite index, there is also reason to analyze democratization along some of the specific institutional dimensions that make it up.Particularly important are the competitiveness of political participation, the openness of executive recruitment, and the extent of the constraints placed on the chief executive.We emphasize these factors and analyze them separately because each has been featured in the theoretical literature on war and democratization.Furthermore, these factors are not closely related.On average, the correlation between any given pair of measures of democratization is quite modest, indicating that they are not tapping the same institutional characteristics.[19]Finally, unlike these factors, the remaining variables that make up Regime are coded in a way that makes distinguishing between democracies and autocracies quite difficult.For each of these three institutional factors, a state’s regime type is assessed using the following coding rules.

The competitiveness of political participation is measured using a five-point scale.We code as autocratic those states characterized by what Gurr, Jaggers, and Moore refer to as “suppressed competition,” a category that includes totalitarian dictatorships, despotic monarchies, and military dictatorships in which no significant political activity is allowed outside of the ruling regime.[20]We code states characterized by “competitive competition” as democratic.In such states, competitive political groupings (usually political parties) are stable and enduring, and their competition rarely leads to violence or widespread disruption.We code as anocratic states falling into any of Gurr, Jaggers, and Moore’s three intermediate categories of the competitiveness of political participation (restricted/transitional, factional, and transitional competition).[21]Gurr, Jaggers, and Moore claim that “transitions to Competitive [i.e. full democracy] are not complete until a national election is held on a fully competitive basis.”[22]Based on this variable, distinguishing among autocracies, anocracies, and democracies is fairly straightforward.

The openness of executive recruitment is measured using a four-point scale.We code as autocratic those regimes with hereditary absolute rulers or with rulers who seized power by force.We code as anocratic those regimes with dual executives, in which a hereditary ruler shares authority with an appointed or elected governing minister.We code as democratic those regimes that Gurr, Jaggers, and Moore classify as having an open system of executive recruitment, regardless of whether the executive is popularly elected or selected through some other regularized process.We rely on the case studies conducted later in this chapter to identify any non-democratic “false positives” that might result from the use of this measure.

Institutional constraints on the chief executive are measured using a seven-point scale.We classify regimes as autocratic if the chief executive has unlimited authority or if the executive’s authority falls in an intermediate category whereby the institutional constraints faced by this individual are less than “slight to moderate.”[23]We classify regimes as democratic if “accountability groups [such as legislatures] have effective authority equal to or greater than the executive in most areas of activity,” or if the constraints on the executive are more than “substantial” based on the Polity scale.[24]Substantial constraints exist when the executive has more effective authority than the legislature, but the legislature can block appointments, funds, or bills proposed by the executive.Regimes in which executive constraints lie in the range between “slight” and “substantial” are classified as anocratic.

The sample analyzed in this paper includes all states coded as members of the interstate system by the Correlates of War (COW) Project during the nineteenth and twentieth centuries.[25]For each measure of regime type (the composite index and the three component indices), we measure democratization over five-year intervals. We analyze regime change over five years because, in various cases, the data needed to code a state’s regime type are missing for years immediately surrounding the change.As such, certain instances of democratization are omitted from the sample when very short intervals are analyzed, a problem that is ameliorated by considering the effects of transitions occurring over five-year periods.To measure democratization, we code each state, i, as democratic, autocratic, or anocratic in every year, t.We then measure i’s regime type in year t-5.Democratization is a variable that equals 1 if state i changes from either an autocracy or an anocracy in year t-5 to a democracy in year t.It equals 0 otherwise.[26]

In the following analysis, we code Democratization in two somewhat different ways to make sure that our results do not depend on which coding procedure is used.Over a given ten-year period from t-9 to t, for example, during which a state was non-democratic until year t-5 and then democratic from year t-4 onward, we would observe five instances in which Democratization equals 1: from t-9 to t-4, from t-8 to t-3, from t-7 to t-2, from t-6 to t-1, and from t-5 to t.One might consider only the initial instance to be a case of democratization since the regime change occurs in year t-4, or one might consider all five instances to be cases of democratization.Since we see no conceptual basis for choosing between these two procedures, we use both.We conduct one set of tests where only the first episode is considered a case of democratization (in this example, from t-9 to t-4).The following four cases are coded as missing.We also conduct a second set of tests after coding all five cases as instances of democratization.It is important to recognize that no democratizing country is omitted altogether, regardless of which procedure is used.The only difference based on these procedures is whether, for a given country, democratization is coded as a single event in the data set or an event occurring over a longer period of time.Estimating our statistical models using both coding procedures will help to assess the robustness of the following results.

It is also important to note that the Polity data include various yearly observations in which a country is coded as -66, -77, or -88 because its institutions are in flux, are difficult to code, or are controlled by a foreign power.We follow the Polity Project in transforming these “standardized authority codes” into values of Regime and each of the three component indices, thereby reducing the number of missing observations in our sample.[27]We refer to this as the Adjusted Polity procedure.[28]However, we also include a set of results in which the values -66,

-77, and -88 are treated as missing to ensure that transforming these values in the way that the Polity Project suggests does not unduly influence our findings.We refer to this as the Polity procedure.

Measuring War

Most studies of conflict’s effect on democratization have focused on the influence of militarized interstate disputes (MIDs).[29]MIDs are a broad class of conflicts ranging in intensity from wars to disputes involving threats to use force but no actual fatalities.[30]Disputes that do not escalate to war constitute the vast majority of MIDs.However, the main causal mechanisms through which war is alleged to cause democratization (such as breaking the grip of authoritarian elites and mobilizing masses) seem more relevant to war than to lower-intensity disputes.Consequently, we analyze wars rather than MIDs in the following empirical analysis.

Like most quantitative studies of war, we rely on the COW Project’s definition of and data on war.[31]These data cover the period from 1816 to 1997.The COW Project defines two types of external wars, which are conflicts in which a state (or group of states) actively fights a foreign enemy.Interstate wars are hostilities between members of the interstate system that generate at least 1,000 battle fatalities.To be considered a participant, a state must have suffered at least 100 fatalities or have sent at least 1,000 troops into combat.Extra-systemic wars are imperial or colonial actions in which a nation-state engages in military conflict against a non-state actor, leading to at least 1,000 battle deaths.States are considered participants in such wars if they sustained – in combination with any allies – at least 1,000 deaths in battle during each year of the conflict.Since existing arguments suggest that any type of external war might promote democratization, we analyze both interstate and extra-systemic wars.

Initially, we analyze interstate and extra-systemic wars together.To this end, we create a variable, All War, that equals 1 if state i is involved in either type of external war in year t-6, 0 otherwise.However, we also examine these types of wars separately to determine if their effects on democratization differ.To this end, we define two variables.Interstate War equals 1 if state i is embroiled in a war with another nation-state in year t-6, 0 otherwise.Extra-Systemic War equals 1 if state i is involved in a foreign war against a non-state actor in year t-6.Recall that Democratization is measured from year t-5 to year t.Measuring war in year t-6 reduces the possibility that our results will be undermined by any simultaneity bias that might exist if a state’s regime type in t-5 influences whether it is enmeshed in a war.Nonetheless, we will also analyze the impact of war when it is measured in year t-5 in a set of additional tests described later.

The Control Variables

In addition to external war, it is important to account for various factors that previous studies have linked to democratization.Due to our interest in tracking the effects of war on democratization over the last two centuries and across as many countries as possible, accumulating data on some of these factors is quite difficult.Nonetheless, we include a series of control variables in the following analyses, all of which are measured in year t-5 unless otherwise noted.

First, we include the natural logarithm of: (1) each state’s per capita energy consumption (Development), (2) each state’s national population (Population), and (3) the ratio of each state’s total military personnel in year t-6 to its total military personnel in year t-11 (Military Personnel).Further, we include the percentage of its population located in urban centers (Urban Population), and the length of time it has been a sovereign entity (Years Sovereign).Data on these variables are taken from the COW Project’s National Material Capabilities version 3.02 and State System Membership List v2004.1.[32]Many studies have concluded that heightened economic development promotes democratization, and per capita energy consumption is a widely-used measure of development.In addition, various studies have argued that smaller countries may be better able to create and maintain democratic forms of government.[33]The log of the ratio of military personnel in year t-6 to t-11 is included to test the argument that, as the government draws a larger portion of society into the military, citizens will demand a more democratic polity.[34]We analyze the percentage of the population living in urban centers because it is commonly held that urbanization contributes to democratization.[35]The length of time that a state has been sovereign is included because it has been argued that states are more likely to democratize as they age.[36]