GLOBAL REPORT ON CONFLICT, GOVER-NANCE, AND STATE FRAGILITY 2008

Monty G. MarshallGeorge Mason University

andBenjamin R. Cole

George Mason University

The Global Report series and its signatureState Fragility Index and Matrix firstappeared in the March 2007 edition of theForeign Policy Bulletin.1 It was designedby Monty G. Marshall and Jack Goldstoneat the Center for Global Policy, GeorgeMason University, and patterned after thePeace and Conflict series created by Mar-shall and Ted Robert Gurr in 2001. Theseglobal report series were designed to satis-fy the imperative for knowing the contrast-ing conditions characterizing the manystates comprising the emerging global sys-tem and gauging general system perfor-mance in an era of dynamic globalization.The original report published in 2000sparked controversy within the global poli-cy community with its prescient observa-tion, and presentation of supporting evi-dence, that “the extent of warfare amongand within states lessened by nearly half inthe first decade after the [end of the] ColdWar.”2 This claim was initially dismissedas either mistaken or misinformed by mostofficials and analysts in the United NationsSecretariat when it was brought to theirattention. The claim clearly challenged theprevailing perception of increasing globaldisorder and that the world was becominga more, not less, dangerous place.3 It tookseveral years before critical reaction turnedaway from examining the claim itself tooffering explanations for the global

decrease in warfare. In the current GlobalReport, we continue the original claim byobserving that global warfare has remainedin decline through 2007 and has dimin-ished by over sixty percent since its peak inthe late 1980s. Consistent with the declinein major armed conflicts has been the con-tinuing increase in the number and consol-idation of democratic regimes, rising toninety-four at the end of 2007 (nearly sixtypercent of the 162 countries examined inthis report). Some cause for concern mustalso be reported: the number of ongoingarmed conflicts may be showing signs ofleveling off, the frequency of onsets of newarmed conflicts in the world has notdecreased substantially since the end of theCold War in 1991, and the occurrence of“high casualty terrorist bombings” has con-tinued to increase through 2007. It appearsthat, while world politics have been suc-cessful in gaining peaceful settlements tomany of the world’s armed conflicts, sever-al long-running wars continue to resistpeaceful settlement and new armed con-flicts continue to break out regularly.

This report begins with a brief discussionof general, systemic trends in global con-flict, governance, and development, with adetailed assessment of changes in StateFragility since 1995. It then presents theState Fragility Index and Matrix 2008(Table 1) which provides an array of mea-sures of individual state fragilities and, byimplication, a systematic assessment of thecapacities and prospects for each of the 162independent countries (with total popula-tions greater than 500,000) that comprisethe global system. The State FragilityIndex combines scores measuring twoessential qualities of state performance:effectiveness and legitimacy; these twoquality indices combine scores on distinctmeasures of the key performance dimen-

sions of security, governance, economics,and social development. The latest versionof the Fragility Matrix has established abaseline set of values for its eight compo-nent indicators in order to measure StateFragility in previous years and examinechanges in each indicator over time.

Global Trends and Systems Analysis

Conventional analyses of security and gov-ernance factors have for too long reliedalmost exclusively on individual or dyadic(bilateral) analysis, that is, on the condi-tions relevant to a particular country orstate or relative to the interactions of twostates. Systems analysis was largely con-fined to the analysis of alliance structuresand treaty organizations. The Cold Warwas, at once, the penultimate example ofdyadic analysis (the “superpower con-frontation”) and a symbolic end to theanarchic, Westphalian state system. It is anatural consequence of the end of the ColdWar that we should begin an era of openglobalization and, with that, widen our per-spectives to recognize the complexities anddensities of interactions, interconnections,and networks among the myriad actors thatconstitute the emerging “global system ofstates.”4

Systems analysis necessarily focuses onthe complex relations between dynamics(human agency and environmental forces)and statics (physical and social attributes,conditions, and structures). Basic societal-systems analysis must take into account theinterconnectedness of three key, or funda-mental, dimensions: conflict, governance,and development (including both physicaland social capital; Figure 1).5 Availabletechnology largely determines the size andcomplexity of viable societal-systems. Thequalities, and prospects, of each of the

Global Report 2008

Foreign Policy Bulletin 3

Global Report on Conflict, Governanceand State Fragility 2008

Monty G. Marshall and Benjamin R. Cole, George Mason University

three fundamental dimensions of societal-systems critically affects the qualities ofthe other two dimensions to such a degreethat it is not possible to meaningfully ana-lyze one dimension without taking theother two into account. Any change in onedimension will have consequences for eachof the other dimensions; any limitation orweakness in one of the key dimensions willlessen the prospects for improvement in theother dimensions. Successful performanceof a societal-system can be expected to beboth incremental and congruent among thekey dimensions. Societal-system perfor-mance, then, depends on the system’s capa-bilities for collective action: applied coor-dination (effectiveness) and voluntarycompliance (legitimacy). A performanceevaluation of a societal-system, whethertaken at the local, regional, or global level,must measure and track performance in allkey dimensions with a mind toward coher-ence, progress, and congruence among thedimensions. Problems that arise in societal-system dynamics can stem from any of thethree fundamental dimensions. The quali-ties of governance and development mustbe taken into account when analyzing orleveraging conflict. Likewise, the qualitiesof conflict and governance must be includ-ed when examining the qualities of andpotential for development and the qualitiesof conflict and development criticallyaffect the nature and prospects of gover-nance.

This report provides general, macro-com-parative evaluations of contemporary qual-ities and trends over time in the three fun-damental dimensions of societal-systems

analysis at the global level. These perfor-mance evaluations are intended to helpinform our audience of the immediate cir-cumstances and prospects of globalization.

Global Governance

Democracy and autocracy are commonlyviewed as contrasting and distinct forms ofgovernance. Principal differences arefound in the ways executive power isacquired and transferred, how politicalpower is exercised and constrained, howsocial order is defined and maintained, andhow much influence public interests andopinion have on the decision makingprocess. Despite fundamental differences,these two ideal forms of governance areoften perceived as comparably stable andeffective in maintaining social order. Inreal terms, however, different countrieshave different mixes and qualities of gov-erning authority; the ideal types are rarelyobserved in practice. Even though somecountries may have mixed features ofopenness, competitiveness, and regulation,the core qualities of democracy and autoc-racy can be viewed as defining oppositeends of a governance scale. We have ratedthe levels of both democracy and autocracyfor each country and year using codedinformation on the general qualities ofpolitical institutions and processes, includ-ing executive recruitment, constraints onexecutive action, and political competition.These ratings have been combined into asingle, scaled measure of regime gover-nance: the Polity score. The Polity scaleranges from -10, fully institutionalized

autocracy, to +10, fully institutionalizeddemocracy.6 A perfect +10 democracy, likeAustralia, Greece, and Sweden, has institu-tionalized procedures for open and compet-itive political participation; chooses andreplaces chief executives in open, compet-itive elections; and imposes substantialchecks and balances on the powers of thechief executive. Countries with Polityscores from +6 to +10 are counted asdemocracies in tracking “Global Trends inGovernance, 1946-2007” (Figure 2). Elect-ed governments that fall short of a perfect+10, like Bolivia, Mozambique, Turkey,and Indonesia, may have weaker checks onexecutive power, some restrictions onpolitical participation, or shortcomings inthe application of the rule of law to opposi-tion groups.

In a perfect -10 autocracy, by contrast,citizens’ participation is sharply restrictedor suppressed; chief executives are selectedaccording to clearly defined (usuallyhereditary) rules of succession from withinthe established political elite; and, once inoffice, chief executives exercise powerwith few or no checks from legislative,judicial, or civil society institutions. OnlySaudi Arabia and Qatar are rated as fullyinstitutionalized autocracies in early 2008.Other monarchies, such as those in Bhutan,Morocco, and Swaziland, share some pow-ers with elected officials. In general, exceptfor a strong presence in the oil-producingstates of the Arabian Peninsula, hereditarymonarchy has nearly disappeared as a formof governance in the early 21st century.Autocratic governance at the turn of thecentury is far more likely to be character-ized by the authoritarian rule of personalis-tic leaders, military juntas, or one-partystructures; Libya, Myanmar (Burma), andVietnam are examples of these non-monar-chical autocracies. Besides having slightlymore open, or less-clearly defined, rules ofsuccession, less-than-perfect autocraciesmay allow some space for political partici-pation or impose some effective limits onexecutive authority; examples includeBelarus, China, and Zimbabwe. Countrieswith Polity scores of -10 to -6 are countedas autocracies in Figure 2.

Many governments have a mix of demo-cratic and autocratic features, for exampleholding competitive elections for a legisla-ture that exercises little effective control on

Global Report 2008

4 Winter 2008

the executive branch or allowing openpolitical competition among some socialgroups while seriously restricting partici-pation of other groups. There are many rea-sons why countries may come to be char-acterized by such inconsistencies, or inco-herence, in governance. Some countriesmay be implementing a staged transitionfrom autocracy to greater democracy; oth-ers may institute piecemeal reforms due toincreasing demands from emerging politi-cal groups; others may simply be losingtheir capacity to maintain strict politicalcontrols and suppress dissent. Societal con-flict and factionalism often stalematedemocratic experiments: some regimesmay be unable to fully institutionalizereforms due to serious disagreementsamong social groups or key political elites;some may harden their institutions inresponse to political crises or due to thepersonal ambitions of opportunistic lead-ers; and others may simply lose control ofthe political dynamics that enable, or dis-able, effective governance. Whereasdemocracy and autocracy are very differentforms and strategies of governance, theyare very similar in their general capacity tomaintain central authority, control the poli-cy agenda, and manage political dynamics.Anocracy, by contrast, is characterized byinstitutions and political elites that are farless capable of performing these funda-mental tasks and ensuring their own conti-nuity. Anocratic regimes very often reflectan inherent quality of instability or ineffec-tiveness and are especially vulnerable tothe onset of new political instability events,such as outbreaks of armed conflict, unex-pected changes in leadership, or adverseregime changes (e.g., a seizure of power bya personalistic or military leader).

Anocracies are a middling category ratherthan a distinct form of governance. Theyare countries whose governments are nei-ther fully democratic nor fully autocratic;their Polity scores range from -5 to +5.7Some such countries have succeeded inestablishing democracy following a stagedtransition from autocracy through anocra-cy, as in Mexico, Nicaragua, Senegal, andTaiwan. A number of African and a fewMiddle Eastern countries have recentlybegun a cautious transition to greater open-ness, among them Burkina Faso, Djibouti,Ghana, Guinea, Jordan, and Tanzania.

Ivory Coast appeared to be headed on asimilar course before stumbling (in 2002)into civil war and regime failure; Iran alsoreversed the course of democratic reformsand tightened autocratic control in 2004.Others have been able to manage conflictbetween deeply-divided social groups forsubstantial periods of time through the useof restrictions on political participation bya substantial out-group as in Malaysia(Chinese), Singapore (Malays), and SouthAfrica (black-Africans under Apartheid).This also appears to be the strategy adopt-ed recently in Fiji to limit political influ-ence by ethnic-Indians (until that policywas challenged by a military coup in late2006). Other anocracies are the result offailed transitions to greater democracy, ascurrently in Algeria, Angola, Cambodia,and Haiti.

In 1946, there were seventy-one indepen-dent states comprising the world’s systemof states (Figure 2; each of the followingglobal trends figures contains a verticalline demarcating the end of the Cold Warperiod in 1991).8 Of these, twenty coun-tries were ruled by democratic regimes andnineteen by autocratic regimes; thirty-twocountries were subject to anocraticregimes. The high proportion of anocraticregimes was largely a consequence of thesevere devastation and disruptions result-ing from the Second World War. Anotherconsequence of that war was a serious ero-sion of European control over its colonialterritories in Asia and Africa. Many newstates gained independence in the 1950s,1960s, and early 1970s, doubling the num-

ber of states in the world by 1975. Duringthis period of decolonization, there was adramatic increase in the number of auto-cratic regimes: to a peak of eighty-nineautocracies in 1977. Although new stateswere about as likely to adopt democratic asautocratic forms of governance upon gain-ing independence, problems of manage-ability caused most new, democraticregimes to fail within several years andgive way to autocratic rule. A dramaticshift away from rigidly autocratic regimesand toward more open governance beganin 1990. This “rush toward democratiza-tion” was led by Latin American countriesand the former-Socialist countries of East-ern Europe. During the Cold War period,there was a steady increase in the numberof democracies at the rate of about one newdemocracy every two years. During theearly 1990s, the number of democraciesincreased by about fifty percent (from 49 in1989 to 76 in 1995). There was an evengreater increase in the number of incom-plete transitions to democracy, as the num-ber of anocracies rose from twenty-sevento forty-eight (falling back to forty-three inlate 2007). The number of autocracies con-tinues to plummet: from a peak of eighty-nine in 1977 to just twenty-five at the endof 2007. There are ninety-four countriesclassified as democracies in late 2007.Countries that have transitioned to, orreturned to, democratic governance since2000 include Burundi, Comoros, Ghana,Guinea-Bissau, Kenya, Lebanon, Liberia,Nepal, Peru, Sri Lanka, and the newlyindependent states of East Timor and Mon-

Global Report 2008

Foreign Policy Bulletin 5

tenegro. The one thing that most clearlydistinguishes the Globalization Era is that,for the first time in human history, theworld has become a predominantly democ-ratic one, at the global level.

While we view the major global shifttoward greater democracy as a very impor-tant and generally positive trend, the sharpincrease in the number of anocracies con-current with the end of the Cold War iscause for concern. Historical research indi-cates that anocracies have been highlyunstable regimes, with over fifty percentexperiencing a major regime change with-in five years and over seventy percentwithin ten years. Anocracies have beenmuch more vulnerable to new outbreaks ofarmed societal conflict; they have beenabout six times more likely than democra-cies and two and one-half times as likely asautocracies to experience new outbreaks ofsocietal wars. Anocracies have also beenabout three times more likely to experiencemajor reversions to autocracy than democ-racies. However, a “new truth” may beemerging regarding the vulnerability ofanocratic regimes in the Globalization Era.In the past fifteen years, there have been farfewer failures of anocratic regimes thanwould be expected from the historicaltrends. Despite continued high numbers ofanocratic regimes, there has been a steadydecrease in global trends in violent conflict(see Figure 3) and fewer than expected out-breaks of new political instability events.We believe that this change in trends foranocratic regimes is due largely to notableincreases in proactive international engage-ment, improved public expectations, and alessening of political activism within mili-taries, which have been far less likely tointervene in politics or support forcefulrepression of public challenges to rulingelites. Counter-examples have occurredrecently as military coups have oustedelected governments in Thailand and Fiji inlate 2006, and Bangladesh in 2007.

Recent research by the US Government’sPolitical Instability Task Force has focusedattention on the problem of “factionalism”in “incomplete democracies.”9 In generalterms, the Polity conceptualization of fac-tionalism refers to an advanced, macro-systemic stage of group polarization thattransforms political behavior in distinctly

contentious ways that are both systematicand sustained. Factionalism transforms theconventional politics of deliberation to theunconventional, anti-system politics of dis-ruption and control. In the Task Force’smodels of the onset of political instability,the factionalism condition stands out ashaving the greatest explanatory poweramong global model indicators.10 Thecondition of factionalism is a precursor toinstability in about half of the countrieswhere it occurs; the other principal out-come of factionalism is the further consol-idation of democratic procedures and dis-course. In accordance with its observedoutcomes, the onset of political polariza-tion or factionalism must be viewed as apolitical crisis condition, a direct challengeto the governing regime, and a key, policydecision-point between two fundamentalcourses: stabilization or destabilization.

The “problem of factionalism” in new orincomplete democracies is not a new find-ing, by any means. In fact, it is probablythe most widely accepted, and least under-stood, problem in the process of democra-tization. In The Federalist No. 10, JamesMadison (1787) makes several prescientobservations in this regard, among theseare 1) the link between “domestic factionand insurrection;” 2) the opportunityafforded by factionalism for “adversariesto liberty” to declaim popular government;3) the observation that the dynamics of“instability, injustice, and confusion” thatfactionalism introduces into public coun-

cils are the “mortal diseases under whichpopular governments have everywhere per-ished;” and 4) the conclusion that the“friend of popular governments” must actwith due diligence to pursue any planwhich “provides a proper cure” to faction-alism “without violating the principles” ofliberty and diversity. Nearly two-thirds ofthe anocracies charted in Figure 2 are char-acterized by factionalism currently. Exam-ples of factionalism in early 2008 can beobserved in Algeria, Egypt, Ethiopia,Georgia, Guinea, Togo, Uganda,Venezuela, and Zimbabwe. Factionalismcan also be a serious problem in the moreinstitutionalized democratic regimes,affecting about one-in-five in early 2008;current examples include Belgium,Bolivia, East Timor, Ecuador, Guyana,Lebanon, and Ukraine. While factionalismpresents a very high risk factor for theonset of political instability, by far thegreatest risk is for the onset of an “adverseregime change” or “autocratic backsliding”whereby democratization is reversedthrough the politicization and activism ofthe military and/or internal security forcesand the oppositional faction(s) is/areforcibly repressed. We are currently con-ducting new research to gain a betterunderstanding of the problem of factional-ism in both the democratization processand in maintaining democratic governance.

Global Armed Conflict

The most encompassing observation that

Global Report 2008

6 Winter 2008

can be made regarding global system per-formance in regard to the conflict dimen-sion concerns the status of regional con-flicts. The global trend in major armed con-flict has continued its dramatic decline inthe globalization era both in numbers ofstates affected by major armed conflictsand in general magnitude (Figure 3).According to our calculations, the generalmagnitude of global warfare has decreasedby over sixty percent since peaking in themid-1980s, falling by the end of 2007 to itslowest level since 1960.11 Civil warfarehas been the prominent mode of warfaresince the mid-1950s; increasing steeplyand steadily through the Cold War period.This linear increase in civil warfare islargely explained by a general tendencytoward longer, more protracted, wars dur-ing this period; internal wars often receiv-ing crucial support from foreign suppliers.The rate of onset of new civil wars hasremained fairly constant throughout theperiod with an average of about four newcivil wars per annum. On the other hand,the general global level of interstate war-fare has remained at a relatively low levelsince the end of the Second World War andthe establishment of the United NationsOrganization, which was specificallydesigned to help prevent interstate wars.Although there was a moderate increase ininterstate wars during the last years of theCold War, from 1977 to 1987, like civilwarfare, interstate warfare has alsodeclined substantially with the end of theCold War. Of the interstate wars that tookplace during the Cold War period, many of

the most serious were wars of indepen-dence fought during the decolonizationphase that coincided with the first half ofthe Cold War. Of the conventional inter-state wars, onsets occurred at the rate ofabout one event per annum, althoughonsets occurred at about double that rateduring the late 1970s and early 1980s.Three-quarters of sixty-seven such warsremained at fairly low levels of violence.High magnitude interstate wars are limitedto the several Israeli wars, the Vietnamesewars, the Afghanistan wars, the Iraqi wars,the India-Pakistan wars, and the recent warbetween Ethiopia and Eritrea; all exceptthe Iraq-Iran war and the first Gulf War hadsome domestic, or former-domestic con-flict aspect (i.e., internationalized civilwars). Over the entire period, since 1946,wars have been quite common: there havebeen over 320 distinct episodes of majorarmed conflict in the world’s 162 coun-tries. During the past twenty-five years(since 1983), just over one-half of all coun-tries have experienced some major armedconflict (83 of 162 countries).

In early 2008, there were twenty-onecountries embroiled in major armed con-flicts; twenty of these countries are besetby civil or communal wars: Afghanistan,Central African Republic, Chad, Colombia,Democratic Republic of Congo, Ethiopia,India, Iraq, Israel (Palestine), Kenya,Myanmar, Nigeria, Philippines, Pakistan,Somalia, Sri Lanka, Sudan, Thailand,Turkey, and Yemen. One of the currentwars is touted as a “global war” (the Unit-ed States’ “global war on terrorism”),

although if war means systematic and sus-tained attacks, that overt war has been con-fined almost entirely to US and US-ledmilitary deployments in Iraq andAfghanistan.12 Increases in militantattacks in Pakistan in 2007 against the US-allied regime may indicate that the largelylocalized wars focused in Iraq andAfghanistan are fueling a larger, regionalconflict in early 2008. There have been fivenew onsets of serious armed conflict in thepast three years: a civil war in the Pakistanprovince of Baluchistan, a cross-borderrebellion along Chad’s border with theDarfur region in Sudan and a relatedepisode of political violence in neighboringareas of the Central African Republic, abrief but intense interstate war involvingIsrael and Lebanon (mainly Hezbollahmilitias), fighting between Lebanese mili-tary and Islamists in Palestinian refugeecamps, a rebellion in the Ogaden region ofEthiopia bordering Somalia, and seriouscommunal violence in Kenya followingflawed elections there. Indeed, many, if notmost, of the active armed conflicts in early2008 show evidence of serious cross-bor-der effects that are increasing inter-stateand regional tensions in the affected areas.

Figure 4, “Annual Numbers of NewOnsets and Ongoing Wars,” provides someadditional evidence that, while the overallmagnitude of global wars continues todiminish, the numbers of ongoing wars inthe global system may be leveling off andthe frequency of onset of new wars remainsa serious problem. There is also some evi-dence of continuing and, even, increasingtensions in several of the countries whereserious civil or communal warfare has beencalmed through negotiated settlements orcease-fires. Wars in Turkey and Sri Lankarekindled recently following long lulls inthe fighting and outbursts of violence havereturned to vex the complex social mosaicof Lebanon. Several countries have man-aged to push warring groups across theirborders into neighboring countries wherethese groups continue to pose seriousthreats to both their home and host coun-tries, as is the case with LRA fighters dri-ven from Uganda and Hutu militias fromRwanda. Several others have managed togain a respite from violence through defacto separation, as has occurred in Azer-baijan, Bosnia, Georgia, Moldova, and Ser-

Global Report 2008

Foreign Policy Bulletin 7

bia.13In regard to the US-led “global war on

terrorism,” two positive observations canbe made: 1) direct attacks on the US and its(non-Muslim) allies have remained rela-tively rare since the dramatic September11, 2001, attacks in the United States(notable exceptions were the 2004 attacksin Madrid and 2005 attacks in London) and2) there have been few instances of attacksanywhere in which biological, chemical, ornuclear “weapons of mass destruction”have been used by non-state actors (withthe exception of some attempts to use chlo-rine gas in Iraq). What we generally per-ceive as “terrorism,” as distinct from theterrifying violence associated with warfare,is the direct and intentional targeting ofcivilian or other non-combatant groups.Civilian populations are inherently vulner-able to political violence and the generallawlessness and disruptions in livelihoodsand essential services that are substance ofattrition in protracted conflict situations;they live and stand in harm’s way.14Explosive devices (concealed bombs, car-bombs, and suicide bombers) are the prin-cipal means by which actors have directlyattacked civilian populations with theintent to inflict high casualties. Figure 5tracks the problem of “high casualty terror-ist bombings” at the global level over thepast eleven years; each of the “high casual-ty” events compiled for the trend graphresulted in at least fifteen people killed in asingle attack in which bombs were a prin-cipal weapon used by the attacker(s).15

What Figure 5 shows is a very steep

increase in the number of people killed inhigh casualty terrorist bombings (HCTB)since the 9/11 attacks in the US. The num-ber killed during the most recent year,2007, was over 5,000; the average duringthe five-year period before prior to Sep-tember 11, 2001, was about 380 per annum(up from about 160 per annum in the pre-ceding five years). The graph also parcelsattacks into three categories: Muslim coun-tries, Muslim attacks in non-Muslim coun-tries, and (non-Muslim) attacks in non-Muslim countries. HCTB incidents havebeen particularly prominent in the ongoingwar in Iraq, increasing dramatically fromthe US invasion in March 2003 throughSeptember 2007. The total number of peo-ple killed in HCTB attacks over the seven-teen-year study period stands at 21,848;however, the actual number of peoplekilled in such attacks is probably substan-tially higher (c20-25%) as reported num-bers of people killed are usually thosekilled immediately by the blast and oftendo not include numbers of people who sub-sequently die from their wounds.16 By far,the largest HCTB event was the coordinat-ed attacks using hijacked airliners as aerialbombs on September 11, 2001, in the US(2,982 killed). The next two largest HCTBevents resulted in 520 killed in coordinatedattacks on Qataniyah and Jazeera in Iraq onAugust 14, 2007, and 331 killed in the Sep-tember 1, 2004, attack in Beslan, Russia.The recent trend in HCTB attacks stands invivid contrast to the more encouraging,downward trend in major armed conflicts.While it may be too much to claim thatHCTB terrorism has a global scope, it is

certain that such terrorism has a globalreach while it remains concentrated in theMiddle East and South Asia. What hasmost characterized the increasing volumeof HCTB attacks since 2001 has been theirconcentration in Muslim-on-Muslim vio-lence (reaching ninety-five percent of allHCTB attacks in 2007), a phenomenon thatmay be best described as “Muslim rage.”17Since September 2007, there has been anotable decrease in the incidence and num-bers of people killed in HCTB events,declining by about fifty percent. Alsonotable is a dramatic shift in the location ofHCTB events: there are substantial increas-es in the use of this tactic in Afghanistan,Pakistan, and Sri Lanka and a dramaticdecline in Iraq. On the other hand, there isa more general emphasis in the use ofHCTB events on striking political targets(public officials and symbols of authority).Taken together, the shift in emphasis topolitical targets and the change in locationfor these events may be construed as signsof a “maturing,” regional rebellion.

Global Development

The third major focus of this Global Reportseries is on global development and thegeneral performance of the economic(material production) and social welfareaspects of globalization and the global sys-tem. The 2007 Global Report highlightedthe great, regional (and, in some cases,intra-regional) disparities in economicdevelopment and the systemic distributionof income. It used a methodology termed“comparative regionalism” to assess therelative economic strength of the statescomprising the global system of States. Itclaimed that the North Atlantic (US, Cana-da, and Western Europe) and South Ameri-can regional sub-systems were “good-per-formance” systems, with a caveat pointingto the increased challenges posed byexpansion of the European Union toinclude the countries of Eastern Europeand Turkey. The 2007 report went on toclaim that the Central American and Southand East Asian sub-systems were “middle-performance” and that the Non-MuslimAfrica and Muslim Countries sub-systemswere comparable “poor-performance” sub-systems. It also highlighted the observationthat the better-performing sub-systems

Global Report 2008

8 Winter 2008

were net-consumers of energy resourceswhile the poorer-performing sub-systemswere characterized by great income dispar-ities between the resource-rich (often, net-producers of petroleum) countries and theresource-poor countries. The report raisedserious concerns regarding the level of ten-sions that would likely occur in a globalsystem characterized by relatively small,super-powerful, resource-demandingregions and large, weak, resource-produc-ing regions. “It would seem that the poten-tial for polarization and factionalism insuch a system is quite high and, given theevidence that the ‘income gap’ is narrow-ing only slowly, will remain high for theforeseeable future. The policy implicationsof this examination can be summarized in asingle word: caution.”18 The report con-cluded by presenting three challenges forthe emerging era of globalization: “one isnarrowing the divide between ‘well being’and ‘fragility’ in constituent societies; asecond is calming the voices of oppositionand transforming their creativity and ener-gy to promote rather than disrupt the glob-al system; and a third is to recognize thefull, disruptive potential of our growingdependence on petroleum and accept thisas a global dilemma, requiring a globalsolution.”19 In this third section, we usemeasured changes in the State FragilityIndex and Matrix from 1995 to 2007 togain a better understanding of progressbeing made toward addressing the firstchallenge, that is, “narrowing the dividebetween ‘well being’ and ‘fragility’ in con-stituent societies.” We will then conclude

our 2008 Global Report by presenting ourmost recent State Fragility assessments foreach of the 162 countries (with populationsgreater than 500,000) that constitute theglobal system in early 2008. The StateFragility Index and Matrix (Table 1, fol-lowing) rates each country according to itslevel of fragility in both effectiveness andlegitimacy across four dimensions: securi-ty, governance, economic development,and social development.

Before we begin our general assessmentof progress in global development weexamine the relationship of state fragilityand the standard measure of a country’seconomic performance: income measuredas gross domestic product (GDP) per capi-ta. Figure 6 plots the relationship betweenour most recent year State Fragility Indexscore and GDP per capita for each of the162 countries included in this study.20 Weconvert GDP per capita figures to their(base 10) logarithmic value because of thevast income disparities among countries inthe global system, wherein the range ofvalues is from $93 to $40,947 and the dis-tribution is highly skewed such that fiftypercent of country income values are lessthan $1,750 and seventy-five percent areless than $6,000. The “best fit” of the rela-tionship between fragility (sfi2008 in Fig-ure 6) and income (Log (base 10) ofGDPPC) is shown to be slightly curvilinear(a quadratic function) with a fairly highcorrelation of 0.716. There is obviously avery strong, negative relationship betweenthe income and fragility of states in theglobal system. However, we can also see a

rather wide variance in fragility scores atany level of income. Countries plotted tothe left of the curve at any level of incomeare performing better than expected by themodel, whereas, countries plotted to theright of the curve are performing morepoorly than expected given their level ofincome. In order to provide an additionalperspective on the “oil curse” we examinedin the 2007 report, we have identified all ofthe top, net oil-producing producing coun-tries (i.e., those with annual net productionper capita greater than ten barrels of oil;identified by shaded-diamond icons). OnlyDenmark, Russia, and Kazakhstan havefragility values near their expected valuesgiven their level of income; all other oil-producing states have fragility scores fargreater than would be expected for theirlevel of income. We have also identifiedtwo referent levels of income: the level ofincome where the model curve intersectsthe value 8 on the fragility index ($1,300)and the level of income at value 4 ($3,800).

Regional Changes in State Fragility: Inorder to gain a better understanding ofchange in the general performance of theglobal system, we use the State FragilityIndex and Matrix assessment methodologyto calculate scores for each country in ear-lier years and, then, examine the changes inassessment values across time. To this pur-pose, we calculated each country’s fragilityscores for each year beginning with theyear 2000. In order to provide additional,temporal depth and a starting point for theemerging era of globalization, we then cal-culated each country’s fragility scores forthe year 1995. The year 1995 was chosenbecause it was well within the post-ColdWar period (which we set as beginning in1992) and had full data coverage on the rel-evant indicators used to construct theFragility Index and Matrix. The original(2007) design for the fragility scores usedquintile cut-points to establish ordinal cat-egories for fragility values on several of theindicators, so, in order to establish baselinevalues for measuring change in values overtime, we (somewhat arbitrarily) chose2004 as the baseline year and the quintilecut-points for the target indicators in thatyear as referents for establishing baselinevalues for the fragility categories. Having,thus, set the parameters for the Matrix wechose the middle-year in the coverage peri-

Global Report 2008

Foreign Policy Bulletin 9

od, 1995-2007, (i.e., the year 2001) as athird point in charting the trajectories ofchange in fragility for each of the countriesunder evaluation. Table 1, following, liststhe State Fragility Index scores for eachcountry in the Matrix for the years 1995,2001, and 2007, and provides an indicatorof the general trajectory of change infragility over the twelve-year period.

Fifty-six of the 162 countries listed inTable 1 show consistent, positive change ofthree points or more over the period (i.e., alower fragility index score for the year2007, as compared with the 1995 score,with the year 2001 score either being thesame or lower than the 1995 score) where-as, only ten countries show consistentlynegative change of two points or more overthe same period (i.e., a higher fragilityindex).21 In all, 110 of 160 countries showlower fragility scores in 2007 than in 1995,with 29 showing the same score and 21showing higher fragility scores (two coun-tries, East Timor and Montenegro did notexist in 1995). The largest improvements infragility score across the study period arein Mali (nine point decrease); Guatemala(eight points); and Bangladesh, Bosnia,and Togo (seven points), with Croatia,India, Nicaragua, and Bhutan improvingtheir scores by six points each. Countrieswith greater fragility scores across the peri-od include Central African Republic (fivepoint increase), Eritrea and Nepal (fourpoints), and Congo-Brazzaville, Ecuador,Israel, Kyrgyzstan, Namibia, SolomonIslands, United States, Uzbekistan, andVenezuela (two point increases).

Figure 7 provides a regional summary ofchanges in State Fragility Index scores dur-ing the study period. States were assignedto one of six politically-salient regions:Non-Muslim Africa (sub-Saharan coun-tries); Muslim Countries (i.e., countries inwhich Muslim confessional groups com-prise fifty percent or more of the total pop-ulation); (non-Muslim) South and EastAsia; Latin America; (non-Muslim) For-mer-Socialist countries; and North Atlanticcountries.22 The bars in the graph showchanges in the mean fragility score for eachregion across the three annual indices(1995 and 2001, 2001 and 2007, and 1995and 2007); the bars are measured on theleft-hand axis. The black-diamond iconsindicate each region’s average StateFragility Index score at the end of 2007;the icons are measured on the right-handaxis. Note that, while Muslim countries arelargely geographically concentrated innorthern Africa and the Middle East, thereare Muslim countries in Europe (Albaniaand Bosnia), the former-Soviet Union(Azerbaijan, Kazakhstan, Kyrgyzstan,Tajikistan, Turkmenistan, and Uzbekistan),and southern Asia and Oceania(Bangladesh, Indonesia, and Malaysia).The regions are arranged according to theirmean State Fragility Index scores, with themost fragile region (Non-Muslim Africa;mean 14.48) on the left and the least frag-ile region (North Atlantic countries; mean0.68) on the right.

The least fragile region in 2007 is theNorth Atlantic region; this region includesWestern Europe, Canada, and the United

States (nineteen countries in 2007).23 TheNorth Atlantic region’s mean State Fragili-ty Index score in 2007 is 0.68, with scoresranging from 0 (ten countries in 2007) to 3(Cyprus). The largest change in fragilityscore is that of the United States, for whichone-point increases in fragility scores arefound in both Security Effectiveness (warin Iraq) and Security Legitimacy (increaseduse of state repression associated with the“global war on terrorism”). Overall, theNorth Atlantic region has long been andstill remains the standard for gaugingregional performance and (lack of) fragili-ty. The question remains whether thisregion has set a reasonable and achievablestandard that is accessible to all countriesin the global system or whether some mod-eration in regional consumption, income,and wealth is a necessary corollary tobroader system access to reasonable stan-dards of achievement.

Closely following the North Atlanticregion in terms of overall fragility is theEastern European region comprising coun-tries that have emerged from the Former-Socialist bloc, including many of the for-mer-Soviet republics (except the predomi-nantly Muslim countries of Albania,Bosnia, Azerbaijan, and the Central Asianrepublics).24 This region’s mean score in2007 is 3.32, with scores ranging from 0(Hungary, Latvia, Lithuania, Poland, andSlovenia) to 9 (Moldova; Georgia followswith a score of 8 and Russia with 7). Thisregion charts the greatest improvement infragility scores since 1995 with an averagedecrease in State Fragility Index scores of2.57 (cutting the regional mean by nearlyhalf). The overall change in fragility scoresfor this region is due equally to improve-ments in effectiveness and legitimacy andthese improvements are nearly equallyspread across the four performance dimen-sions. Lesser improvements in fragility arenotably in areas where this region hadalready made substantial achievements:security effectiveness and legitimacy andeconomic legitimacy. Improvements in thelatter half of the period were somewhat lessthan the earlier half, probably due to thefact that the region was drawing closer tothe “ceiling” of improvement. Of particularnote is Croatia, which reduced its StateFragility Index score by six points, andLatvia and Georgia, which reduced their

Global Report 2008

10 Winter 2008

fragility scores by five points between1995 and 2007.

Latin America countries improved theirfragility scores by just over two points, onaverage, while the mean fragility score forthe region stands at about double that of theFormer-Socialist countries (6.52 in2007).25 Scores for Latin American coun-tries range from 0 (Costa Rica) to 13(Haiti; Ecuador follows with 12; Boliviaand Colombia score 11). Like the Former-Socialist countries, the Latin Americaregion shows greater improvement infragility scores during the earlier period,1995-2001 with the pace of improvementdropping off substantially in the latter halfof the period, 2001-2007. Latin Americanimprovement was driven largely by gainsin effectiveness, with the biggest gain overthe entire period counted in Political Effec-tiveness ( 0.61). Improvements in Securi-ty, equally driven by gains in effectivenessand legitimacy, also accounted for a largepart of the decrease in regional fragilityscores ( 0.30 each; 0.60 total). In 2007,the legitimacy component of the fragilityscores for the region (4.26 points) wasnearly double that of the effectivenesscomponent (2.26 points). The region per-formed particularly poorly in improvingPolitical Legitimacy ( 0.13) and Econom-ic fragility more generally ( 0.13 in Eco-nomic Effectiveness and 0.17 points inEconomic Legitimacy). Guatemala led theregion in improvement over this period,reducing its fragility score by eight points;Nicaragua improved by six points, Mexicoby five points, and Cuba, El Salvador,Honduras, Panama, and Peru all reducedtheir fragility scores by four or more pointsover this period. Offsetting these improve-ments was a net increase in fragility forEcuador and Venezuela (two points each).

As noted in our 2007 Global Report, therate of growth of the regional income forthe South and East Asia region, as a whole,nearly doubled the rate of economicgrowth in the world’s richest countries;with much of the gains accounted for bythe emergence of China as a major produc-er on the global market.26 Fragility scoresfor this region show moderate improve-ment during the emerging era of globaliza-tion period, 1995-2007, with an averagedecrease in overall fragility of just overtwo points; the regional mean score stands

at 7.68 in 2007. This region shows thebroadest range of fragility scores, from 0(Japan and South Korea) to 21 (Myanmar;East Timor and Nepal score next at 15points). Improvements in this region wereslow in the earlier half of the period andincreased in the latter half; change is near-ly equally spread across the two principalcomponents: effectiveness ( 1.09) andlegitimacy ( 0.96). Only modest gains areshown for Security, in general; PoliticalEffectiveness; and Social Effectiveness; nonet change is noted for Economic Legiti-macy. Most of the region’s improvementsin fragility come in Political Legitimacy( 0.50), Economic Effectiveness ( 0.45),and Social Legitimacy ( 0.32). Improve-ment was particularly strong in India andBhutan (six points each), followed by Laosand Papua New Guinea (five points each)and Cambodia, South Korea, and Vietnam(four points each). During the same period,the fragility ratings for Nepal increased byfour points with increased fragility onSecurity and Political Effectiveness andLegitimacy, offset somewhat by improvedSocial Effectiveness and Legitimacy.

The Muslim Countries were identified inthe 2007 Global Report as one of theworld’s two “poor-performance” regions interms of economic development (alongwith Non-Muslim Africa).27 Between1995 and 2007, the Muslim Countriesrecorded moderate improvement in theirregional mean fragility scores ( 2.38);gains in effectiveness outpaced gains inlegitimacy by nearly two-thirds ( 1.45 and

0.92 respectively). The range of fragilityscores spans from a low of 2 (United ArabEmirates; Kuwait and Albania score 3points) to a high of 22 (Somalia and Sudan;close behind are Afghanistan with 21 andChad and Iraq with 20). Notable improve-ments in regional fragility are found acrossthe Political ( 0.45 and 0.35) and Social( 0.47 and 0.47) dimensions. The Mus-lim Countries region stands out because ofits relatively large net fragility increase inEconomic Legitimacy (i.e., it is becomingmore dependent on revenues from primarycommodities, mainly oil). The NorthAfrican state of Mali is identified as havingthe region’s (and the world’s) largest netdecrease in fragility (nine points), reducingits fragility rating from 19 to 10 over the1995-2007 period. Also notable are Bosnia

and Bangladesh which improved theirfragility scores by seven points each, whileAlbania, Azerbaijan, Indonesia, andTunisia improved by five points. Iraq, Kyr-gyzstan, and Uzbekistan are noted withslightly increased fragility ratings between1995 and 2007; in all these countries,increased fragility was recorded on thelegitimacy component.

Countries comprising the Non-MuslimAfrica region have the world’s highestmean State Fragility Index score (14.48)and showed the least net improvement infragility ratings across the period ( 1.43;discounting the “non-fragile” NorthAtlantic region).28 Fragility scores for thisregion range from 4 (Botswana) to 20(Rwanda and the Democratic Republic ofCongo). Despite the general stagnation infragility ratings for this region, someAfrican countries are noted as havingreduced their fragility ratings substantiallyacross the study period: Togo improvedseven points; Angola and EquatorialGuinea improved by five points; andLiberia, Madagascar, and Ugandaimproved by four points each. The Africaregion also had the most states thatincreased their fragility rating over thisperiod: Central African Republic worsenedby five points, Eritrea by four points,Congo (Brazzaville) and Namibia by twopoints, and Burkina Faso, Cote d’Ivoire,Guinea-Bissau, and Zimbabwe worsenedby one point over this period. Particularlydisheartening is the lack of improvement inthe region’s Social Effectiveness (mea-sured by Human Development Index) andSocial Legitimacy (measured by infantmortality rate) scores. Although we wouldexpect to see the most improvement inthese areas due to NGO and internationaldonor efforts in these areas since 1995, wein fact see almost no net change in the ear-lier half of the period and only modestgains in the more recent half of the period.In terms of social legitimacy, we actuallysee worsening in many cases. Almost allthe net decrease in fragility scores for theNon-Muslim Africa region is accounted forby improvements in Political Effectiveness( 0.57) and Political Legitimacy ( 0.36).

Global Summary of Changes in StateFragility: In keeping with the global sys-tem perspective of this report, we concludeour 2008 report with a global summary of

Global Report 2008

Foreign Policy Bulletin 11

changes in State Fragility during the periodof study. These changes are presented inFigure 8 below. The chart is organized likethe State Fragility Matrix (Table 1) in orderto facilitate comparison; vertical bars areread on the left-hand scale and the black-diamond icons are read on the right-handscale. In all, the global total of “fragilitypoints” assessed for the world’s 162 coun-tries (State Fragility Index; SFI) decreasedby 292 points (17.8 percent) from the 1995assessment. Breaking the Index into its twoprincipal components, we see that theimprovements were accounted for to amuch greater degree by gains in Effective-ness (180 points; 22.4 percent decrease)than gains in Legitimacy (112 points; 13.4percent decrease). This imbalance charac-terizes three of the four fragility dimen-sions.

Consistent with the dramatic decrease inglobal warfare presented in Figure 3 above,the Security Effectiveness category showsthe lowest total fragility score of the eightfragility categories: 91 total points in 2007(18.8 percent decrease from 1995). Theother seven categories contribute similarpoint totals to the global total in 2007,ranging from 157 points in the PoliticalEffectiveness category to 198 points in theEconomic Legitimacy category. SecurityLegitimacy (state repression) shows verymodest improvement since 1995 (17points; 9.2 percent decrease). PoliticalEffectiveness, reflecting the “third wave ofdemocratization” and stabilization/consoli-dation of more open political systems inthe Globalization Era, shows the most dra-matic improvement (75 points; 32.3 per-cent decrease in fragility). The PoliticalLegitimacy category shows strongimprovement over the period (48 points;21.8 percent decrease). The economicdimension shows only modest gains inEconomic Effectiveness (37 points; 16.7percent decrease) and no real change inEconomic Legitimacy at the global systemlevel, reflecting the general failure of pri-mary commodity producers (rentier states)to reinvest foreign exchange earnings intogreater manufacturing capacity. On thehand, steady progress can be noted in gen-eral improvements in Social Effectiveness(47 points and a 19.8 percent decrease infragility) and Social Legitimacy (45 points;19.6 percent decrease).

Concluding Remarks

The end of the Cold War ushered in an eraof globalization that is, for the first time,governed predominantly by democraticregimes; this marks a watershed moment inmodern human history and the beginningof a new world order. However, this newworld order encompasses a global systemthat, while improving steadily according toour analysis, lacks the capacity andresiliency that would provide a solid foun-dation for a stable and durable societal-sys-tem. The Global Report 2007 charted aglobal distribution of income among itsconstituent states characterized by highlyunequal regional development and profileda “system that is profoundly split into‘Haves’ (about 15% of the global popula-tion) and ‘Have-nots.’ [A system in which]the potential for polarization and factional-ism…is quite high and…will remain highfor the foreseeable future.”29 The currentGlobal Report on development under-scores Africa’s continuing malaise andhighlights a general imbalance betweengains in effectiveness and continuingdeficits in legitimacy. This imbalance isespecially problematic when considered inthe context of our growing investment indemocracy. While governance at the statelevel has become predominantly democrat-ic, the nature and quality of governance atthe global system level is challenged by itslarge number of anocratic states strugglingto maintain political stability and a smallnumber of classic autocracies controlling

some of its most vital and coveted oilreserves. Governance at the global level,whether formal or informal, is bound toreflect the nature and quality of the con-trasts inherent in the system. While violentconflict in the global system continues todiminish in total magnitude, some protract-ed societal wars continue to contradict thegeneral trend and defy proactive engage-ment, new wars break out regularly, andextremist violence and radical tactics drawcrucial resources away from critical sys-temic development. We believe that ourobservations have compiled an encourag-ing report on global system performance inthe emerging era of globalization. Howev-er, we caution that this progress has largelybeen purchased with a “peace dividend”that may now be largely spent. Furtherprogress and consolidation of the newworld order will surely demand a deter-mined and active commitment amongstates and citizens to reason and modera-tion in managing the challenges that defineour common predicament.

The State Fragility Index and Matrix2008

Having examined the general performanceof the Global System of States in the areasof security, governance, and developmentand discussed changes in the fragility ofstates since 1995, we conclude this GlobalReport 2008 with our assessments of thefragility of the system’s constituent units:the 162 independent (macro) states. Theidea of a using a matrix of effectiveness

Global Report 2008

12 Winter 2008

and legitimacy dimensions as a method forassessing state fragility was developed atthe University of Maryland’s IRIS center,in response to a research request from theUS Agency for International Development(USAID). Contributions to developing theidea were made by a number of people atIRIS and those involved in parallel effortsat USAID (including not only Marshall butalso Jack Goldstone, Dennis Wood, KarlSoltan, Ron Oakerson, Jonathan Houghton,Patrick Meagher, Lewis Rasmussen,Joseph Siegel, Clifford Zinnes and TjipWalker), but the current matrix of indica-tors was designed and applied by Marshalland Goldstone (2007). The idea is similarto other multi-dimensional schemes foraddressing state fragility, failure, or peace,including earlier indices developed byMarshall and Ted Gurr for the Peace andConflict series, models designed by the USGovernment’s Political Instability TaskForce (in which Marshall, Goldstone, andGurr continue to serve key roles), thosedeveloped by Frederick Barton and associ-ates at CSIS, metrics developed for theOffice of the Coordinator for Reconstruc-tion and Stabilization under CarlosPasquale in the State Department, the Fundfor Peace’s “Failed States Index,” and themore recent “Global Peace Index” devel-oped by the Economist Intelligence Unitfor the Vision of Humanity organization.

All of these schemes recognize thatassessing a state’s ability to win the loyaltyof its people depend on its performance inmultiple spheres, spanning governance,economic performance and opportunity,security, and delivery of social services.What the IRIS research team added was tomake explicit the need for governingregimes to exhibit both effectiveness andlegitimacy in its performance of thosetasks. That is, to achieve maximum stabili-ty a regime must both carry out the tasksexpected of a competent government, andmaintain legitimacy by being perceived asjust and fair in the manner it carries outthose tasks. A state may remain in a condi-tion of fragile instability if it lacks effec-tiveness or legitimacy in a number ofdimensions; however a state is likely tofail, or to already be a failed state, if it haslost both. The collaboration between theCenter for Global Policy at George MasonUniversity and the Center for Systemic

Peace makes the State Fragility assess-ments unique in that they are based on real-time monitoring of security and politicalconditions in each of the 162 countriesunder examination and they use well-respected and annually updated sources forthe Economic and Social assessments.These primary information resources makethe State Fragility Index as current andconsistent as possible. Table 1 presents theState Fragility Index and Matrix and thecorresponding ratings of the global sys-tem’s 162 countries in early 2008. It isaccompanied by Technical Notes that iden-tify the data sources used and describe howthe various indicators were constructed.

Global Report 2008

Foreign Policy Bulletin 13

14 Winter 2008

Table 1. State Fragility Index and Matrix 2008

S

tate

Fra

gilit

y In

dex

2007

Sta

te F

ragi

lity

Inde

x 20

01

Sta

te F

ragi

lity

Inde

x 19

95

Tra

ject

ory

1995

-200

7

Arm

ed C

onfli

ct In

dica

tor 2

007

Effe

ctiv

enes

s Sc

ore

2007

Leg

itim

acy

Scor

e 20

07

Sec

urity

Effe

ctiv

enes

s 20

07

Sec

urity

Leg

itim

acy

2007

Pol

itica

l Effe

ctiv

enes

s 20

07

Pol

itica

l Leg

itim

acy

2007

Eco

nom

ic E

ffect

iven

ess2

007

Eco

nom

ic L

egiti

mac

y 20

07

Soc

ial E

ffect

iven

ess

2007

Soc

ial L

egiti

mac

y 20

07

Reg

iona

l Effe

cts

Somalia 22 21 22 War 11 11 2 3 3 2 3 3 3 3 Mus Sudan 22 22 23 War 10 12 3 3 3 3 2 3 2 3 Mus Afghanistan 21 24 24 ▼ War 11 10 3 3 2 1 3 3 3 3 Mus Myanmar (Burma) 21 20 20 War 10 11 3 3 2 3 3 3 2 2 Chad 20 21 21 War 11 9 2 2 3 1 3 3 3 3 Mus Dem. Rep. of Congo 20 23 22 War 11 9 2 3 3 1 3 3 3 2 Afr Iraq 20 19 19 War 9 11 3 3 3 3 1 2 2 3 Mus Rwanda 20 21 21 * 10 10 2 2 2 2 3 3 3 3 Afr Burundi 19 21 21 X 11 8 2 3 3 0 3 3 3 2 Afr Liberia 19 22 23 ▼ X 11 8 2 2 3 0 3 3 3 3 Afr Nigeria 19 20 21 War 10 9 2 2 3 1 2 3 3 3 Afr Sierra Leone 19 23 22 * 11 8 2 1 3 1 3 3 3 3 Afr Central African Republic 18 15 13 ▲ War 9 9 1 3 2 2 3 1 3 3 Afr Ethiopia 18 19 19 War 9 9 2 2 1 2 3 2 3 3 Afr Guinea 18 20 18 * 9 9 0 1 3 3 3 2 3 3 Mus Angola 17 22 22 ▼ X 8 9 3 2 1 1 1 3 3 3 Afr Guinea-Bissau 17 18 16 * 9 8 0 1 3 1 3 3 3 3 Afr Zambia 17 17 18 8 9 0 1 2 2 3 3 3 3 Afr Burkina Faso 16 16 15 9 7 0 1 3 0 3 3 3 3 Afr Cameroon 16 19 18 * 6 10 0 2 2 3 2 3 2 2 Afr Congo-Brazzaville 16 18 14 X 6 10 1 2 2 2 1 3 2 3 Afr Eritrea 16 14 12 ▲ * 9 7 2 2 1 0 3 3 3 2 Afr Ivory Coast 16 18 15 X 9 7 1 2 3 0 2 2 3 3 Afr Niger 16 17 19 ▼ * 8 8 0 1 2 1 3 3 3 3 Mus Uganda 16 19 20 ▼ X 9 7 2 2 2 1 3 2 2 2 Afr Zimbabwe 16 17 15 * 8 8 0 2 3 3 3 1 2 2 Afr Algeria 15 17 17 X 5 10 2 3 1 3 1 3 1 1 Mus East Timor 15 * 8 7 1 2 2 1 3 3 2 1 Nepal 15 12 11 ▲ X 9 6 2 3 2 2 3 0 2 1 Pakistan 15 16 17 War 8 7 2 2 2 3 2 0 2 2 Mus Yemen 15 14 16 War 6 9 1 2 1 2 2 3 2 2 Mus Djibouti 14 14 17 * 6 8 0 1 2 1 2 3 2 3 Mus Iran 14 16 16 * 6 8 2 2 2 2 1 3 1 1 Mus Kenya 14 15 14 War 6 8 1 2 1 2 2 2 2 2 Afr Malawi 14 14 14 7 7 0 1 1 1 3 2 3 3 Afr Mauritania 14 17 17 ▼ * 7 7 0 1 3 2 2 2 2 2 Mus Mozambique 14 18 17 * 7 7 1 1 0 0 3 3 3 3 Afr Uzbekistan 14 11 12 ▲ 4 10 0 2 1 3 2 3 1 2 Mus Azerbaijan 13 15 18 ▼ * 6 7 2 1 2 2 1 2 1 2 Mus Benin 13 16 16 ▼ 7 6 0 1 1 0 3 2 3 3 Afr Comoros 13 16 13 7 6 0 0 2 1 3 3 2 2 Mus Gambia 13 14 13 8 5 0 1 3 0 3 2 2 2 Mus Haiti 13 13 15 X 8 5 1 2 2 1 3 0 2 2 Tajikistan 13 15 15 * 8 5 1 1 2 2 3 2 2 0 Mus Bangladesh 12 14 19 ▼ * 6 6 0 2 2 2 2 0 2 2 Mus Cambodia 12 14 16 ▼ * 9 3 2 1 3 0 2 0 2 2

Foreign Policy Bulletin 15

S

tate

Fra

gilit

y In

dex

2007

Sta

te F

ragi

lity

Inde

x 20

01

Sta

te F

ragi

lity

Inde

x 19

95

Tra

ject

ory

1995

-200

7

Arm

ed C

onfli

ct In

dica

tor 2

007

Effe

ctiv

enes

s Sc

ore

2007

Leg

itim

acy

Scor

e 20

07

Sec

urity

Effe

ctiv

enes

s 20

07

Sec

urity

Leg

itim

acy

2007

Pol

itica

l Effe

ctiv

enes

s 20

07

Pol

itica

l Leg

itim

acy

2007

Eco

nom

ic E

ffect

iven

ess2

007

Eco

nom

ic L

egiti

mac

y 20

07

Soc

ial E

ffect

iven

ess

2007

Soc

ial L

egiti

mac

y 20

07

Reg

iona

l Effe

cts

Ecuador 12 12 10 ▲ * 4 8 0 1 2 3 1 3 1 1 Equatorial Guinea 12 13 17 ▼ 4 8 0 1 2 1 0 3 2 3 Afr Ghana 12 10 15 * 7 5 0 1 2 1 3 2 2 1 Afr Laos 12 15 17 ▼ * 4 8 0 1 0 1 2 3 2 3 Madagascar 12 12 16 ▼ 7 5 0 1 2 1 3 2 2 1 Afr Sri Lanka 12 14 13 War 6 6 3 3 1 3 1 0 1 0 Tanzania 12 15 14 6 6 0 1 0 1 3 2 3 2 Afr Togo 12 15 19 ▼ 6 6 0 1 1 3 3 0 2 2 Afr Bolivia 11 9 12 4 7 0 1 1 2 1 2 2 2 Colombia 11 11 11 War 4 7 2 3 0 2 1 1 1 1 Egypt 11 12 14 ▼ * 4 7 0 2 2 3 1 1 1 1 Mus Guatemala 11 15 19 ▼ * 6 5 2 2 1 2 1 0 2 1 India 11 13 17 ▼ War 7 4 3 2 0 1 2 0 2 1 Kyrgyzstan 11 8 9 ▲ 7 4 0 1 2 2 3 1 2 0 Mus Philippines 11 11 14 ▼ War 7 4 3 2 2 1 1 0 1 1 Solomon Islands 11 14 9 X 7 4 1 0 2 0 2 3 2 1 Bhutan 10 12 16 ▼ * 3 7 0 0 0 3 1 1 2 3 Gabon 10 9 11 3 7 0 1 1 0 0 3 2 3 Afr Lebanon 10 11 11 X 4 6 2 1 1 3 0 0 1 2 Mus Lesotho 10 10 12 * 6 4 0 0 2 1 2 0 2 3 Afr Mali 10 15 19 ▼ * 7 3 0 0 1 0 3 0 3 3 Mus Papua New Guinea 10 13 15 ▼ * 4 6 0 1 0 0 2 3 2 2 Peru 10 11 14 ▼ * 5 5 2 1 1 1 1 2 1 1 Senegal 10 12 14 ▼ * 6 4 0 1 1 1 2 0 3 2 Mus Turkmenistan 10 9 10 3 7 0 1 1 1 1 3 1 2 Mus Venezuela 10 9 8 ▲ 2 8 0 2 1 2 0 3 1 1 China 9 9 11 * 4 5 2 2 0 2 1 0 1 1 Guyana 9 11 11 2 7 0 1 0 3 1 2 1 1 Indonesia 9 12 14 ▼ X 6 3 2 2 1 0 2 0 1 1 Mus Israel 9 8 7 ▲ War 3 6 2 3 1 3 0 0 0 0 Kazakhstan 9 9 9 3 6 0 1 1 2 1 2 1 1 Mus Moldova 9 10 13 ▼ * 4 5 0 1 1 3 2 1 1 0 South Africa 9 12 12 ▼ * 3 6 1 1 0 2 0 0 2 3 Afr Georgia 8 10 13 ▼ * 3 5 0 1 1 3 1 1 1 0 Honduras 8 10 12 ▼ * 4 4 0 1 1 1 1 1 2 1 Libya 8 9 10 2 6 0 1 2 1 0 3 0 1 Mus North Korea 8 10 11 ▼ 3 5 0 2 1 1 1 1 1 1 Paraguay 8 10 9 3 5 0 1 1 1 1 2 1 1 Saudi Arabia 8 11 10 X 1 7 1 1 0 3 0 3 0 0 Mus Swaziland 8 7 9 4 4 0 1 1 0 1 0 2 3 Afr Syria 8 11 12 ▼ 2 6 0 1 0 2 1 2 1 1 Mus Turkey 8 10 11 ▼ War 3 5 2 2 0 2 0 0 1 1 Mus Bosnia and Herzegovina 7 9 14 ▼ * 3 4 1 0 1 2 1 2 0 0 Mus Mongolia 7 9 9 4 3 0 1 0 0 2 1 2 1 Namibia 7 7 5 ▲ 3 4 0 1 0 0 1 0 2 3 Afr Nicaragua 7 10 13 ▼ * 3 4 0 1 0 0 2 2 1 1 Russia 7 9 9 X 3 4 2 2 1 0 0 2 0 0 Armenia 6 8 7 * 4 2 0 1 2 0 1 0 1 1 Serbia 6 10 10 ▼ * 3 3 0 1 2 2 1 0 0 0

16 Winter 2008

S

tate

Fra

gilit

y In

dex

2007

Sta

te F

ragi

lity

Inde

x 20

01

Sta

te F

ragi

lity

Inde

x 19

95

Tra

ject

ory

1995

-200

7

Arm

ed C

onfli

ct In

dica

tor 2

007

Effe

ctiv

enes

s Sc

ore

2007

Leg

itim

acy

Scor

e 20

07

Sec

urity

Effe

ctiv

enes

s 20

07

Sec

urity

Leg

itim

acy

2007

Pol

itica

l Effe

ctiv

enes

s 20

07

Pol

itica

l Leg

itim

acy

2007

Eco

nom

ic E

ffect

iven

ess2

007

Eco

nom

ic L

egiti

mac

y 20

07

Soc

ial E

ffect

iven

ess

2007

Soc

ial L

egiti

mac

y 20

07

Reg

iona

l Effe

cts

Tunisia 6 9 11 ▼ 2 4 0 1 1 2 0 0 1 1 Mus Brazil 5 5 6 1 4 0 2 0 1 0 0 1 1 El Salvador 5 8 9 ▼ * 3 2 1 1 0 0 1 0 1 1 Dominican Republic 5 7 7 1 4 0 1 0 1 0 1 1 1 Fiji 5 9 7 4 1 0 0 2 0 1 1 1 0 Jordan 5 6 8 ▼ 2 3 0 1 0 2 1 0 1 0 Mus Morocco 5 7 9 ▼ * 3 2 0 1 0 0 1 0 2 1 Mus Oman 5 6 6 2 3 0 0 2 0 0 3 0 0 Mus Panama 5 7 9 ▼ * 1 4 0 0 1 1 0 3 0 0 Qatar 5 6 5 1 4 0 0 1 0 0 3 0 1 Mus Romania 5 8 9 ▼ * 1 4 0 1 0 2 1 0 0 1 Thailand 5 6 8 ▼ War 3 2 1 2 1 0 0 0 1 0 Ukraine 5 5 5 3 2 0 1 1 1 1 0 1 0 Vietnam 5 8 9 ▼ * 3 2 0 1 0 1 2 0 1 0 Bahrain 4 4 8 ▼ 0 4 0 0 0 1 0 3 0 0 Mus Belarus 4 4 4 3 1 0 1 2 0 1 0 0 0 Botswana 4 5 7 ▼ 2 2 0 0 0 0 0 0 2 2 Afr Cuba 4 6 8 ▼ 0 4 0 1 0 1 0 2 0 0 Macedonia 4 4 5 2 2 0 1 1 1 1 0 0 0 Malaysia 4 6 7 ▼ 0 4 0 1 0 3 0 0 0 0 Mus Mexico 4 5 9 ▼ * 0 4 0 2 0 1 0 0 0 1 Montenegro 4 2 2 0 1 1 1 1 0 0 0 Trinidad 4 3 4 0 4 0 1 0 1 0 1 0 1 Albania 3 6 8 ▼ * 2 1 0 1 1 0 1 0 0 0 Mus Bulgaria 3 4 7 ▼ 1 2 0 1 0 1 1 0 0 0 Croatia 3 4 9 ▼ * 1 2 0 0 1 2 0 0 0 0 Cyprus 3 3 3 0 3 0 0 0 3 0 0 0 0 Jamaica 3 4 4 1 2 0 1 0 0 0 0 1 1 Kuwait 3 4 6 ▼ * 0 3 0 0 0 0 0 3 0 0 Mus Argentina 2 3 5 ▼ 0 2 0 1 0 0 0 1 0 0 Australia 2 1 2 0 2 0 0 0 0 0 2 0 0 Chile 2 2 4 * 0 2 0 0 0 0 0 2 0 0 Norway 2 2 1 0 2 0 0 0 0 0 2 0 0 United Arab Emirates 2 5 4 0 2 0 0 0 0 0 2 0 0 Mus United States 2 1 0 ▲ War 1 1 1 1 0 0 0 0 0 0 Belgium 1 0 0 0 1 0 0 0 1 0 0 0 0 Czech Republic 1 1 1 0 1 0 0 0 1 0 0 0 0 Estonia 1 1 5 ▼ 0 1 0 0 0 1 0 0 0 0 France 1 0 0 0 1 0 0 0 1 0 0 0 0 Greece 1 0 1 0 1 0 1 0 0 0 0 0 0 Italy 1 1 1 1 0 0 0 1 0 0 0 0 0 New Zealand 1 1 1 0 1 0 0 0 0 0 1 0 0 Singapore 1 2 2 0 1 0 0 0 1 0 0 0 0 Slovak Republic 1 2 4 ▼ 0 1 0 0 0 1 0 0 0 0 Spain 1 0 2 0 1 0 1 0 0 0 0 0 0 Switzerland 1 1 1 0 1 0 0 0 1 0 0 0 0 Taiwan 1 1 2 1 0 0 0 0 0 1 0 0 0 Uruguay 1 0 1 0 1 0 0 0 0 0 1 0 0 Austria 0 0 0 0 0 0 0 0 0 0 0 0 0

Foreign Policy Bulletin 17

Sta

te F

ragi

lity

Inde

x 20

07

S

tate

Fra

gilit

y In

dex

2001

Sta

te F

ragi

lity

Inde

x 19

95

Tra

ject

ory

1995

-200

7

Arm

ed C

onfli

ct In

dica

tor 2

007

Effe

ctiv

enes

s Sc

ore

2007

Leg

itim

acy

Scor

e 20

07

Sec

urity

Effe

ctiv

enes

s 20

07

Sec

urity

Leg

itim

acy

2007

Pol

itica

l Effe

ctiv

enes

s 20

07

Pol

itica

l Leg

itim

acy

2007

Eco

nom

ic E

ffect

iven

ess2

007

Eco

nom

ic L

egiti

mac

y 20

07

Soc

ial E

ffect

iven

ess

2007

Soc

ial L

egiti

mac

y 20

07

Reg

iona

l Effe

cts

Canada 0 0 0 0 0 0 0 0 0 0 0 0 0 Costa Rica 0 0 2 0 0 0 0 0 0 0 0 0 0 Denmark 0 0 0 0 0 0 0 0 0 0 0 0 0 Finland 0 0 0 0 0 0 0 0 0 0 0 0 0 Germany 0 0 2 0 0 0 0 0 0 0 0 0 0 Hungary 0 0 1 0 0 0 0 0 0 0 0 0 0 Ireland 0 0 0 0 0 0 0 0 0 0 0 0 0 Japan 0 0 0 0 0 0 0 0 0 0 0 0 0 Latvia 0 1 5 ▼ 0 0 0 0 0 0 0 0 0 0 Lithuania 0 0 2 0 0 0 0 0 0 0 0 0 0 Mauritius 0 1 1 0 0 0 0 0 0 0 0 0 0 Netherlands 0 0 0 0 0 0 0 0 0 0 0 0 0 Poland 0 0 2 0 0 0 0 0 0 0 0 0 0 Portugal 0 0 0 0 0 0 0 0 0 0 0 0 0 South Korea 0 2 4 ▼ 0 0 0 0 0 0 0 0 0 0 Slovenia 0 0 1 0 0 0 0 0 0 0 0 0 0 Sweden 0 0 1 0 0 0 0 0 0 0 0 0 0 United Kingdom 0 2 2 * 0 0 0 0 0 0 0 0 0 0

and introduced in the “Global Report onConflict, Governance, and State Fragility2007.” In order to standardize proceduresfor scoring each of the eight componentindicators so as to make the indicators andindices comparable across time, we haveset threshold values for the categoricalfragility scores based on quintile cut-pointsderived from values in a baseline year(2004). This methodology effects continu-ous measures used for Economic Effective-ness (GDP per capita in constant 2000 USdollars); Economic Legitimacy (manufac-turing exports as a percent of merchandiseexports); Social Effectiveness (humandevelopment indicator; HDI); and SocialLegitimacy (infant mortality rate); baselinespecifications are provided in the relevantindicator explanations that follow. In addi-tion, a fourth indicator has been added tothe calculation of the Political LegitimacyScore (scores for previous years have beenrecalculated); see below.

Security Indicators

Security Effectiveness Score: Total Resid-ual War, a measure of general security andvulnerability to political violence, 1983-2007 (25 years). Source: Monty G. Mar-shall, Major Episodes of Political Violence,1946-2007, (www.systemicpeace.org),variable name “actotal.” The formula tocalculate this score is based on twoassumptions: (1) the residual effects of lowlevel and/or short wars diminish relativelyquickly; and (2) the residual effects of seri-ous or protracted wars diminish graduallyover a 25-year period. Three indicators areused to calculate each country’s “residualwar” score (reswartot): warsum1-4 (sum ofannual scores for all wars in which thecountry is directly involved for each con-tinuous period of armed conflict);yrnowar1-3 (interim years of “no war”between periods of armed conflict); andyrpeace (years of peace, or no war, sincethe end of most recent war period). Forstates with one war episode: reswartot =warsum – [yrpeace + ( .04yrpeace x war-sum1)]. For countries with multiple periodsof war, a reswar value is calculated foreach, in chronological order. Thus, for astate with two episodes of war, to calculatethe first episode: reswar1 = warsum1 –[yrnowar1 + ( .04yrnowar1 x warsum1)];

and for the second episode: reswartot =(reswar1 + warsum2) – {yrpeace +[.04yrpeace x (reswar1 + warsum1)]}; andso on. Any negative residual war (reswar)scores are converted to zero before calcu-lating additional residual war scores. Thefinal reswartot value is then converted to afour-point fragility scale, where: 0 = 0; 1 =0.1-15; 2 = 15.1-100; and 3 = greater than100.

Security Legitimacy Score: State Repres-sion, a measure of state repression, 1993-2006. Source: Mark Gibney, Linda Cor-nett, and Reed Wood, Political Terror Scale(PTS; www.politicalterrorscale.org). ThePTS provides separate annual indicatorsdrawn from U.S. State Department andAmnesty International reports; each indica-tor is coded on a five-point scale, from 1:“no repression” to 5: “systemic, collectiverepression.” To calculate the 2007 staterepression score, we calculate the follow-ing: (1) nine-year average, 1993-2001; (2)four-year average, 2002-2005; and (3)most recent value, 2006; the three, meanindicators are then compared according toa fragility categorization: 0 = 1.0-2.0; 1 =2.1-3.0; 2 = 3.1-4.0; and 3 = greater than4.0. If the most recent year value agreeswith the previous four-year average, thenthese two means are used to identify therepression category. When the most recentyear score is not in agreement with the pre-vious period, then the earlier nine-yearmean is used to help determine a more gen-eral pattern in state repression. Historicaltreatments, that is, calculations of SecurityLegitimacy Scores for previous years, arefurther aided by reference to patterns in“future” PTS values. The exact year ofchange in the general practice of staterepression and, so, the Security LegitimacyScore can be more confidently identified inthe historical treatment.

Political Indicators

Political Effectiveness Score:Regime/Governance Stability, 1993-2007.Sources: Monty G. Marshall, Keith Jag-gers, and Ted Robert Gurr, Polity IV Pro-ject: Political Regime Characteristics andTransitions, 1800-2007; Henry S. Bienenand Nicolas van de Walle, LeadershipDuration (updated by Monty G. Marshall);

and Monty G. Marshall and Donna RamseyMarshall, Coups d’Etat, 1960-2007,datasets (www.systemicpeace.org). Threeindicators are used to calculate theRegime/Governance Stability score:Regime Durability (Polity IV, 2007); Cur-rent Leader’s Year’s in Office (LeadershipDuration, 2007); and Total Number ofCoup Events 1993-2007, including suc-cessful, attempted, plotted, alleged coupsand forced resignations or assassinations ofchief executives (Coups, 1993-2007), butnot including coup events associated withPolity adverse regime changes (these majorregime changes cause the “durability”score to be reset to “0” and, so, would bedouble-counted, see above). These indica-tors are scored such that: Durability < 10years = 1; Leader Years in Office > 12years = 1; and Total Coup Events: 1-2 = 1and >2 = 2. These indicators are then addedto produce the Regime/Governance Stabil-ity score (scores of 4 are recoded as 3).Note: Countries coded in the Polity IVdataset as an “interregnum” (i.e., total ornear total collapse of central authority,

77) for the current year are scored 3 onthe Political Effectiveness indicator.

Political Legitimacy Score: Regime/Gov-ernance Inclusion, 2007. Sources: PolityIV, 2007; Ted Robert Gurr, Monty G. Mar-shall, and Victor Asal, Minorities at RiskD i s c r i m i n a t i o n(www.cidcm.umd.edu/mar); and TedRobert Gurr and Barbara Harff, Elite Lead-ership Characteristics datasets (updated byMonty G. Marshall). In the 2007 report,four indicators were used to determine theRegime/Governance Inclusion score: Fac-tionalism (Polity IV, parcomp value 3 = 1);Ethnic Group Political Discriminationagainst more than 5% of Population (Dis-crimination, 2007: POLDIS values 2, 3, 4 =1); Political Salience of Elite Ethnicity(Elite Leadership Characteristics, 2007:ELETH values 1 or 2 = 1); and Polity Frag-mentation (Polity IV, fragment valuegreater than 0 = 1). To these indicators, wehave added Exclusionary Ideology of Rul-ing Elite (Elite Leadership Characteristics,2007: ELITI value 1 = 1). The PoliticalLegitimacy Score is calculated by addingthese five indicators; scores of 4 or 5 (rare)are recoded as 3.

Global Report 2008

18 Winter 2008

Economic Indicators

Economic Effectiveness Score: GrossDomestic Product per Capita (constant2000 US$), 2000-2006. Source: WorldBank, World Development Indicators,2007 (www.worldbank.org/data). Threeindicators are considered in gauging eco-nomic effectiveness: five-year averageGDP per capita, 2000-2004; the mostrecent year’s GDP per capita, 2006 (bothmeasured in constant US dollars); and thefifteen-year average GDP per capitagrowth rate, 1992-2006. Values for thefive-year average and most recent yearvalue are coded into a four-point fragilityscale, based on cut-points derived from theupper threshold values for the lower threequintiles of GDP per capita in a baselineyear (2004). The standardized categoriesare as follows: 3 = less than or equal to$400.00; 2 = $400.01 to $1000; 1 =$1000.01 to $2500.00; and 0 = greater than$2500. The five-year and most recent yearcategories are compared. If they are thesame, that category value is assigned as theEconomic Effectiveness Score. If the twovalues differ or are near the borderlinevalue separating categories, the fifteen-year income growth indicator is used toassign the final score: selecting the highercategory if long-term growth is negative orthe lower category if long-term growth ispositive.

Economic Legitimacy Score: Share ofExport Trade in Manufactured Goods,1993-2005. Source: UN Development Pro-gramme, Structure of Trade, 2007, andWorld Bank, World Development Indica-tors (WDI), 2007, (manufacturing as a per-centage of merchandise exports). Merchan-dise exports include two classes of prod-ucts: manufactured goods and primarycommodities; low percentage of manufac-tured goods indicates a high reliance onprimary commodities for foreignexchange. The annual values of this vari-able are examined to ensure that the mostrecent annual value is a representativevalue within the established range for thatcountry. The manufacturing percentage ofmerchandise exports is then converted to afour-point fragility score, where: 3 = lessthan or equal to 10; 2 = greater than 10 andless than or equal to 25; 1 = greater than 25

and less than or equal to 40; and 0 = greaterthan 40.

Social Indicators

Social Effectiveness Score: Human CapitalDevelopment, 2005. Source: UNDPHuman Development Report 2007/2008,Human Development Index (HDI), 2005(www.undp.org). Reported HDI values areconverted according to a four-point fragili-ty scale based on the cut-points of thelower three HDI quintiles in the baselineyear, 2004. The Social Effectiveness Scoreis assigned as follows: 3 = less than orequal to .500; 2 = greater than .500 and lessthan or equal to .700; 1 = greater than .700and less than or equal to .800; and 0 =greater than .800.

Social Legitimacy Score: Human CapitalCare, 2005. Source: US Census Bureau,International Data Base, 2007, (IDB;www.census.gov/ipc/www/idb), InfantMortality Rate, 2005. This indicator isbased on the infant mortality rate (numberof deaths of infants under one year of agefrom a cohort of 1,000 live births), withvalues converted to a four-point fragilityscale based on the upper cut-points of thelower three quintiles of the infant mortalityrates in the baseline year, 2004. The SocialLegitimacy Score is assigned as follows: 3= greater than 75.00; 2 = less than or equalto 75.00 and greater than 45.00; 1 = lessthan or equal to 45.00 and greater than20.00; and 0 = less than or equal to 20.00.These scores are then adjusted according toranking comparisons between the country’sincome level (GDP per capita) and humancapital development (HDI). If the country’sHDI ranking among the 162 countries list-ed is more than twenty-five places aboveits GDP per capita ranking (meaning it pro-vides better human capital care thanexpected by its level of income) the SocialLegitimacy Score (fragility) is lowered byone point. If HDI ranking is more thantwenty-five places below GDP per capitaranking, the fragility score is increased byone point.

Note on Historical State FragilityIndices and Scores

The historical SFI scores presented here(i.e. 1995, 2001) were calculated using thatyear’s values for each of the indicators,whereas the current year scores (i.e., 2007)necessarily rely on the values of the mostrecent year reported (this limitation affectsdata drawn from the World Bank andUNDP only). For example, while socialeffectiveness for 2007 relies on 2005 HDI,which is the value reported in the 2007-08Human Development Report, social effec-tiveness for 1995 is based on 1995 HDI,reported in the 1997 Human DevelopmentReport. Our best estimations of currentyear scores are revised when current yearvalues become available.

Global Report 2008

Foreign Policy Bulletin 19

Notes:

1. The Global Report 2008 appears in the March2008 issue of Foreign Policy Bulletin (CambridgeUniversity Press); www.foreignpolicybulletin.com.The Report is prepared at the Center for GlobalPolicy at George Mason University;globalpolicy.gmu.edu. The lead author can be con-tacted by email at mmarsha5@gmu.edu.

2. Ted Robert Gurr, Monty G. Marshall, and DeepaKhosla, Peace and Conflict 2001: A Global Surveyof Armed Conflicts, Self-Determination Move-ments, and Democracy (College Park, MD: Centerfor International Development and Conflict Man-agement, 2000), p.9.

3. This general reaction was intimated by AndrewMack who was heading the Secretary General’sStrategy Unit at that time and who personally cir-culated our findings through UN channels.

4. Large portions of descriptive text in the “systems-analysis,” “governance,” and “armed conflict” sec-tions have been carried over from last year’s Glob-al Report to provide basic information on howglobal trends have been measured and charted. SeeMonty G. Marshall and Jack Goldstone. 2007.“Global Report on Conflict, Governance, and StateFragility 2007: Gauging System Performance andFragility in the Globalization Era.” Foreign PolicyBulletin 17.1 (Winter 2007):3-21.

5. See Monty G. Marshall, Third World War: System,Process, and Conflict Dynamics (Lanham, MD:Rowman & Littlefield, 1999) for a more detailedexplanation of the societal-systems approach topolitical analysis.

6. The Polity IV data set was originally designed byTed Robert Gurr and coded by Erica Klee Gurr andKeith Jaggers; the Polity IV Project is now direct-ed by Monty G. Marshall at the Center for GlobalPolicy (CGP), George Mason University, and theCenter for Systemic Peace. It has annually codedinformation on the qualities of political institutionsfor all independent countries (not including micro-states) from 1800 through 2006 and is updatedannually by Marshall and Jaggers; the most recentyear (2007) values have been estimated by Mar-shall for use in Global Report analyses. The PolityIV data series is available on the Center for Sys-temic Peace Web site atwww.systemicpeace.org/polity/polity4.htm.

7. Also included in the anocracy category are coun-tries that are undergoing transitional governments(coded “-88” in the Polity IV dataset), countrieswhere central authority has collapsed or lost con-trol over a majority of its territory (coded “-77” inthe dataset), and countries where foreign authori-ties, backed by the presence of foreign forces, pro-vide a support structure for maintaining localauthority (coded “-66”).

8. The Polity IV dataset does not include informationon micro-states; a state must have reached a totalpopulation of 500,000 to be included.

9. In the Polity IV data set, “factionalism” is definedoperationally as a code “3” on the “competitive-ness of political participation” (PARCOMP). The

Political Instability Task Force (PITF; formerlyknown as the State Failure Task Force) was createdin 1994 at the request of senior policy makers inthe US Government; it is tasked with developingdata-driven, global and regional models to helpexplain, and anticipate, the emergence of seriouspolitical instability situations in the world’s inde-pendent states.

10. The Political Instability Task Force (PITF) find-ings are reported regularly; the problem of faction-alism was first identified in the most recent (PhaseV) findings. Task Force reports, including papersdetailing the Phase V findings, can be accessedfrom the PITF Web site at URL: http://globalpoli-cy.gmu.edu/pitf.

11. Interstate and civil wars must have reached amagnitude of over 500 directly-related deaths to beincluded in the analysis. The magnitude of eachmajor armed conflict is evaluated according to itscomprehensive effects on the state or states direct-ly affected by the warfare, including numbers ofcombatants and casualties, size of the affected areaand dislocated populations, and extent of infra-structure damage. It is then assigned a single scoreon a ten-point scale; this value is recorded for eachyear the war remains active. See Monty G. Mar-shall, “Measuring the Societal Effects of War,”chapter 4 in Fen Osler Hampson and David Mal-one, eds., From Reaction to Conflict Prevention:Opportunities for the UN System (Boulder, CO:Lynne Rienner, 2002) for a detailed explanation ofthe methodology used. A full list of the “majorepisodes of political violence” from which the datafor figure 3 is compiled is posted on the Center forSystemic Peace (CSP) Web site at www.systemic-peace.org/warlist.htm. The CSP Web site alsotracks other aspects of global and regional armedconflict trends; see the “Conflict Trends” sectionon the CSP Web site atwww.systemicpeace.org/conflict.htm.

12. Scattered and sporadic terrorist attacks haveoccurred throughout the predominately Muslimregion stretching from northwestern Africa throughthe Middle East and in the Muslim areas of south-eastern Asia and Oceania. However, there is scantevidence that Islamic militants have established a“global reach” capability for systematic and sus-tained attacks beyond the Muslim region itself.

13. The “de facto” separation of Kosovo from Serbiawas accomplished by fiat following a seventy-eightday aerial bombardment of Serbia by NATO airforces; its separation was administered by the UNMission in Kosovo and enforced by a NATO-ledpeacekeeping force (KFOR). The Kosovo Assem-bly declared independence for the territory on Feb-ruary 17, 2008; its declaration was accepted bymany states but rejected by Serbia and many otherstates.

14. The greater problem of direct, intentional, andsystematic attacks on civilian populations is mostoften attributed to “state terror” or state-sponsored“death squads.” These types of atrocities have notbeen uncommon during the contemporary periodand fall somewhere between the conceptual terms

“lethal repression” and “genocide or politicide.”For a contemporary survey of the problem of“genocide and politicide,” see Barbara Harff, “NoLessons Learned from the Holocaust? AssessingRisks of Genocide and Political Mass Murder since1955,” American Political Science Review 97.1(2003): 57-73. Indirect (collateral) attacks on civil-ian populations can be similarly terrifying duringwarfare characterized by substantial power discrep-ancies between opposing militaries and militias,particularly when relatively indiscriminate aerialbombardment is utilized by state forces againstnon-state groups.

15. A full list of HCTB events used in this analysiscan be found with the “Conflict Trends” report onthe Center for Systemic Peace Web site (note9).

16. This simple accounting of reported deaths doesnot intend to discount the far greater number ofpeople seriously injured and otherwise traumatizedby such attacks. There are rarely any follow-upreports on the newswires that provide informationon the fate of people wounded in specific attacks(with the notable exception of attacks in the US,UK, and Spain). Given the generally poor condi-tion of medical and health services in the areasmost affected by HCTB attacks, it is reasonable toassume that a relatively large proportion of thoseseriously wounded in HCTB attacks would diefrom their wounds.

17. This term “Muslim rage” references a term“Black rage” associated with a social psychologi-cal condition of intense anger and frustration thatinduces irrational behavior and an ultimately, self-destructive paranoia; the term was introduced inthe 1968 book, Black Rage, by psychiatristsWilliam H. Grier and Price M. Cobbs.

18. Marshall and Goldstone 2007, p. 11.19. Marshall and Goldstone 2007, p. 11.20. GDP per capita figures are taken from the World

Bank, World Development Indicators 2007 (con-stant 2000 US$; values listed for year 2006). Weuse this measure of income due to its comparabili-ty at the global level of analysis. Income measuredas “purchasing power parity” may provide a moreaccurate measure of the relative value of income atthe local level of analysis.

21. Serbia is considered to be the successor state forboth Yugoslavia and Serbia and Montenegro. Assuch, it is identified as a country with consistentlypositive change in its State Fragility Index score:10 (Yugoslavia) in 1995, 10 (Serbia and Montene-gro) in 2001, and 6 (Serbia) in 2007.

22. Israel and Mauritius are regionally isolated and,so, are not included in the regional calculations.East Timor (2002) and Montenegro (2006) are newstates that have no earlier fragility scores; thesetwo countries are removed from the comparativeregional analyses.

23. Nineteen countries comprise the North Atlanticregion: Austria, Belgium, Canada, Cyprus, Den-mark, Finland, France, Germany, Greece, Ireland,Italy, Norway, Netherlands, Portugal, Spain, Swe-den, Switzerland, United Kingdom, and the UnitedStates.

Global Report 2008

20 Winter 2008

24. Twenty countries comprise the Former-Socialistregion: Armenia, Belarus, Bulgaria, Croatia, CzechRepublic, Estonia, Georgia, Hungary, Latvia,Lithuania, Macedonia, Moldova, Montenegro,Poland, Romania, Russia, Serbia, Slovak Republic,Slovenia, and Ukraine. Montenegro became anindependent state in 2006 and, so, is not includedin the comparative regional analyses.

25. The Latin America region comprises twenty-threecountries: Argentina, Bolivia, Brazil, Chile,Colombia, Costa Rica, Cuba, Dominican Republic,Ecuador, El Salvador, Guatemala, Guyana, Haiti,Honduras, Jamaica, Mexico, Nicaragua, Panama,Paraguay, Peru, Trinidad & Tobago, Uruguay, andVenezuela.

26. The (non-Muslim) East and South Asia regionconsists of twenty-three countries: Australia,Bhutan, Cambodia, China, East Timor, Fiji, India,Japan, Laos, Mongolia, Myanmar (Burma), Nepal,New Zealand, North Korea, Philippines, PapuaNew Guinea, Singapore, Solomon Islands, SouthKorea, Sri Lanka, Taiwan, Thailand, and Vietnam.East Timor became an independent state in 2002and, so, is not included in the comparative analy-ses.

27. Muslim Countries are identified as countries inwhich Muslim confessional groups comprise fiftypercent or more of the country’s total population.This region is defined by forty-two countries span-ning from West Africa to the Pacific Ocean,including Afghanistan, Albania, Algeria, Azerbai-jan, Bahrain, Bangladesh, Bosnia, Chad, Comoros,Djibouti, Egypt, Gambia, Guinea, Indonesia, Iran,Iraq, Jordan, Kazakhstan, Kuwait, Kyrgyzstan,Lebanon, Libya, Malaysia, Mali, Mauritania,Morocco, Niger, Oman, Pakistan, Qatar, SaudiArabia, Senegal, Somalia, Sudan, Syria, Tajikistan,Tunisia, Turkey, Turkmenistan, United Arab Emi-rates, Uzbekistan, and Yemen.

28. Non-Muslim Africa comprises thirty-three coun-tries: Angola, Benin, Botswana, Burkina Faso,Burundi, Cameroon, Central African Republic,Republic of Congo, Democratic Republic ofCongo, Equatorial Guinea, Eritrea, Ethiopia,Gabon, Ghana, Guinea-Bissau, Ivory Coast,Kenya, Lesotho, Liberia, Madagascar, Malawi,Mozambique, Namibia, Nigeria, Rwanda, SierraLeone, South Africa, Swaziland, Tanzania, Togo,Uganda, Zambia, and Zimbabwe.

29. Marshall and Goldstone 2007, p. 11.

Global Report 2008

Foreign Policy Bulletin 21