the political economy of war and peace in afghanistan

The Political Economy of War and Peace in

Afghanistan

BARNETT R. RUBIN *

Center on International Cooperation, New York University, NY, USA

Summary. Ð The 20-year old Afghan con¯ict has created an open war economy, a�ectingAfghanistan and surrounding areas. Not only has Afghanistan become the worldÕs largest opiumproducer and a center for arms dealing, but it supports a multibillion dollar trade in goodssmuggled from Dubai to Pakistan. This criminalized economy funds both the Taliban and theiradversaries. It has transformed social relations and weakened states and legal economiesthroughout the region. Sustainable peace will require not just an end to ®ghting and a politicalagreement but a regional economic transformation that provides alternative forms of livelihoodand promotes accountability. Ó 2000 Elsevier Science Ltd. All rights reserved.

Key words Ð Asia, Afghanistan, war, smuggling, Islam, refugees

Quid est aliud omnia ad bellum civile hosti arma larg-iri, primum nervos belli, pecuniam in®nitam? (What isthis but to lavish on an enemy all the weapons for civilwar, ®rst of all the sinews of war, unlimited money?)

Cicero M. T. (1951). The Fifth Phillipic of M. TulliusCicero against M. Antonius (v.ii.5, pp. 260±261).(Walter C. A. Ker, Trans., Cicero Philippics). Cam-bridge, MA: Harvard University Press

1. INTRODUCTION

Classical interstate war may be, as vonClausewitz wrote, nothing else but the pursuitof politics with the admixture of other means,but the pursuit of politics through both peace-ful and violent means requires money. Politicalleaders speak in public about their ideals andgoals, but much of their activity is devoted toraising resources to exercise power and rewardsupporters or themselves. How political leadersraise and distribute these resources determinesthe outcome of their acts, as much as if notmore than their stated goals and intentions.

The dominant current form of war is neitherClausewitzian interstate war nor classic civilwar (government versus insurgency), buttransnational war involving a variety of o�cialand uno�cial actors, often from several states.Such wars create conditions for economicactivity, though often of a predatory nature,

and such economic returns to the use of vio-lence may both provoke such wars and nourishinterests that perpetuate them. A few actorspro®t, while most have no say in the develop-ment of their own society. Peacemakingrequires not only political negotiations buttransforming the war economy into a peaceeconomy and creating institutions foraccountability over economic and politicaldecision making (Collier & Hoe�er, 1999;Collier, 1999; Collier, Hoe�er & S�oderbom,1999; Holsti, 1995; Jean & Ru®n, 1996; Kaldor,1999; Keen, 1998; Reno, 1998).

World Development Vol. 28, No. 10, pp. 1789±1803, 2000Ó 2000 Elsevier Science Ltd. All rights reserved

Printed in Great Britain0305-750X/00/$ - see front matter

PII: S0305-750X(00)00054-1www.elsevier.com/locate/worlddev

* I presented an earlier version of this paper to the

Afghanistan Support Group, Stockholm, Sweden, June

21 1999. I thank UN Under Secretary General Lakhdar

Brahimi, and the Ministry of Foreign A�airs of Sweden

for that opportunity. I owe special thanks to Ashraf

Ghani for many exchanges of ideas over the years.

David Mans®eld provided me with reports from the

United Nations Drug Control Program and his own

observations. I also bene®ted from comments and

assistance from Sadiq Ahmed, William Byrd, Anthony

Davis, Anders F�ange, Bernard Frahi, M. Jamal Hani®,

R. Scott Horton, Didier Leroy, William Maley, Zareen

F. Naqvi, Ahmed Rashid, Michael von den Schulen-

burg, and two anonymous reviewers, but they, like my

ex-employer, the Council on Foreign Relations, bear no

responsibility for the views expressed herein, which are

mine alone. Final revision accepted: 25 February 2000.

1789

The war economy of Afghanistan exempli®esthis phenomenon. Devastated Afghanistan hasbecome both the worldÕs leading producer ofopium (75% of world production in 1999) and atransport and marketing corridor where armedgroups protect a region-wide arbitraging centerwhere pro®ts are made o� policy-inducedprice di�erentials. The region in questionincludes Dubai, the worldÕs largest duty-freeshopping mall; Pakistan, a state where the twoISIsÐthe Directorate of Inter-Services Intelli-gence and import-substitution industrializa-tionÐhave created a highly armed and corruptsociety where economic interest in evading hightari�s and the imperatives of covert actioncombine to undermine enforcement of ®scalrules and public order; Iran, where subsidizedgasoline sells for three cents a liter; Afghani-stan, a barely governed territory that includesthe remnants of a road network that links Iran,Central Asia, and Pakistan; Central Asianstates opened to the world without institutionsto govern markets; and linked wars in Tajiki-stan, Afghanistan, and Kashmir, plus a grow-ing insurgency in the Ferghana Valley inUzbekistan (Nunn, Lubin & Rubin, 1999).

This economy developed in response to thedemands of warlords for resources and of theAfghan people for survival in a countrydevastated by over 20 years of war. Over amillion have died in a country whose popula-tion is now estimated at 26 million, and theproportion of disabled in the population maybe the worldÕs highest. Afghanistan ranks at thebottom of all measures of human welfare(Table 1), and illicit activities have become keyelements of its peopleÕs survival strategies.Though most elements of this war economyhad developed before the rise of the IslamicMovement of Taliban, the consolidation bythat group of its ``Islamic Emirate of Afghani-stan'' over most of the country has suppressedlocalized predation, enabling this group torealize what Collier and Hoe�er (1999, p. 3)call ``economies of scale'' in looting as the wareconomy has grown. Depending on both deci-sions by the Taliban and international policy,this development could prove a prelude to statebuilding, with potential for more legitimategovernance and development, or to a morerapidly expanding criminalized economy undera stigmatized leadership.

Table 1. Measures of humanitarian emergency in Afghanistana

Indicators Afghanistan SouthAsia

Developingcountries

Industrialcountries

Human development index rank(out of 174)b

169 N/A N/A N/A

Population % with access toHealth care (1985±93)b 29 65 79 100Safe water (1990±95)c 12 (rural 5,

urban 39)77 69 100

Daily calorie supply per capita (1992)c 1,523 2,356 2,546 3,108Infant mortality per 1,000 live births(1993)c

165 85 70 N/A

Under ®ve mortality per 1,000 livebirths (1993)d

257 122 101 N/A

Maternal mortality per 100,000 livebirths (1993)

1,700e or 640f 469 351 10

Life expectancy at birth in years(1993)b

44 60 62 76

Adult literacy rate (%, 1993)b;c 28 (men 45,women 14)

48 68 98

a Sources: All comparative data from other regions are from source (b) below. One indicator of humanitarianemergency in Afghanistan is the collapse of institutions able to produce such statistics. Hence, unlike suchpresumably better governed countries as Sierra Leone and Burundi, Afghanistan has not been listed in the standardsource for such data, UNDPÕs Human Development Report, since 1996.b UNOCHA (1996, p. 4); citing UNDP, Human Development Report 1996.c UNOCHA (1997, p. 4); citing UNDP, Human Development Report 1997.d UNOCHA (1997); citing UNICEF, State of the world's children report 1996.e UNOCHA (1997); citing Study by UNICEF/World Health Organization, 1996.f UNDP (1997).

WORLD DEVELOPMENT1790

This illicit economy is not con®ned toAfghanistan. Through the development of anAfghan diaspora linked to neighboring societ-ies, the opening of borders, and lack of customsenforcement in many areas, the Afghan wareconomy has generated a pattern of regionaleconomic activity and associated social andpolitical networks that compete with andundermine legal economies and states. Thisregional economy is in turn linked through thedrug and arms trade to globalized crime(Castells, 1998, pp. 166±205). Transformationof this war economy is thus essential not onlyto Afghanistan but to neighboring regions andthe world.

2. STAGES OF THE WAR ECONOMY'SDEVELOPMENT

Before the outbreak of war in Afghanistan in1978, a gradually expanding, foreign-supportedstate coexisted with a rural sector based onsubsistence agriculture and pastoralism. TheCommunist coup of April 1978, the Sovietinvasion of December 1979, and the reaction tothese by the United States, Pakistan, SaudiArabia, China, and others, destroyed thissystem. Both sides of the war depended onmilitary technology and cash provided byforeign sponsors. The pipelines for arms andhumanitarian aid supplied capital to build upregional smuggling networks.

After the Soviet withdrawal in February1989, both Soviet and Western aid decreased,and both ended with the dissolution of theUSSR at the end of 1991. Regional powers(mainly Pakistan, but also Iran and Russia)stepped in, but local commanders and somereturning refugees pursued new survival strat-egies in a context of highly fragmented powerand no e�ective central state (UNDCP, 1999a,p. 1). Predation by commanders, opium culti-vation by peasants, and smuggling to Pakistanand elsewhere constituted adaptations to thishigh-risk environment.

Predation by commanders imposed heavycosts on commerce, blocked PakistanÕs accessto Central Asia, and prevented consolidation ofan Islamic or any other order. Hence a coali-tion of Pakistani authorities, Afghan andPakistani traders, and ultra-conservativeAfghan and Pakistan religious leaders createdthe Taliban. The Taliban, a transnationalmovement bene®ting from social capitalcreated in madrasas (Islamic academies) in the

Afghan±Pakistan border areas during 15 yearsof Afghan dispersion, managed by 1998 toconsolidate control over nearly all the countryÕsroads, cities, airports, and customs posts,thereby drastically lowering the cost and risk oftransport and consolidating AfghanistanÕsposition at the center of a regional war economy.

(a) Prewar economy, state, and society

In the 1970s Afghan society was splitbetween a rural, largely subsistence economyand an urban economy dependent on a statethat in turn drew most of its income from linksto the international state system and market.About 85% of the population lived from therural economy. As late as 1972, the casheconomy constituted less than half of the total(Rubin, 1995a, pp. 62±75; Fry, 1974, pp. 135±162). Government expenditure consumed lessthan 10% of the whole, and in the 1960s foreignaid accounted for over 40% of the budget.When aid declined, it was replaced by exportsof natural gas from northern Afghanistan tothe Soviet Union. Such rentier incomescontinued to ®nance nearly half the budget(Rubin, 1995a, pp. 296±297). Urban societydepended on state redistribution. After theintroduction of a state-led development modelin the mid-1950s, the private sector was largelycon®ned to trade, and the governmentcontrolled most urban employment.

Changes in the role of women, includingvoluntary unveiling, and womenÕs secular edu-cation and professional employment, wereentirely urban phenomena dependent on thestate sector. They were decreed by the highest(male) leadership of the state in order toimplement a (lightly) imposed vision ofmodernization. The subsequent collapse andloss of legitimacy of the weakly modernizingstate also meant the weakening of the institu-tional support for womenÕs public roles(Dupree, 1998).

(b) Soviet occupation and politically dependentwar economy

During the Soviet occupation (1979±89) anumber of new phenomena emerged:

Ðdependence of competing leaders onopposing ¯ows of politically motivated mili-tary assistance;Ðdependence of the population for subsis-tence on politically motivated humanitarianaid;

WAR AND PEACE IN AFGHANISTAN 1791

Ðdestruction of the rural subsistence econ-omy through counterinsurgency;Ðforced urbanization, including internaldisplacement to Afghan cities and the ¯ightof millions of mainly rural refugees to campsand cities in Pakistan and Iran;Ðcreation of refugee-warrior communitiesin Pakistan and Iran and of a regionwide Af-ghan diaspora; andÐrapid monetization of the economy.

On one side, the stateÕs dependence onforeign aid and sales of natural gas becameeven more pronounced, but aid came exclu-sively from the Soviet bloc. While the state lostaccess to much of the countryside, more of theswollen urban population came to depend on it.By the Soviet withdrawal, nearly all of KabulÕsfood and fuel was donated by the USSR anddistributed by the government throughcoupons. As men under government controlwere enrolled in the war e�ort, womenÕs civilianroles expanded.

A di�erent culture of dependency developedon the other side. Food production fell by halfto two-thirds as Soviet counterinsurgencydevastated the rural economy (SCA, 1988, p.37). This destruction not only impoverished therural population but weakened the elites whosepower depended on control of rural resources.Much of the rural population ¯ed to Pakistanand Iran, where it entered monetary economies.Islamic parties recognized by PakistanÕs Direc-torate of Inter-Services Intelligence (ISI) asrecipients of US- and Saudi-supplied militaryassistance acted as gatekeepers for distributionof international aid.

In these communities, as well as in rural areasof Afghanistan, patriarchal strictures onwomen were retained or reinforced. Theserestrictions resulted from male reaction to boththe insecurities of life in exile and reformsassociated with the disaster that had overtakenthe country. 1

The infrastructure of support for the resis-tance poured cash into several social networks.Before reaching its intended bene®ciaries, bothmilitary and humanitarian aid passed throughmany international, Pakistani, and Afghanintermediaries, some of whom skimmed o�cash and resold arms and commodities. Theseresources provided capital to expand smugglingand other businesses. While the Pakistanimilitary delivered arms to mujahidin parties inits own trucks, private teamsters moved thesupplies to the border region and intoAfghanistan. Many of these trucks were

already active in Pakistani±Afghan smugglingderived from violations of the Afghan TransitTrade Agreement (ATTA). Under this agree-ment, listed goods can be imported duty-free insealed containers into Pakistan for onwardshipment to land-locked Afghanistan. Many ifnot most of the goods were instead sold insmugglersÕ markets (bara bazaars) in Pakistan.During the war the trucks used in this lucrativetrade were also leased for arms transport,income from which expanded the capitalavailable for investment in smuggling linked tothe ATTA, as well as the growing drug trade.

(c) Soviet withdrawal: monetization and thegrowth of predation

Within Afghanistan itself, the main economicactors were the commanders, a group mostlydrawn from new elites that bene®ted from US,Pakistani, and Saudi policies of supportingonly Islamist parties rather than the nation-alist former rulers (Roy, 1986; Rubin, 1995a).When the Soviet withdrawal reduced bothmilitary pressure and external aid, however,commanders pursued economic strategies toincrease their power, wealth, and autonomy,establishing bazaars and providing local secu-rity to traders in return for tribute. They alsosought aid from Western or Islamic humani-tarian organizations engaged in cross-borderassistance from Pakistan. Such aid providedservices and employment that increasedresources under their control (Dorronsoro,1996).

The Soviet withdrawal also led to a rapidincrease in the Afghan money supply. TheSoviet-supported government of Najibullahturned to an expensive policy of ``nationalreconciliation,'' which included increasing theeconomic dependence of the population on thestate, enlarging the local security forces, ando�ering political recognition and subsidies todefecting commanders. Just as these policiesincreased expenditures, the government lostincome. Soviet aid declined, natural gas reve-nues fell with poor maintenance and endedwhen the Soviet troops left, along with thetechnicians who ran the gas ®elds. Thegovernment ®nanced the resulting de®cit byprinting money. From 1987 to the fall ofNajibullah in 1992 the value of banknotes incirculation increased by an average of 45% peryear. Food prices rose by factors of ®ve or 10.The afghani rapidly lost value against thedollar, trading at 1,000 to the dollar, or about


20 times the o�cial rate, by the summer of 1991(Rubin, 1995a, pp. 153±164). The destructionof agriculture and trading networks created afood de®cit. For several months out of the yearAfghans had to rely for food on aid or cashpurchases at the new high prices (UNDCP,1998a, p. ii).

Such a situation created tremendous incen-tives for cash-producing activities, mainlysmuggling of consumer goods and opiumgrowing. As early as 1987 roads became moresecure, and trade and humanitarian assistancethat had previously traveled by pack animalover mountain trails could now go by truck(SCA, 1988). Trade increased, including thedrug trade, the import of goods into Afghani-stan, and the transit trade, consisting of theexport (smuggling) into the neighboring coun-tries of goods imported into Afghanistan fromboth Dubai (by air or via Iran) and the FarEast (via the trans-Siberian railway and linksup to the northern Afghan border). In north-east Afghanistan, the gem trade also ®nancedthe war, as diamonds do in Angola, Congo, orSierra Leone: Ahmad Shah Massoud, theethnic Tajik commander in northeast Afghan-istan who built up the most extensive resistanceorganization inside the country, controlledemerald and lapis lazuli mines in or near hisnative valley, Panjsher (Chipaux, 1999).

Opium, however, became the main expand-ing source of cash incomes. Opium can begrown in most parts of Afghanistan, and insome regions the yields are the highest in theworld. Opium provides cash not only throughsale, but through credit and demand for labor.Farmers sell the crop to wholesale traders.When faced with cash ¯ow problems or foodde®cits, especially in the winter months beforethe harvest, they can obtain loans from tradersunder a system of futures contracts calledsalaam. Finally, the opium harvest requiresintensive labor, which provides many landlessor land-poor young men with earning oppor-tunities (UNDCP, 1998b, c, 1999a, b, c, d).

The fall of the Najibullah government inApril 1992 brought mainly non-Pashtunmujahidin groups to power in Kabul. Led byPresident Burhanuddin Rabbani but domi-nated by Massoud as military commander, theIslamic State of Afghanistan failed to establishits power over most of the countryÕs territory.Attacks on Kabul by groups supported byPakistan kept the capital insecure. Thoseattacks, combined with the unwillingness ofPashtuns to accept a Tajik-led government,

kept it from expanding control beyond thecapital and its ethnic base in the northeast.Regional warlords developed ®efdoms.

Kabul remained dependent on internationalaid and printing money. The UN and interna-tional humanitarian agencies stepped in toreplace the USSR as the supplier of food to themost vulnerable, but they delivered only abouthalf the amount supplied by the Soviets(120,000 tons versus 250,000 tons of wheat peryear). Printing currency remained probably thesingle most important source of state expendi-ture. Banknotes printed ®rst under contract inRussia and then by other internationalcompanies continued to be delivered to the``government.'' The resulting devaluation of theafghani and in¯ation were so severe that thegovernment introduced new currency notes of®rst 5,000 and then 10,000 afghanis (previouslythe largest note was 1,000). The formercommunist ethnic Uzbek warlord of northernAfghanistan, Abdul Rashid Dostum, had hisown notes printed after breaking with Rabbaniin January 1994. By September 1996, whenKabul fell to the Taliban, the afghani, theo�cial rate of which had been 50 to the USdollar under Najibullah, was trading at 17,800to the dollar in Kabul. The afghani was wortheven less (25,600/dollar) in DostumÕs capital,Mazar-i Sharif, indicating the lack of a nationalmarket. 2

Though road transport was no longerthreatened by Soviet bombing, predation alongroadways disrupted trade and the nationalmarket that had begun to develop with theconstruction of a highway network in the1960s. Each region became increasingly inte-grated with its neighboring foreign market. Thewar economy, like the political structure,remained fragmented.

(d) Taliban: toward predatory monopoly/duopoly

In 1994 a number of changes in Afghanistanand the international environment combined tosupport the growth of the Taliban. Within fouryears, as the Taliban gained control over thecountryÕs main roads, cities, airports, andcustoms posts, they implemented a transitionfrom localized predatory warlordism to weakrentier state power based on a criminalizedopen economy. The opposition, which hadformerly included Uzbek, Hazara, and Tajikleaders, shrank to a mainly Tajik core, led by


Ahmed Shah Massoud in his northeasternmountain bastion.

The breakup of the USSR raised theeconomic stakes in Afghanistan and pitted Iranand Pakistan against each other in competitionfor access to the oil and gas-rich Central Asianstates. Pakistan saw commercial and politicalconnections to Central Asia via Afghanistan askey to the development of ``strategic depth'' inits confrontation with India. The US de®ned aninterest in the independence and economicdiversi®cation of the Central Asian states,without relaxing sanctions on Iran, aneconomically feasible pipeline route. Pipelinesthrough Afghanistan would reconcile thoseoften contradictory goals. Various companies,including the US-based UNOCAL in alliancewith the Saudi company Delta (whose consor-tium received US government encouragement)and their rival, the Argentine ®rm, Bridas,began negotiations with the Rabbani govern-ment and de facto powerholders. Traderschafed at the growing insecurity along themajor routes crossing the country.

In their ®rst major operation, in October1994, Taliban freed a Pakistani trade convoy.Led by a Pakistani intelligence o�cer whohad played a leading role in supporting themujahidin, this convoy was headed for Turk-menistan via Qandahar and Herat, along theprojected pipeline route. When it encountereda checkpoint set up by tribal (Achakzai)militia demanding exorbitant tolls, waves ofnewly armed Taliban ¯ooded across theborder to break the blockade. The convoyrolled on to Turkmenistan as the Talibanmarched into Qandahar (Rubin, 1995b;Maley, 1998; Rashid, 1998). This event wasemblematic of what the Taliban themselvessee as their main accomplishmentÐ``bringingsecurity''Ðand provided both a model and aneconomic basis for their nearly nationwideconsolidation.

Overcoming predation poses a collectiveaction problem: each predatory actor bene®ts,while a larger but di�use constituency wouldbene®t from suppressing predation (Collier,1999). Both social capital that strengthensnetworks of solidarity and investments or side-payments from groups bene®ting from thesuppression of predation can help overcome theobstacles to collective action. The Taliban bothmobilized social capital created in madrasas tocreate a homogeneous leadership group linkedto political networks in Pakistan and usedassistance from Pakistan and Saudi govern-

ments and traders to build up a military forceand buy o� opponents.

The years of war had destroyed much ofAfghanistanÕs social capital as communitiesand institutions were dispersed or destroyed.The prevalence of predatory economic activitiesre¯ected the fragmentation of social power(Rubin, 1995a). During those same years,however, Afghan rural ulama, especially in thetraditionalist south, continued to teach or studyin either their rural madrasas, away from thecenters of war, or in kindred (and much larger)institutions in Pakistan, largely linked to theconservative Deobandi movement. 3 Thou-sands of Pashtun refugee boys received the onlyeducation available in these schools, funded byPakistani or Saudi donors.

The links among these madrasa students andteachers provided an e�ective form of socialcapital. In response to the crisis of anarchy insouthern Afghanistan, a group of teachers andstudents (taliban) from such madrasas formed amovement to overcome warlordism andcorruption. They enjoyed the support of theirPakistani colleagues and could recruit troopsfrom madrasa students in both Afghanistanand Pakistan. They were linked across tribes bytheir common madrasa background, as well asby subethnic solidarity among QandahariPashtuns.

Taliban success, however, also requiredhuman capital (training), technology (weap-ons), and ®nance. For the strategic reasonsdescribed above, Pakistan decided in 1994 toprovide such assistance to the Taliban, includ-ing military training and advisors. SaudiArabia also provided funds until June 1998,when it ended aid to the Taliban in a disputeover Bin Laden. Hoping that the Talibanwould provide security for pipeline routes,stabilize the country, and further isolate Iran,the United States originally did not object tothese policies.

Afghan and Pakistani traders, too, werewilling to pay for suppression of predation.These crossborder traders form a coherent,organized group with an interest in endingextortion, and they have therefore contributedto the Taliban (Rashid, 1998). The Taliban alsoassess them when the need arises.

Taliban o�cials tell visitors, ``You can drivefrom one end of the country to the other evenat night with a car full of gold, and no one willdisturb you.'' This expression is hardly ametaphor. Driving across southern Afghani-stan from Qandahar to Farah and back in June


1998, I saw many trucks doing just that, thoughtheir cargoes consisted of consumer goodsrather than gold. In two daysÕ drive I encoun-tered only three, unobtrusive checkpoints. Thegreater security provided by Taliban alsoimproved the conditions for the trade in opium(UNDCP, 1998c).

The expansion of Taliban power led to theelimination of all but the strongest opposition.A bandwagon e�ect combined with the shift inPakistani support enabled the Taliban quicklyto eliminate rivals in the Pashtun areas ofsouthern Afghanistan. They were then opposedby a coalition of non-Pashtun groups, includingUzbeks, Tajiks, and the Shia Hazaras,supported by Iran, Russia, and Central Asianstates. By 1998, however, except for occasionalpockets of resistance, only Massoud remained.MassoudÕs areas included a small thoughexpanding opium-growing area, but he appearsto have made more money through printingmoney and the international marketing of gems(Chipaux, 1999).

Collier, Hoe�er, and S�oderbom (1999) arguethat ethnic homogeneity enables insurgentgroups to coordinate and prolong civil war. Inthis case subethnic homogeneity of the leader-ship of both armed contenders contributed totheir endurance: the Taliban leadership iscomposed exclusively of Qandahari Pashtuns,while Massoud relies almost entirely on Panj-sheri Tajiks. The use of ethnic and subethnicsolidarity to coordinate military and politicalaction has increased ethnic polarization ofAfghan society.

3. THE CONTEMPORARY WARECONOMY

The projected oil and gas pipelines have beenstymied by the continuing war and the Tali-banÕs harboring of Bin Ladin. TodayÕs wareconomy in Afghanistan consists of the transittrade, the drug trade, the gem trade, serviceindustries stimulated by the growth of theformer three, and the emission of currency.Foreign exchange earned by exports ®nancesAfghanistanÕs imports of arms as well as foodand other necessities (Naqvi, 1999). The Tali-ban control the transit trade, which seems to bethe largest of these sectors. Massoud controlsthe gem trade. Opium production and trade isexpanding in regions controlled by both sides,but in 1999 areas controlled by the Taliban

produced 97% of AfghanistanÕs poppy(UNDCP, 1999a, d).

Control by the Taliban of most of the mainroad system has cleared a corridor for thesmuggling of duty-free consumer goods fromDubai to Pakistan. Until a ban on international¯ights from Taliban territory imposed underUN Security Council sanctions on November14, 1999, some goods were ¯own directly toAfghanistan from Dubai (the airlineÕs onlyinternational destination). Most goods crossthe Persian Gulf by ship to Iran, from wheretruckers haul them through Afghanistan toPakistan. 4 This trade complements smugglinginto Pakistan under cover of the ATTA. InJune 1998 I noted that many of the trucksappeared to be carrying automotive vehicletires and spare parts rather than the electronicappliances I had heard so much about. I laterlearned that since automotive parts hadrecently been eliminated from the list of goodseligible for import under the ATTA, they werebeing imported to Pakistan by this alternateroute (Naqvi, 1999).

A World Bank study (Naqvi, 1999) estimatedthe value of this trade at $2.5 billion in 1997,the ®rst year after the Taliban capture ofKabul, equivalent to nearly half of Afghani-stanÕs estimated GDP and around 12±13% ofPakistanÕs total trade. After the Talibancapture of most of the north in August 1998,the amount increased, according to diplomatsin Central Asia. Naqvi (1999) estimates that theTaliban derived at least $75 million in 1997from taxing Afghanistan±Pakistan transittrade. While this is a signi®cant income in thecontext of Afghanistan, it is far less than theamount of Pakistani duties that would be owedon these goods, so that the more indirectcontraband route is still pro®table.

Before the appearance of the Taliban,Afghanistan was already a major opiumproducer (Table 2). About 56% of the poppycrop was grown in the areas of southernAfghanistan that the Taliban captured in thefall of 1994, and 39% was grown in easternAfghanistan, which they took two years later.These remain the principal opium growingareas, though poppy has also spread to newregions (UNDCP, 1999d).

Two main factors a�ect the amount of opiumgrown in Afghanistan. First, the total hastended to rise through di�usion of a pro®tabletechnology and the decrease in risk a�orded byTaliban ``security'' (UNDCP, 1998b, 1999d).Second, production ¯uctuates around the trend


as a result of weather and price changes,creating a ``cobweb'' cycle typical of unregu-lated agricultural commodities. Rains and¯oods damaged much of the crop in 1998,leading to high prices and indebtedness thatencouraged increased production in 1999.Combined with good weather conditions, thisresulted in a record crop estimated at 4,581tons, or about 75% of global production. Thisin turn led to a drop in price and higher laborcharges, which may lead to cutbacks in plantingfor 2000 (UNDCP, 1999a).

Afghans, including the Taliban, earn rela-tively little from this crop. Superpro®ts in theglobal drug market derive from the riskpremium of marketing an illegal commodityin wealthy societies. Producers and marketersof the raw material share in these pro®ts onlyif they develop vertical integration through tothe retail markets, as the Colombian cocainecartels did in the 1980s (Castells, 1998, pp.166±205). Afghan opium traders, however,generally sell only to the border. A few areinvolved as far as the Persian Gulf, but notin the lucrative retail markets (UNDCP,1998c).

Within Afghanistan, while opium growingand trading involves economic risk, neither theTaliban nor their opponents treat these ascriminal activities, and there is consequentlyneither a high risk premium nor violentcompetition for markets. The opium trade inAfghanistan is by and large peaceful andcompetitive (UNDCP, 1998c). In easternAfghanistan the market is more centralizedwith higher markups between the farmgate andbazaar prices than in the south.

The UN Drug Control Program estimatesthat the 1999 bumper crop was worth $183million at the farmgate at harvest prices. 5 In

1998, when prices were higher for a smallercrop, the markup from the farmgate to theborder was in the vicinity of 50% (UNDCP,1999a,1998c). 6 Processing, of course can bemore pro®table. It appears that opium fromeastern Afghanistan is processed into heroin inborder laboratories controlled by the Pakistan-based Afridi tribe, while there is less processing,and often only into morphine base, in the south(UNDCP, 1998c, p. 4). There are also reportsof heroin being ¯own out of the country onprivate aircraft, by Arabs to the Persian Gulf,or by the Russian ma®a to Central Asia, but noinformation is available about the economicvalue of such transactions.

It is di�cult to estimate how much revenuethe Taliban derive from this trade. Growers paythe Islamic tithe (ushr) at the farmgate onopium and other produce, mostly in kind. Lessconsistent reports indicate that the Taliban alsolevy zakat of 20% on traders in opium andopium derivatives. Some evidence indicates thatthis zakat is collected only in the south, not inthe east, where Taliban control is less stable(Torabi, 1996±97, pp. 141±142; UNDCP,1998c, p. 13; Rashid, 1998). 7 A very roughestimate would be that ushr (which seems to beused for local expenses, not the war e�ort)might amount to up to $15 million in 1999,while zakat of 20% on the remaining opium inthe south (60% of the total) marked up by 50%(the border price) would total about $30million. These rough calculations indicate thatTaliban raise less revenue from opium tradethan the transit trade. How much they mightderive from taxing trade in morphine base andheroin remains an open question, though mostprocessing seems to occur outside Afghanistan,and the Taliban have destroyed some heroinlaboratories.

Table 2. Estimated opium poppy cultivation in Afghanistan, 1994±99a

Year Hectares-cultivated

Opiumharvested

(metric tons)

Average yield(kg/ha)

Districts withreported pop-py cultivation

Hectares cultivated inTaliban areas

(percentage of total)b

1994 71,470 3,416 48 55 (56)c

1995 53,759 2,335 43 60 651996 56,824 2,248 40 63 651997 58,416 2,804 48 60 931998 63,674 2,692 42 73 941999 90,983 4,581 50 104 97

a Sources: UNDCP (1998a; 1999a).b Taliban control at time of planting.c There was no Taliban control at time of planting in 1994. This ®gure is the percentage planted in provincescontrolled by the Taliban in 1995.


The transit and drug trades are comple-mented by service industries, such as fuelstations, shops, and tea houses. Much of thefuel is smuggled from Iran, where its subsidizedprice is approximately $0.03 per liter, less thana soft drink. The o�cial budget in Kabul(which does not include military expenses)seems to be paid for by direct foreign aid fromPakistan (Rs. 500 million or about $10 millionin 1998), and a few taxes from Kabul itself.Until late 1998 the Taliban also received direct®nancial assistance from Saudi Arabia, whichprovided subsidized fuel, as well as cash grants.These were ended in protest over the TalibanÕsfailure to expel or curb Usama Bin Ladin. BinLadin himself is reputed to have put some ofhis wealth at the TalibanÕs service, paying,reportedly, for the capture of Kabul inSeptember 1996.

Though the Taliban control all major bran-ches of Da Afghanistan Bank (the centralbank), they have not printed their own money.The Taliban continue to recognize the notesdelivered to the Massoud-Rabbani forces,despite their protest against this funding oftheir enemies and the resulting devaluation oftheir currency. Taliban banking o�cials saythey recognize the Rabbani currency becausethey do not wish to undermine national unityby circulating two currencies. 8 In practice,the Taliban would probably have di�cultyobtaining professionally printed notes. 9

Northeast Afghanistan, controlled byMassoud, produces only 3% of AfghanistanÕsopium today. Commanders levy ushr on opiumfarmers, and at least some local authorities taxopium traders as well. 10 There are a number ofheroin re®neries, though authorities havedestroyed some. Besides the aid he receives,mainly from Iran, and the continued delivery ofnew Afghan currency, MassoudÕs main incomecomes from the gem trade. Since the beginningof the war Massoud taxed trade in lapis lazuliand emeralds, collecting ushr from mine ownersand zakat from traders. In 1997, however,Massoud established a monopoly in purchaseof the gems and in 1999 signed an agreementwith a Polish ®rm, Inter Commerce, to marketthem. His aides estimate that, while the tradenow brings in $40±60 million per year, the newjoint venture might make as much as $200million in annual income (Chipaux, 1999).

The Taliban, like their opponents, are thusnot throwbacks to medieval times but actors intodayÕs global economy and society. For the®rst time in history, ulama dominate political

and military life in Afghanistan because ofgeopolitics and resources made available byglobalization. Pakistan, Saudi Arabia, and theUS supplied massive quantities of weaponsonly to religious parties rather than to Afghannationalists who might ally with India orchallenge Pakistan, while turning a blind eye tothe growth of drug trade and other forms ofcriminality. Meanwhile, factions of the eliteseducated in AfghanistanÕs state institutions andin the West and the USSR busied themselveskilling, arresting, and exiling each other withthe help of foreign powers.

As the educated elite was destroyed, privatemadrasas o�ered almost the only educationavailable for Pashtun refugee and rural boysafter 1978, since the West did little to providethem with modern education. The madrasanetworks accumulated social capital whileother institutions were destroyed. The mullaslost, however, the ties to the landlord-domi-nated local economy and society that hadcircumscribed their power. Both the state andthe rural economy that had sustained triballeaders collapsed. The Deobandi ulama becamemore autonomous in exile and in warlord-dominated Afghanistan, resulting in theirbecoming more extremist and deracinated. Inexile they also became linked to internationalnetworks, both political and economic, includ-ing Pakistani political parties and intelligenceagencies and the Arab Islamists who aided thejihad. Foreign aid, commercial agriculture(opium), and long-distance contrabandprovides this newly armed elite with theopportunity to mobilize resources to exercisepower directly, as it never did before.

The domination of the country by thispreviously marginalized group has reversed thepattern of social, political, and economicbifurcation developed under the royal regimeand intensi®ed under the communists. Thecapital city is now ruled by a force from thecountryside, which has reversed the reforms ofpast decades. The Taliban attitude toward thestate and reforms are not the continuation ofsome unchanging ``tradition,'' but the result oftheir own uprooting and trauma of the past 20years, during much of which a central statedominated by a foreign ideology destroyed thecountry in the name of progressive reform. Theannihilation of the state and the developmentand reformist agenda it had pursued underseveral governments has spelled the end fornow of the emancipation of urban womenthrough decrees by modernizing male leaders.


4. TRANSNATIONAL NETWORKS

The Afghan war economy has spread inter-nationally through a variety of social networks.The Taliban themselves are a transnationalmovement led by Afghans but organizationallypresent in both Afghanistan and Pakistanthrough Deobandi madrasas and parties. TheTaliban recruit from both sides of the border.During the 1999 summer o�ensive, as many as25±30% of their troops were estimated to bePakistani volunteers, mostly recruited frommadrasas and political groups, not the mili-tary. These ®ghters are often organized inseparate groups, as are groups of Arabs andother international supporters. Besides UsamaBin Ladin, who still has followers in SaudiArabia, the Taliban host members of radicalIslamist groups from Pakistan, Egypt, Uzbe-kistan, Algeria, and many other countries.Several groups engaged in sectarian violence inPakistan have training camps in Afghanistan,and the Pakistani intelligence (ISI) has usedthese groups in its ``covert'' operations inAfghanistan and Kashmir. As a result, asPakistan confronts a severe economic andpolitical crisis, it faces thousands of madrasastudents who have returned after ®ghting forthe Taliban, eager to establish a similar regimein Pakistan.

The religious and political networks aresupported by the transborder economicnetworks described above that link traders tothe Taliban. These in turn have close relationswith the local administration in Pakistan,where the goods are sold in smugglersÕ markets(Rashid, 1998). Given that countryÕs ®scaldependence on customs duties and sales taxeson luxury goods, the toleration of such a largeblack market contributes signi®cantly to Paki-stanÕs ®nancial crisis (Naqvi, 1999). Soon afterhe took power on October 12, 1999, GeneralPervez Musharraf halted unlicensed export ofwheat to Afghanistan, suspended the operationof the ATTA pending the negotiation of strictercontrols, and announced a crackdown onsmuggling from Afghanistan. It is unclear,however, if he will succeed in implementing theeconomic reforms he has announced, as suchmeasures are not only opposed by Pakistaniconstituencies that bene®t from smuggling butalso con¯ict with an unchanged policy ofsupporting the Taliban and using Talibanterritory as a base for operations in Kashmir.

The economic networks involved extendbeyond Afghanistan and Pakistan. Dubai now

contains the third largest urban population ofPashtuns (Karachi is the ®rst and Peshawarthe second). These networks made Afghani-stan the second largest trading partner of theUnited Arab Emirates (along with SaudiArabia and Pakistan, one of the only threestates that recognize the Taliban regime). Thisre¯ects the purchase of duty-free goods inDubai by Afghan and Pakistani tradersshipping them onward for smuggling intoPakistan.

The drug and arms trade have broughtinternational organized crime into the region.Though over 97% of the opium is grown inTaliban-controlled southern and easternAfghanistan, increasing portions of it aresmuggled northward in cooperation with theRussian ma®a. Russian organized crime groupshave sold arms to Massoud and reportedlypurchased heroin from traders on all sides.Opium products cross from territory controlledby Massoud into Tajikistan, where some aretransported by Russian troops and borderguards to Osh, Kyrgyzstan, for transshipment.There are reports that the Russian ma®a also¯ies heroin out of Kunduz, a Taliban-con-trolled town in northern Afghanistan. Arabswho for years have ¯own private planes tosouthwest Afghanistan for hunting expeditionsare now also reported to be ¯ying out opiumproducts to the Persian Gulf. Iran has deployedtroops and police along the border, wherehundreds have lost their lives in clashes withdrug smugglers. The money involved in thedrug and arms trades is undermining stateinstitutions throughout Central Asia and is alsoa�ecting Russia and the Persian Gulf (Nunn,Rubin & Lubin, 1999).

5. MICRO POLITICAL ECONOMY OFCONFLICT

Socioeconomic con¯icts related to the warhave also developed at the local level. Thecollapse and partial revival of the state, thedestruction of assets, and the mass displace-ment and partial return of the population havecreated a crisis in property relations. The oldregime in Afghanistan had established privateproperty in land and pasture and used theseregulations in favor of Pashtun nomads andsettlers in northern and central Afghanistan.State protection of property in land also madepossible absentee landlordism around majorcities. Some landlords were better able to


a�ord emigration or were enticed into theregimeÕs program of ``national reconciliation,''which promised state protection of propertyand exemption from land reform, whiletenants and the landless stayed to ®ght in thejihadÐand for the land of those who sidedwith the regime.

Nomadic sections of Pashtun tribes such asthe Ahmadzai formerly shifted their ¯ocks intothe high pasture of Hazarajat in summer. Thenomads, whose mobility gave them a naturalvocation as traders, also had more access tocash and acted as bankers for the local peas-antry, who were often heavily indebted. Whenthe ShiÕa Hazaras revolted against thecommunists in 1979, they also, in e�ect,declared autonomy from the Pashtun-domi-nated Sunni Muslim state. Since that timePashtun nomads could not use pastures inHazarajat or collect their debts. Hence anumber of the tribes that formerly migratedthere have supported the Taliban reconquest ofthe area. Armed young men from Pashtunnomadic tribes have returned to Hazarajat withthe intention of collecting 20 year old debts. 11

Similarly, the fertile steppes north of Haza-rajat are inhabited by largely Tajik (thereforeSunni) populations who were also favored bythe state against their Shia neighbors. Thesegroups lost land and other assets to theHazaras over the past 20 years and now formthe base of Taliban support in these areas. AsBamiyan, the center of the Hazara-controlledareas in central Afghanistan, changed hands inearly 1999 from the Taliban back to the Shiaparties and then back again, Hazaras and localTajiks engaged in several rounds of burninghomes, killing, and expelling populations. Thiswas due as much to local con¯icts as to thenational one.

Pashtun nomads also lost pasture on thenorthern steppes, and Pashtun landlords lostcontrol of their agricultural lands in theseareas. Landlords around Herat and Qandaharlost control to their tenants. Wealthier land-lords were better able to a�ord the journeyabroad and therefore were more likely tobecome refugees. In such cases local ulamaoften allotted use of their land to the families ofmujahidin or martyrs. As people return fromexile, they demand their land back, sometimestaking disputes to Taliban courts. All of thesetransformations have created a rich ®eld forproperty and monetary disputes, sometimesconnected to ethnic, tribal, or clan con¯icts, aswell as class.

6. POLITICAL ECONOMY OFPEACEMAKING

The war economy makes peacemaking muchharder. Standard international plans for endingcivil wars involve negotiating a cease-®re withinternational monitors; establishing interimpower-sharing leading to elections; integratingrebels into government security forces;rebuilding the economy and society underinternational auspices; and institutingaccountability for abuses of human rights. Butbelligerents negotiate when they are exhaustedor reach a stalemate, while continued foreignassistance and the open war economy ofAfghanistan assure both sides of resources tocontinue ®ghting. Some may in fact prefercontinued war that allows them to pro®t(Collier, 1999).

Pakistan and Iran both have clear interests ina stable Afghanistan but competing interests inhow power should be distributed and exercised.Negotiations between them confront the factthat PakistanÕs interest in Afghanistan derivesfrom its quest for strategic depth against India,not from items than can be negotiated withTehran.

The re-establishment of a weak state in mostof the country by one party would in othercircumstances favor peace, by establishing astructure into which others could be integrated.Generalized predation plus accountability candevelop into taxation and state building. But inthis case the two sides do not even agree onwhich is the government and which the insur-gents. The ideology that is a key component ofthe TalibanÕs social capital has thus farprecluded both genuine power sharing inter-nally and international recognition externally.Their bans on female education and employ-ment as well as their harboring of Usama BinLadin have made them the target of condem-nation and sanctions. Even their genuineaccomplishments seem at risk. Accounts ofcorruption and indiscipline in the ranks grow.Shopkeepers charge that in January 2000Taliban guards burglarized KabulÕs moneymarket, making o� with over $200,000 (AFP,2000).

Furthermore, ending war in Afghanistanmight transform the criminalized war economyinto an even faster-expanding criminalizedpeace economy. Whoever rules Afghanistan,the incentives for misgovernment are nearlyirresistible. Only the drug, transit, and gemtrades are worth taxing. The rest of the


economy is hardly productive enough torecover the cost of governing it. Such a polit-ical economy would leave the power holders asunaccountable to most Afghan people as theywere under previous regimes. Most of thepopulation would be left to fend for them-selves, perhaps in conditions of greater secu-rity, but without a development agenda, publicservices, or reforms, notably in the status ofwomen. 12

A more challenging alternative would be toconsider peacemaking in Afghanistan as part ofa larger problem, of transforming the politicaleconomy of a region. It has ®nally dawned onEurope and the United States that nothing lesswill work in the Balkans. There is no reason tothink that Central and Southwest Asia will be asimpler problem. Of course, even in Europe therhetoric has outstripped the ®nancing. Butintegrating economic and considerations intothe strategic picture reveals a few, weak pointsof leverage.

The military coup of October 12, 1999, wasonly a symptom of the deep crisis of Pakistanwhose policies toward Afghanistan arguablythreaten that state itself. Pakistan cannotresolve its economic and political crisis withoutreforms that would also promote peace inAfghanistan, and the countryÕs heavy interna-tional indebtedness provides leverage. Paki-stanÕs policy of aiding the the Taliban and usingAfghan territory to train groups for jihad inKashmir empowers armed groups guilty ofsectarian terrorism and the hijacking of theIndian airliner in December 1999. It alsounderwrites smuggling that bankrupts thePakistani state. If General Musharraf wants topromote public order, he will have to suppressarmed groups and move toward law-bound,civic politics. To stabilize the economy, hisregime will have to reform the ATTA not inisolation but as part of a general change awayfrom the high duties of an import substitutionregime toward the legal institutionalization ofthe greater openness that exists de facto outsidethe law. This would reduce the incentive fortransit trade, the TalibanÕs main source offunds. Pakistani businesses could bene®t fromlegitimate trade with Afghanistan and evenmore from its reconstruction under interna-tional auspices. Pakistan should not be o�ereda ®nancial rescue package if it is not willing toreform its own policies that create and intensifythe crisis.

No major institution has started planning forreconstruction of Afghanistan or involved

Afghans in thinking about it. 13 Yet a majoruntapped resource today is the hidden exhaus-tion of a vast majority of Afghans with the warand a historically unprecedented demandÐarising from the people, not the stateÐforeducation and development. Starting a seriousinternational process involving Afghan intel-lectuals and community organizations in plan-ning for a reconstruction that would beconditional on a cessation of hostilities andobservance of minimal humanitarian andhuman rights principles (the right of both sexesto available education and health care) mighta�ect the current dynamic of con¯ict. Localactors assume that major aid for reconstructionwill not be forthcoming, however they behave.It is not surprising that they ®nd it relativelyeasy to dismiss international professions ofconcern about Afghanistan and their ownbehavior.

Aid for reconstruction of Afghanistan shouldbe decided upon and disbursed in such a way asto build reciprocity between state and societyand make the former more accountable to thelatter. For instance, programs aimed atreplacing opium poppies would also have to®nd alternative sources for ®nancing stateactivities. The existing economic actors wouldhave to be drawn into alternative forms ofactivities from which they could realizereasonable pro®ts.

The international private sector could also beinvolved as a source of funds. Today positionson the proposed pipeline and other potentialinternational investments in Afghanistan arepolarized around attitudes toward the Taliban,and the United States has imposed unilateralsanctions on such investment. A more creativesolution might ®nd ways to ®nance the invest-ment while reducing the risk that it would fundwar or oppression. Could ®nancing be o�eredon the condition that the rental income go notto any armed group, but to a fund forcommunity development and reconstruction?Airlines crossing Afghan territory on long-dis-tance ¯ights pay fees into a Swiss accountmanaged by the International Air TransportAuthority (IATA), which holds them in trustfor spending on civil aviation requirements(AFP, 1999). Perhaps such a fund from pipelineincomes could be administered by an interna-tional organization with Afghan participation.

It be virtually impossible to sustain curbs onsmuggling in a region with such wide disparitiesin trade policy. International institutionsshould work with the regional powers toward


something approximating a customs union thatwould both make legitimate trade moreattractive and reduce incentives to smuggling.Greater crossborder cooperation and con®-dence-building measures with Iran, alsoundergoing a struggle over reform, might alsohelp reduce regional tensions and create acommon stake in rebuilding the country.

Opium production presents particularlydi�cult obstacles. UNDCPÕs research hasbegun to outline what would be needed tomove away from opium production: cropsubstitution, the growth of o�-farm incomeopportunities, and the spread of education.These will require massive foreign involvementand investment, and none will be possiblewithout legitimate governance.

Most important is working with Afghans tochange the image and role of the state, seenlargely as a distant and indi�erent if not hostilepower. Local power structures that have largelygrown up as defensive measures of self-rule tokeep the state or powerholders away have to beincorporated into o�cial structures of planningand service provision. Afghanistan needs adecentralized governance structure in whichprovinces and localities receive authority to taxand plan in consultation with local shuras(councils). In the past local societies developed

uno�cial power structures to shield themselvesfrom the state, rather than participate in it, andthe centralizing mentality shared by the Talibanand much of their opposition reproduces thatpast pattern. Instead, modest local resourcesunder local control could be directed intolocally accountable planning processes. Thecentral state will still be needed for provision ofbasic security and dispute resolution, but aclear division of labor among levels of gover-nance will promote greater accountability overthe reconstruction process.

The disintegration of the state paradoxicallyopens such possibilities, though the criminal-ized economy has created interests that willresist it. Peacemaking also has dangers:attempts to exercise economic pressure onPakistan risk precipitating a worse crisis there.Attempts to weaken or replace the Talibancould easily lead to the return of anarchy andpredation and a yet more bloody civil war. Butunless peacemaking can transform powerfuleconomic actors into agents of peace, it will belimited at best to halting ®ghting in one placebefore social and economic forces provoke itonce again elsewhere in this dangerous region.Without such an e�ort, spread of both con¯ictand the regional war economy remain the mostlikely prospect.

NOTES

1. For a literary treatment of this process, see Majrooh

(1984).

2. Data collected by the World Food Program,

supplied by Zareen F. Naqvi of the World Bank. In

May 1999 the afghani was trading at 42,675/dollar in

Kabul and at 94,250/dollar in Mazar.

3. This movement, named after a town in northern

India where conservative ulama established a madrasa in

the 19th century, was founded to combat Islamic

modernism, and, in particular, reforms in Muslim

education.

4. In protest against the murder of Iranian diplomats

and a journalist by Taliban troops in Mazar-i Sharif,

Iran closed the border between August 1998 and

November 1999. During that period the goods took a

detour via Turkmenistan.

5. Farmers who sold some of their crops through

futures contracts (salaam) received less, while those able

to hold the opium could bene®t from higher prices

during the winter.

6. In southern Afghanistan in 1998, the average

farmgate price was about $60 per kilogram, with various

qualities selling for as low as $44 and as high as $82,

twice the price in eastern Afghanistan. The southern

border price was $73±95, with the highest quality dried

opium selling for $126 per kilogram. In 1999 fresh

opium prices ranged from $27 to $72 per kilogram

throughout the country (UNDCP 1998c, 1999a).

7. According to sharia, zakat is a tax on wealth levied

at 2.5%, or one-fortieth. It is unclear on what legal basis

the Taliban impose this tax at a much higher rate and on

a ¯ow of commerce rather than a stock of wealth. It is

unclear if the zakat is assessed on gross income or on

pro®t.

8. Interviews with director of Bank-i Milli Afghani-

stan, Qandahar, and deputy director of Da Afghanistan

Bank, Kabul, June 1998. Both of these o�cials were


graduates of Pakistani madrasas with no economic,

commercial, or ®nancial background or experience.

9. The international legal regime for currency printing

is complex and decentralized. In controversial cases the

few companies that do ``security printing'' (of currency,

passports and other o�cial documents) look to their

host governments (usually their major customers) for

guidance. These governments generally use political

criteria in giving opinions about such contracts. The

major security printers are in the United States, United

Kingdom, Germany, and France, none of which looks

favorably on the Taliban. I thank R. Scott Horton for

clarifying these points for me.

10. Anthony Davis and Bernard Frahi, personal

communications.

11. Bernt Glatzer and Ashraf Ghani, personal commu-

nications.

12. For an analysis of this phenomenon in Africa, see

Reno (1998).

13. The following ideas draw, to an extent that I can

no longer identify, on seven years of discussion with

Ashraf Ghani, though I alone am responsible for the

speci®c proposals here.

REFERENCES

Agence France-Press (AFP), (1999). Kabul's war-bat-tered airport beats the millennium bug. Kabul, June29.

Agence France-Press (AFP), (2000). Kabul's moneymarket burgled. Kabul, January 13.

Chipaux, F. (1999). Des mines d'�emeraude pour ®nancerla r�esistance du commandant Massoud. Le Monde,Paris, July 17.

Castells, M. (1998). End of millennium. The informationage: economy, society, and culture, vol. 3. Oxford:Blackwell.

Collier, P. (1999). Doing well out of war. Washington,DC: World Bank (www.worldbank.org/research/con¯ict/papers.htm).

Collier, P., & Hoe�er, A. (1999). Justice-seeking andloot-seeking in civil wars. Washington, DC: WorldBank (www.worldbank.org/research/con¯ict/pa-pers.htm).

Collier, P., Hoe�er, A., & S�oderbom, M. (1999). On theduration of civil war. Washington, DC: World Bank.(www.worldbank.org/research/con¯ict/papers.htm).

Dorronsoro, G. (1996). Afghanistan: des r�eseaux desolidarit�e aux espaces r�egionaux. In F. Jean, J-C.,Ru®n, �Economie des guerres civiles (pp. 147±188).Paris: Hachette.

Dupree, N. (1998). Afghan women under the Taliban. InW. Maley, Fundamentalism reborn? Afghanistan andthe Taliban (pp. 145±166). New York: St. Martins.

Fry, M. (1974). The Afghan economy: money, ®nance,and the critical constraints to economic development.Leiden: E. J. Brill.

Holsti, K. (1995). War and the state and the state of war.Cambridge: Cambridge University Press.

Jean, F., & Ru®n, J. (1996). �Economie des guerres civiles.Paris: Hachette.

Kaldor, M. (1999). New and old wars: organized violencein a global era. Cambridge: Polity Press.

Keen, D. (1998). The economic functions of violence incivil wars. Adelphi Paper 320, London: OxfordUniversity Press for the International Institute forStrategic Studies.

Majrooh, S. (1984) End of a sojourn in the abode ofrefugees: Gul andam [body like a ¯ower], or the story

of laughing lovers. Ashraf Ghani (trans.). In Izhda-yiKhudi (The Ego Monster), Book 1, vol. 5. Peshawar:Unpublished manuscript.

Maley, W. (1998). Fundamentalism reborn? Afghanistanand the Taliban. New York: St. Martins.

Naqvi, Z. F. (1999). Afghanistan±Pakistan trade rela-tions. Islamabad: World Bank.

Nunn, S., Lubin, N., & Rubin, B. R. (1999). Calming theFerghana valley: development and dialogue in theheart of Central Asia. preventive action reports, vol. 4.New York: The Council on Foreign Relations andThe Century Foundation.

Rashid, A. (1998). Pakistan and the Taliban. In W.Maley, Fundamentalism reborn? Afghanistan and theTaliban (pp. 72±89). New York: St. Martins.

Reno, W. (1998). Warlord Politics and African States.Boulder, Co.: Lynne Rienner.

Roy, O. (1986). Islam and Resistance in Afghanistan.Cambridge: Cambridge University Press.

Rubin, B.R. (1995a). The fragmentation of Afghanistan:State formation and collapse in the internationalsystem. New Haven, CT.: Yale University Press.

Rubin, B.R. (1995b). The search for peace in Afghani-stan: From bu�er state to failed state. New Haven,CT.: Yale University Press.

Swedish Committee for Afghanistan (SCA). (1988). TheAgricultural survey of Afghanistan: ®rst report.Peshawar: Swedish Committee for Afghanistan.

Torabi, B. (1996-1997). Entretien avec Mollah Moham-mad Omar. Politique internationale, 74, 135±143.

United Nations Development Programme (UNDP)(1997). Afghanistan PEACE Initiative (1997±1999).Islamabad: UNDP.

United Nations International Drug Control Programme,Afghanistan Programme (UNDCP). (1998a). Annualopium poppy survey 1998. Islamabad: UNDCP.

United Nations International Drug Control Programme,Afghanistan Programme (UNDCP). (1998b). Strate-gic study #1: analysis of the process of expansion ofopium poppy cultivation to new districts in Afghanistan,preliminary report. Islamabad: UNDCP.

United Nations International Drug Control Pro-gramme, Afghanistan Programme (UNDCP).


(1998c). Strategic study #2: The dynamics of thefarmgate opium trade and the coping strategies ofopium traders, ®nal report. Islamabad: UNDCP.

United Nations International Drug Control Pro-gramme, Afghanistan Programme (UNDCP).(1999a). Annual opium poppy survey 1999. Islamabad:UNDCP.

United Nations International Drug Control Pro-gramme, Afghanistan Programme (UNDCP).(1999b). Strategic study #3: the role of opium as asource of informal credit, preliminary report. Islama-bad: UNDCP.

United Nations International Drug Control Pro-gramme, Afghanistan Programme (UNDCP).(1999c). Strategic study #4: access to labour: the roleof opium in the livelihood strategies of itinerantharvesters working in helmand province, Afghanistan,Final Report. Islamabad: UNDCP.

United Nations International Drug Control Pro-gramme, Afghanistan Programme (UNDCP).(1999d). Strategic study #5: an analysis of the processof expansion of opium poppy to new districts inAfghanistan, second report. Islamabad: UNDCP.

United Nations O�ce of the Coordinator of Humani-tarian A�airs for Afghanistan (UNOCHA). (1996).Afghanistan consolidated appeal for assistance 1997.Islamabad: UNDCP.

United Nations O�ce of the Coordinator of Humani-tarian A�airs for Afghanistan (UNOCHA). (1997).1998 Consolidated appeal for Afghanistan. Islama-bad: UNDCP.

United Nations Development Program (UNDP) (1996).Human Development Report. New York: OxfordUniversity Press.

UNICEF (1996). State of the world's children report.New York: UNICEF.


the political economy of war and peace in afghanistan

Documents