The World Bank 90186 DECEMBER PREMnotes 2006 NUMBER 111 ECONOMIC POLICY & DEBT Drugs, Security, and Development 1 William A. Byrd, Adviser, SASPR, World Bank South Asia Region Introduction drugs, success has been elusive. On the pro- Illicit narcotics are a global issue. duction side, a few countries have been Consumption is found in virtually all coun- able to eliminate cultivation of illicit drug tries, although the kinds of drugs involved crops on a sustainable basis, but because of and the incidence and level of drug con- increases in production in other countries, sumption vary greatly across countries, this has not translated into sustained regions, and groups of people. Drug traf- declines in global output. On the consump- ficking also is widespread, although the tion side, it has not been possible to sharply number of countries where trafficking is reduce drug consumption in mature con- large-scale and constitutes a major threat to suming markets, and there have been siz- national governance and development is able increases in consumption in new mar- much smaller. Production of agriculture- kets, especially in some transitional and based illicit narcotics (primarily opium and developing countries. Where declines in coca) is highly concentrated in a small consumption have occurred, they appear to handful of developing countries.2 Overall, be related more to changing fads, demo- the size of the global illicit drug industry graphic factors, competition from synthet- and its direct and indirect costs are very ics, and so on, rather than attributable to large and worthy of serious attention from specific counternarcotics interventions. both research and policy perspectives. Finally, the adverse side effects of drug con- Although seen by many decision makers trol interventions have become increasingly and practitioners in the drug control com- evident, as will be discussed later in this munity as primarily a criminal or a health paper. problem, illicit narcotics constitute a devel- These considerations highlight the need opment issue of the first order. The impact to “think out of the box” about illicit drugs, of the drug industry on security and gover- and to treat this squarely as an important nance in major producing and trafficking development issue. The 2006 PREM countries, as well as its more direct adverse Conference session “Drugs, Security, and effects on development, are of great, even Development,” on which this paper is overriding importance in some developing based, was intended to bring broader countries such as Afghanistan, Colombia, development perspectives to bear on the and Myanmar. The adverse development drug issue. The session was also intended effects of illicit drug consumption are more to encourage the rethinking of approaches diffuse but add up to something very large in light of treating illicit narcotics as a in global terms (including the spread of development and security issue, rather HIV/AIDS), and are very significant in than merely as a narrower problem of crim- countries with major drug consumption inality and health. Last but not least, the problems, such as Iran. Finally, the costs of session raised questions about what the combating drugs divert scarce resources World Bank should do or can do in this from other priority uses, constituting area—one it has traditionally avoided. another adverse impact from a broader The presentations and discussion at the development perspective. session focused on the following three Despite the large and costly efforts that questions: have gone into the fight against illicit FROM THE POVERTY REDUCTION AND ECONOMIC MANAGEMENT NETWORK 1. What do we know about the implica- ward some conclusions, including possible tions of drug production and traffick- roles for the World Bank in this field. ing in poor countries involved in pro- The Broader Context of Illicit Activities duction / trafficking / consumption of Worldwide illicit drugs? (Based on Presentation by Moises Naim) 2. What do we know about the develop- Although they have existed on an organized ment consequences, particularly the basis for a very long time, illicit drug produc- indirect consequences operating tion and trade constitute only one compo- through security and violence, of nent (albeit a very important one) of the bur- developed country responses to drug geoning global illicit economy. Trafficking of production and trafficking? people, counterfeit products, timber and 3. Is it possible for developed countries to other natural resources, and art are among pursue their main policy goal (reduced the major (and in some cases rapidly grow- drug consumption) in a way that better ing) illicit activities worldwide. Although takes into account the development these activities have been present to some impacts of efforts to achieve that goal? extent all along, a number of factors have Focusing broadly on these questions but contributed to the boom in illicit trade dur- also expanding beyond them as appropri- ing the 1990s and in the initial years of the ate, this paper provides a thematic sum- 21st century. These include, among other mary of the four presentations and subse- developments, improvements in commu- quent discussion at the session, and draws nications technology, declining transporta- some conclusions and possible implica- tion costs, trade liberalization by many tions for strategy and policies. Rather than countries, conflicts and associated poor mechanically following the order of the governance and weak rule of law, and sev- presentations and summarizing the dis- eral specific developments, such as the cussion, however, the approach is topical transition in the former Soviet Union coun- and issue-oriented, with the sequence tries and China’s economic boom. The intended to lay out the facts in a cohesive trends and events that have spurred global- manner. ization in general have also facilitated and stimulated the growth of illicit activities. After this introduction, the paper starts by placing illicit narcotics within the broad- Illicit activities and trade have to be con- er phenomenon of illicit activities, based on sidered in assessing the size of a country’s Moises Naim’s presentation (also drawing economy and in setting macroeconomic on his recent book, Illicit3). Then, based on policies. In a considerable number of coun- Peter Reuter’s presentation, it reviews glob- tries, illicit economic activities are at the al patterns and trends for illicit opium and heart of economic and political life. These opiates and coca and cocaine. The evidence activities can be profoundly destabilizing to available on the impact of various drug governments as well as costly to societies, intervention strategies is summarized, and despite their economic benefits. policy implications and dilemmas put for- For many countries (especially poorer ward. Based on Rodrigo Soares’ presenta- developing countries), illicit activities con- tion, the next section of the paper outlines a stitute the most effective way to integrate cost-benefit approach to evaluating drug themselves into world markets. For these control interventions. Also, based on this countries and their people, such activities presentation, the paper looks at the unin- represent globalization at work. Notable tended consequences of the illegality of illic- examples in the drugs sphere include it narcotics, and some possible implications opium in Afghanistan and coca in Bolivia, for thinking and strategy. Drug control and which bring farmers and traders into a close development policies in Afghanistan, a pre- relationship with global drug markets. mier example of a poor country affected by The sources and enabling factors behind massive drug production and trafficking, countries getting into illicit activities and are the subject of the next section, based on trade vary greatly, with considerable con- David Mansfield’s presentation. The paper text-specificity. Common themes include closes by drawing out key points from the the following: discussion at the session and putting for- 2 PREMNOTE DECEMBER 2006 • Traditional activities (historical basis): However, farmers, refiners, and local traf- for example, a traditional agricultural fickers receive less than 2 percent of this— base for production of illicit narcotics which, nevertheless, is locally significant in appears to be an important factor in some of the drug-producing developing some cases. countries.4 International traffickers receive • Locational advantages (geography 5–10 percent—that is, up to $15 billion. matters): for example, Haiti is deeply Thus the bulk of gross revenues from drugs integrated with Colombian drug traf- accrues to distributors in the rich consum- ficking, Costa del Nol in Spain is a cen- ing countries. ter for a variety of trafficking activities, Production and trafficking of illicit nar- and Surinam is a connection for drug cotics have broad adverse consequences for transactions between the Andean coun- development. In the case of drug produc- tries and developed countries’ markets. tion, these include conflict between farmers • Manufacturing base: for example, and the government, corruption, erosion of China for counterfeit goods, Ukraine the state and weakening of governance, for weapons, Romania for electronics and the international stigma attached to and hacking. being a major drug-producing country. Drug trafficking can generate considerable • Financial system: countries and locali- organized violence, in extreme cases consti- ties with a well-functioning, weakly tuting a challenge to the government. In regulated financial system have an contrast to production (which tends to gen- advantage since they can engage in erate small-scale corruption), trafficking money laundering, a service that virtu- usually involves high-level, systemic cor- ally all illicit activities require. ruption. The large share of drug revenues • Weak rule of law can be an important accruing to a few principals can aggravate enabling factor for illicit activities. inequality and associated tensions. The evi- • Path dependency: there seems to be a dence as to whether drug production and considerable role for historical factors, trafficking create an increased risk of “drug experience, and enabling logistics in epidemics” in the countries concerned is determining the location of illicit mixed, with Mexico, Turkey, and, at least activities. up to this point, Afghanistan, being coun- terexamples. In some countries “criminal governance” has emerged. These countries become An interesting feature of the situation is structurally involved in illicit activities, that illicit drug trafficking and especially with the government becoming in effect a production of agriculture-based narcotics is criminal enterprise to carry out and protect highly concentrated globally. Afghanistan such activities. Countries can even become and Myanmar have accounted for more “specialists” in certain illicit activities, not than 90 percent of global illicit opiate pro- least in money laundering—several very duction since 1990; Bolivia, Colombia, and small countries are in effect operating pri- Peru are the only producers of cocaine marily as money laundering facilities. (although a number of countries have culti- vated coca and opium in the past). More In sum, illicit narcotics form part of a rap- countries are involved in trafficking; never- idly growing and increasingly globalized theless, opiates are shipped primarily illicit economy. Strategic thinking, as well as through Iran, Pakistan, Tajikistan, Turkey, law enforcement and regulation, lag far China, Thailand, and Mexico, and cocaine behind dynamic illicit trading, calling for trafficking is concentrated in Mexico and greater attention and policy priority as well the Caribbean. Drug abuse is much more as rethinking of past neglect, and in some widespread, but also concentrated in cer- cases (for example, narcotics) counterpro- tain regions. Most opiate abusers are in ductive policies. Asia, and cocaine abuse is primarily in rich Illicit Drugs: Global Patterns and Trends nations. Table 1 summarizes these patterns. (Based on Presentation by Peter Reuter) Global consumption of illicit narcotics is Global retail sales of cocaine and heroin likely to continue to grow, although proba- generate $100-150 billion per year. bly more slowly in the case of heroin and DECEMBER 2006 PREMNOTE 3 three categories: demand-side interven- Table 1: Developing Countries Strongly tions (prevention, treatment), supply-side Affected by Illegal Drugs interventions against production and traf- Cocaine Heroin ficking, and harm reduction programs. Production Bolivia Afghanistan While demand-side interventions have Colombia Colombia played a significant role in addressing drug Peru Laos consumption in some rich countries, the Myanmar main focus of counternarcotics efforts as Trafficking Mexico Iran they relate to developing countries has Dominican Pakistan been on supply-side interventions. Republic However, evaluation of drug control strate- Panama? Tajikistan gies and instruments has been uneven and generally weak. In particular, there have Consumption Iran been relatively extensive evaluations of Thailand drug treatment programs in some coun- India Tajikistan tries, but few and weak evaluations of other Russia interventions. Source: Peter Reuter’s presentation at PREM Evaluations in several countries have Conference Session on Drugs, Security, and shown that methadone maintenance Development (April 2006). reduces opiate use, HIV risks, and crime. Other heroin substitutes appear promising. cocaine than for synthetic drugs. Demand Cocaine treatment programs are less strong for cocaine and heroin tends to be “sticky” but also have substantial short-term effects. downwards. Drug abuse spreads through However, it must be recognized that even “epidemics,” leaving aging cohorts of in good programs, most patients drop out dependent users. No country has quickly or relapse, and there have been few such reversed a drug use epidemic. More specif- evaluations in developing countries. ically, heroin consumption in rich countries Evaluations of prevention programs have is stable, a pattern likely to continue, but is been much weaker, and predominantly growing in some regions, notably Russia, conducted in the United States (primarily Central Asia, and China. Cocaine consump- of school-based programs for children tion is decreasing in the United States but 12–14). Some experimental programs have may be increasing in Western Europe. Thus, had sizable effects on marijuana use, but developing countries will continue to face the longer-term effects have not been stud- high levels of demand for drug production ied, and it is not clear whether prevention and trafficking. programs affect people most likely to Policy Options and Dilemmas become problematic cocaine or heroin (Based on Presentation by Peter Reuter) users—evaluations of implemented pro- As Table 2 shows, existing drug control grams show little or no effect. Programs strategies can be broadly classified into may need to be culturally specific to achieve results. Table 2: Classification of Existing Drug Control Strategies Demand-Side Interventions • Prevention • Treatment Supply-Side Interventions • Production controls — Eradication — Alternative development — General economic development • Trafficking, smuggling, and distribution controls — Interdiction — Domestic enforcement Harm Reduction Programs • Needle exchange — Heroin maintenance? Source: Peter Reuter’s presentation at PREM Conference Session on Drugs, Security, and Development (April 2006). PREMNOTE DECEMBER 2006 Little is known about the effectiveness of once they are established, this leads to the supply-side controls. Some interventions, question of whether the world would be like eradication and general economic better off if one nation is the dominant pro- growth, have not been evaluated for their ducer for each drug. Although obviously impact on drug production. Alternative extremely harmful for the country con- development programs have been studied, cerned, this approach might minimize the but only at the local level, and without global costs in terms of corruption, and assessment of spillover effects.5 Contrary to limit the adverse effects on governance and arguments from some quarters, supply-side the state from drug production to one or controls in source countries are unlikely to two countries. However, even if such an have a substantial effect on foreign con- approach were to be considered, is there an sumption. This is because consumption can acceptable mechanism for selecting the be affected only through price, but as world producers? shown in Table 3, only a small share of the final A Cost-Benefit Approach to Drug retail price is accrued in source countries. Control Efforts There are also very few evaluations of (Based on Presentation by Rodrigo Soares) supply-side controls directed against traf- A cost-benefit approach is essential (and ficking outside the source countries. Even surprisingly, not much used hitherto) to in the United States, few papers have esti- assess whether drug control policies are mated the impact of tougher enforcement working. On the benefit side, the assump- on drug availability and use. There are no tion is that reduced drug consumption pro- very positive findings on domestic enforce- tects the individuals concerned from them- ment in consuming countries. For example, selves (based on a paternalistic approach, the methodologically strongest published or time-inconsistent preferences on the part study estimated that tripling incarcerations of addicts). Reduced drug consumption between 1986 and 1996 resulted only in a 10 also protects society from dangerous or percent higher price of cocaine. The find- incapacitated addicts (reflecting externali- ings with respect to international interdic- ties associated with drug use). On the cost tion are mixed, with some studies reaching side, the costs of interventions include pessimistic conclusions and others some- direct expenditures on police, judicial sys- what more optimistic.6 tem, prisons, and so forth, as well as the All in all, it is hard to escape the conclu- loss of human capital of those incarcerated, sion that drug-affected countries, especially loss of lives due to violence associated with developing countries, have relatively nar- crackdowns on drugs, and other related row policy options and a very weak evi- costs.7 dence base for action. It must be empha- In simple terms, the effectiveness of an sized, moreover, that policies need to be intervention depends on (i) the extent (size) analyzed in a context wider than that of a of the intervention; (ii) the elasticity of sup- single country. For example, a crackdown ply; and (iii) the elasticity of demand. The against drugs in one country could well stylized facts on these three aspects are shift production elsewhere, as has hap- summarized below. pened in the Andean countries. Combined The size and associated cost of drug con- with the fact that the effects of drug produc- trol measures have been very substantial in tion and trafficking are difficult to reverse Table 3: Structure of Cocaine and Heroin Prices Stage Cocaine (per pure kg equivalent) Heroin (per pure kg equivalent) Farmgate $650 (Colombian leaf) $500 (Afghan opium) Export $1,000 (Colombia) $3,000 (Afghanistan) Import $20,000 (Miami) n/a Wholesale (kg) $30,000 (Chicago) $50,000 (London) Wholesale (ounce) $50,000 (Chicago) n/a Retail (100 mg pure) $100,000 (Chicago) $175,000 (London) Source: Based on Boyum and Reuter, An Analytic Assessment of U.S. Drug Policy (Washington, DC: American Enterprise Institute, 2005). DECEMBER 2006 PREMNOTE 5 some countries. In the United States, rough- military expenditure in 2004 was equiva- ly $35 billion annually was spent on fight- lent to 4.3 percent of GDP, compared with ing drugs at the end of the 1990s, up from 4.9 percent for education and 6 percent for $10 billion per year in the mid-1980s. One health. Military personnel composed 1.5 in four prisoners is in jail for drug-related percent of the labor force by 2004. Although offenses, mostly nonviolent.8 As shown in not solely for fighting against drugs, these Figure 1, drug-related incarcerations have resources and personnel were to a large increased steeply over time. extent devoted to this purpose. It does not appear, however, that these Interventions did have a substantial impact major law enforcement efforts have had on crime, with significant reductions in much impact on opium prices. As shown in homicides, kidnappings, and acts of terror- Figure 2, prices for cocaine, heroin, and ism between 2002 and 2004 (although marijuana have all been trending down- whether these reductions will be sustained ward in real terms. Nor do there appear to is questionable). But whether there has have been major changes in the incidence of been any impact in reducing overall coca drug abuse. Prevalence of heroin and production, especially on a global basis, is cocaine abuse among high-school seniors not clear (see Figure 3). appears to have remained relatively con- Turning to the elasticity of supply, the stant since the early 1990s, and marijuana effect of interventions on supply seems to use has increased following a low point in be relatively limited. To the extent that 1992 (although it still remains below peak prices may be inversely related to supply, levels of the late 1970s). And the rates at the downward trends seen in Figure 2 do which these three drugs are mentioned in not suggest any significant impact of inter- hospital emergency room cases (per 100,000 ventions on supply. This should not be sur- population) have increased continuously prising, for several reasons. The quantities over the past two decades and longer. transacted are relatively small (for example, Turning to Colombia, a prominent exam- the estimated annual import of cocaine into ple of a drug-affected developing country, the United States is 400 tons, and total glob- Figure 1: The Number of Adults Incarcerated for Drug Law Violations in the US Has Grown Sharply Over Time 500,000 400,000 Jails State Prisons 300,000 Federal Prisons 200,000 100,000 0 1972 1977 1982 1987 1992 1997 2002 6 PREMNOTE DECEMBER 2006 Figure 2: Real Cocaine, Heroin, and Marijuana Prices, 1975–2003 2 1.8 1.6 real prices (1975 =1) 1.4 1.2 1 0.8 0.6 0.4 0.2 0 75 77 79 81 83 85 87 89 91 93 95 97 99 01 03 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 real cocaine price real heroin price real marijuana price al production of heroin has been on the easily stimulate a shift of production to order of 500 metric tons per year). The cost other regions. of inputs and transportation constitutes The price elasticity of demand also does only a small fraction of the final retail price not appear to be very high. Taking an exam- (and the latter already builds in a large mar- ple from another addictive product, it has gin reflecting risks associated with law been estimated that the prohibition against enforcement). Finally, there are many alter- alcohol reduced consumption by no more native regions for cultivation of illicit drug than 10–20 percent (and perhaps not at all crops, and reductions in supply from one over the longer term). Although estimates area (and associated price increases) can for illicit narcotics suggest that the elastici- DECEMBER 2006 PREMNOTE 7 ty may be somewhat higher, it is still rela- (iii) security, safety, and institutions. tively low in comparison to the costs of In the primary producing countries, intervention. For example, recent work sug- drugs can account for a substantial propor- gests that the price of cocaine is only 2–5 tion of total economic activity. Similarly, the times what it would be in a legalized market. value of drugs trafficked through some of This cursory summary of drug control the main transit countries is equivalent to a efforts from the perspective of cost-benefit significant share of GDP and total exports. It analysis clearly calls for consideration of should also be noted that income from illic- alternatives. The question arises as to it narcotics is highly skewed—as between whether, from this perspective, a form of the transit and especially source countries legalization with regulation would result in and high-income consuming countries (see something any worse than the present situa- earlier discussion), and within the source tion. This could be the case, if (i) governments countries as between farmers, wage labor- were not able to effectively regulate the legal ers, and small-scale traders on the one market, or (ii) addiction increased uncontrol- hand, and the more substantial traders in lably and there were no means of treating it. the distribution chain on the other. With respect to the first issue, given In the area of public health, the adverse the relatively low elasticity of demand, effects of illegality include those of unregu- legalization with regulation may be lated, variable, and often substandard qual- more effective than prohibition, even ity, notably the risk of overdose because of taking into account the possibility of a uncertain purity, and poisoning because of black market. Moreover, governments in the adulteration of drugs with other harm- many countries (especially high-income ful substances. Contagious diseases, includ- countries) do regulate markets for sever- ing HIV/AIDs, are spread by sharing of al different drugs, most of them quite dirty syringes among addicts. Another pos- effectively.9 sible adverse effect is environmental dam- With respect to the second issue, it is dif- age due to aerial spraying. ficult to make a firm prediction. However, But most serious for countries heavily the fact that the number of Americans with affected by illicit drugs are the adverse substantial drug problems is limited and implications for institutions. Under condi- has remained roughly constant (in the tions of large rents and the necessity for dis- range of two million people) suggests that tribution networks with private enforce- the incidence of harmful drug addiction ment of contracts in the illicit drug econo- might not increase very substantially fol- my, organized groups using violence tend lowing legalization. Moreover, effective to emerge. In a weak institutional setting, treatments do exist for some drugs, notably this may lead to perverse effects of intensi- methadone for heroin addicts. Even in the fied law enforcement and policing—higher case of crack cocaine—probably the most crime rates, more corruption, and, per- damaging drug from personal and social versely, a larger role for organized crime. perspectives—the indirect social costs due Emergence of strong organized crime has to repression are estimated to be higher several negative implications. There are than the direct social costs resulting from typically turf wars and escalation of gang addiction. violence. In Colombia, 27,000 people died The Unintended Consequences of per year as a result of the violence, which is Illegality estimated to have had a welfare cost corre- (Based on Presentation by Rodrigo Soares) sponding to a present value of 250 percent In addition to the considerations outlined of GDP. Increased illegal drug trafficking in the previous section, there are other was also associated with the advent of adverse consequences of the illegality of organized crime and increased violence in illicit narcotics, which may be particularly Brazil (see Table 4). The social costs of vio- important and damaging for developing lence typically are concentrated among the countries. These can be broadly divided prime-age population. into three categories: impacts on (i) Other adverse effects on a country’s insti- resource endowment, (ii) public health, and tutions include the symbiotic relationship 8 PREMNOTE DECEMBER 2006 between illicit drugs and guerrilla move- Drug Control and Development Policies ments and terrorism in some countries, in Afghanistan FARC (Fuerzas Armadas Revolucionarias (Based on Presentation by David Mansfield) de Colombia—Revolutionary Armed The case of Afghanistan brings out in stark Forces of Colombia) in Colombia and the form many of the issues discussed earlier. Shining Path in Peru being notable exam- Afghanistan is by far the world’s largest ples. Such relationships can start out from producer of illicit opium; its share in global either side but rapidly build synergies. output has risen sharply over the last cou- Corruption associated with illicit drugs ple of decades and is now in the range of 90 ranges from petty corruption (for example, percent. Opium has become the country’s police, customs) to grand corruption, leading economic activity, accounting for involving political campaigns and the judi- on the order of one-third of total (opium- cial system. Displacement of vulnerable inclusive) GDP. Not surprisingly the opium populations is another adverse side effect— economy, and the ongoing counternarcotics for example, the estimated 2.5 million efforts to fight it, are having profound desplazados in Colombia in the past effects on Afghanistan’s economy, society, decade. More generally, a culture of vio- and governance, as well as politics. lence and lawlessness becomes entrenched. However, there is great diversity within All of these factors can generate political the country in the incidence and importance instability, whether through murders of top of opium, and even at current levels of pro- officials (Colombia), effective power wield- duction, only a small proportion of the total ed by drug interests in certain slums cultivated land area is being devoted to the (Brazil), or popular protests resulting in crop. Afghan rural households have complex toppling of a government (Bolivia). livelihood strategies, depending on the To summarize the arguments in this and assets, opportunities, and constraints that the previous section, the effectiveness of the they face, with opium an element of varying existing drug control efforts in actually importance for different households. reducing consumption and production of Extensive rural fieldwork in recent years illicit drugs is, at the very least, question- suggests that the degree of dependency of able. On the other hand, the costs of these rural households on opium is inversely relat- interventions are obvious and very large. ed to households’ access to assets of various Hence there is a need to think critically and kinds, including those of the household objectively about this issue, and to consider itself (size of household, number of able- whether there are viable and realistic alter- bodied adults, landholding, livestock assets, natives to present approaches. and so forth), the resources of the locality (irrigation water, climate, whether double- cropping is possible), access to labor and Table 4: Evolution of Violence in Brazil commodity markets, and local governance and rule of law. These relationships are sum- 1980 2002 marized graphically in Figure 4, which is Homicide as Fraction intended to capture the spectrum of depend- of Total Mortality 2% 5% ency lying between the two extremes of access to assets and dependence on opium. Homicide Rate (per 100,000) 12 28 For the relatively resource-rich (in the Afghanistan context), opium provides an Homici as Fraction opportunity to earn higher returns on their of Total Mortality land, profit from holding inventories of Males 15–29 18% 44% opium, and mobilize outside labor for Homicide Rate opium through favorable (to the landown- (per 100,000) er) share-cropping arrangements. Given Males 15–29 36 105 their assets and location, for these types of households law enforcement measures Source: Brazilian Ministry of Health, Sistema de against opium poppy cultivation are appro- Informacoes sobre Moralidade. priate and are likely to be effective and sus- tainable, as they can move into licit activi- DECEMBER 2006 PREMNOTE 9 ties and earn sufficient (even if lower) ritories for one year) and on a more spo- returns to cover their livelihood needs. radic, localized basis in subsequent years— On the other hand, for resource-poor most notably in the southeastern province households the opium economy provides of Nangarhar in the 2004–05 growing sea- in many cases indispensable access to son (when a 96 percent reduction in cultiva- essential resources to help meet their mini- tion was achieved), can have variegated mum basic needs. These include, most and often counterproductive effects. A notably, access to land and credit, as well as notable example of the latter is the wage labor opportunities in opium harvest- increased burden of opium-related debt on ing. Dependency of poorer households on many rural households, as a result of the opium is exacerbated by the opium-related sharp price increase for opium in response debts that many of them have accumulated, to the Taliban ban, and the inability of which require continuing opium poppy households not cultivating opium poppy to cultivation to be in a position to service the service their debts. Cases of poor house- debt. For these households, choices become holds providing underage girls in marriage much more limited and stark in the face of to opium traffickers to settle their opium- eradication or other enforcement efforts related debts have been widely reported. against drugs. Many of them simply have Generally, the impact on better-off house- no viable alternative for making ends meet holds in localities with better resources and in the absence of the opium economy, and access to markets has been quite different under such conditions may be forced to from the impact on poorer households in take drastic, counterproductive coping more remote areas, lacking access to mar- actions, for example selling remaining kets and poorly served by infrastructure. assets or outmigration. Figure 5, which builds on Figure 4 and In the face of this diversity, the imposi- reflects analysis from recent fieldwork in tion of an effective ban on opium poppy Nangarhar province, divides the spectrum cultivation, as occurred under the Taliban of households in terms of access to assets regime in 2000–01 (which virtually elimi- and dependency on opium into four cate- nated cultivation in Taliban-controlled ter- gories, with declining access to assets and 10 PREMNOTE DECEMBER 2006 increasing dependency on opium from left of these households and areas have already to right. The figure distills the experience of returned to opium poppy cultivation, and these four categories of households during others are likely to do so in the near future. the two years since the opium ban was In summary, extensive fieldwork and imposed in Nangarhar. Households with research in Afghanistan highlight the diver- the most assets and with access to markets sity of rural households and their relation- and other resources (the leftmost category) ships with the opium economy. In this dif- were not only able to weather the opium ferentiated context, imposition of a ban on ban (albeit with significant loss of income), opium poppy cultivation has widely vary- but were able in the following year to fur- ing short-run impacts and difficult prospects ther develop a variety of licit activities and for sustainability. In particular, it is only in improve their income positions. At the the case of localities and households with opposite extreme, for the asset-poor house- viable alternative livelihoods that a ban has holds in remote locations (also more diffi- good prospects of being sustained beyond cult to reach by law enforcement), the the initial year. Efforts by the aid communi- opium ban in Nangarhar never worked in ty to provide alternative livelihoods in the the first place; households continued to cul- short run to households affected by an tivate some opium poppy even in the first opium ban, for example through cash-for- year of the ban, with increases in the fol- work programs, are not sufficient or sus- lowing year. For the middle two groups, the tainable. Moreover, they do not address the opium ban was largely effective in the first benefits beyond short-run income provided year but with clear signs of becoming by the opium economy, notably access to unsustainable in the second year. Assets land and credit for poor, landless house- and licit income opportunities, while better holds. Thus, short-term alternative liveli- than for the most resource-poor group, are hoods programs do not fundamentally nevertheless insufficient for viable and sus- change the patterns depicted in Figure 5. tainable alternative livelihoods. Thus some DECEMBER 2006 PREMNOTE 11 This gives rise to a policy dilemma: entry points, and how could the Bank efforts to eliminate opium poppy cultiva- have an impact? tion will not work unless households have The first question led to generally nega- access to viable licit alternative livelihoods, tive conclusions from the panel: the global but except where these already exist as a policy environment—in particular the result of household assets and local strong commitment to prohibition and a resources and opportunities, it is not possi- law enforcement approach on the part of ble to generate alternative livelihoods key high-income countries—makes any quickly or sustainably through develop- prospects for bold rethinking of policies ment assistance. Moreover, opium bans and consideration of a different approach to imposed on poorer areas and households the legality of illicit narcotics very unlikely. tend to be counterproductive, actually It was also stated that there is very little increasing the potential dependency on popular support in the United States for opium by forcing households to engage in legalization of drugs. This is the case harmful asset sales and other coping mech- despite the positive experience with the anisms in the face of the ban, and leaving ending of the prohibition against alcohol a them with even fewer alternatives to opium number of decades ago, and sharp reduc- poppy cultivation in the future. tions in smoking in the United States Thus, there is no alternative to a longer- achieved through regulation of tobacco. term, well-sequenced approach that targets Developing countries’ policy options are eradication and other efforts to eliminate severely constrained by the drug control opium poppy cultivation in the short run at policies of high-income countries. better-off localities, which already are in a Nevertheless, it was felt that the issue of position to shift to licit livelihoods to meet legalization should continue to be studied their basic needs. Development of viable and raised in policy forums, from technical alternative livelihoods for resource-poor (for example, cost-benefit) as well as insti- households in remote localities with limited tutional and governance perspectives. resources will take a long time (and is like- With respect to the role of the World ly to involve outmigration from the most Bank, there was broad agreement that the remote, resource-poor localities). As Bank cannot ignore illicit narcotics, both as demonstrated by experience with the a global issue and as an important develop- Taliban ban and more recently in ment issue for heavily drug-affected coun- Nangarhar province, forcing such house- tries. Leaving drugs out of the picture risks holds to stop cultivating opium poppy, weakening the quality and relevance of the before development progresses to the point Bank’s analytical work and policy advice in where viable licit livelihoods become a rea- contexts where drugs are a significant prob- sonable prospect, can exacerbate poverty lem. At the global level, taking a broader and increase dependency on opium over development perspective in analytical and the medium term. policy work would be useful. At the country Discussion and Conclusions level, the importance of narcotics issues The presentations summarized in this varies greatly across countries, but clearly paper provoked a lively discussion at the this area should not be ruled out (in advance session, which focused to a large extent on of examining country contexts) from being two key questions: included in country analytical work and pol- • Even if the presentations point toward icy dialogue as appropriate. Drugs are also legalization with regulation as an alter- very important in some regional contexts native to current drug control policies (for example, the Andean region and near- that may be worth considering is this at by transit countries, and the wider Central all a realistic possibility, and what has Asia region around Afghanistan); thus, this been the experience with legalization of issue needs to be brought into the Bank’s other once-prohibited goods? regional work as necessary. • What can and should the Bank do in Where important (for example, in the this field (which it has traditionally case of major drug producing and transit avoided) —what are its comparative countries, and in work on HIV/AIDS), advantages (if any), what are good drugs cannot be ignored in the Bank’s 12 PREMNOTE DECEMBER 2006 financial operations, even though direct Although the presentations highlight the financial support to drug control programs complexity and difficulty of the drug issue, is not a prospect. For example, given the this does not mean it is impossible for the pervasive nature of the opium economy in Bank to have an impact. In the case of Afghanistan, efforts are underway by the Afghanistan, for example, Bank engage- Bank and some other donors to “main- ment through analytical work and dialogue stream” the counternarcotics dimension in has been productive. However, expectations development programs.10 need to be kept modest, given the enormity In terms of comparative advantage, and intractable nature of the problem. although the Bank has no expertise on Finally, we return to the three main ques- drugs per se, it does have a global develop- tions that were intended to guide the pre- ment reach and perspective as well as sentations and discussion at the session. demonstrated skills and resources to take Based on the presentations and the discus- on complex cross-cutting issues, and to do sion as summarized in this paper, the fol- innovative thinking on such issues from a lowing conclusions emerge: broader development perspective. There is ample evidence of the adverse Moreover, Bank technical expertise on spe- development and governance implications cific sectors closely related to drugs—for of drug production and trafficking in the example, agriculture and health—can be producing, trafficking, and poor consum- brought to bear. The Bank’s expertise on ing countries. In some countries the econo- governance and corruption also is relevant my becomes distorted and uncompetitive to work on drugs. As in other areas, a way as a result of drugs, and the drug industry forward would be through partnerships can be a serious threat to the state as well as with other agencies working in the drug a major source of corruption. field to exploit synergies and differing com- parative advantages. There is also evidence that drug control policies, often vigorously advocated and Turning to entry points, at the country strongly pursued by developed countries, level they would vary, depending on the do not achieve much effect (particularly in context. In countries where drug produc- relation to their costs), and moreover have tion is a major problem, the broader macro- adverse development consequences for economic and development impact is an developing countries, including through obvious area for Bank engagement and con- violence and displacement and, in extreme cern.11 In these and also in major trafficking cases, political destabilization. countries, the impact on governance and corruption may be a good entry point as It would appear that there are possibili- well. In some cases sector work (for exam- ties to pursue drug control objectives in ple, in agriculture or health) may be a good ways that better take into account develop- way to engage initially. Patient dialogue ment impacts. Even if legalization with reg- with governments and proactive response ulation is not a realistic option, evidence to demand, as well as taking advantage of from fieldwork suggests that more targeted opportunities when they arise, are likely to and better sequenced approaches involving be essential. law enforcement and development pro- grams could work better. In any case, the At the global level, one entry point may lack of success of the drug control policies be HIV/AIDS and other serious diseases pursued hitherto points to the need to look spread through illicit drug use. The Bank is for new and improved approaches. engaged in global programs in these areas, Endnotes and it would make sense to incorporate 1 drug issues in this work. The Bank fre- This paper summarizes and builds on the four quently gets invited to international forums presentations at the World Bank PREM Conference session on “Drugs, Security, and on drugs, and its presence at such meetings Development” on April 26, 2006, by Moises could serve as an entry point. Continuing Naim, Peter Reuter, Rodrigo Soares, and David research on drugs and their nexus with Mansfield . It also draws on the discussion at the security and other issues from a global session following the presentations, and to some development perspective may create fur- extent from other recent work on illicit narcotics ther entry points. issues. The session was organized by William DECEMBER 2006 PREMNOTE 13 Byrd and Philip Keefer. The views, findings, least in the very short run, the impact of lower interpretations, and conclusions expressed in drug consumption on consumer surplus is not the paper are those of the author and (where included on the cost side. However, this issue is indicated) the presenters, and should not be relevant on the production side, and the loss of attributed to the World Bank, its affiliated organ- income and producer surplus if drug control izations, its Executive Board of Directors, or the policies are successful in reducing consump- governments they represent. tion—which necessarily implies a reduction in 2 Production of synthetic drugs, such as meta- production over time, if not immediately— should be taken into account. amphetamines, which are based on chemical rather than agricultural raw materials, is much 8 This puts the U.S. imprisonment rates for drug- more dispersed. related offenses above the overall imprisonment 3 Moises Naim, Illicit: How Smugglers, rate in most western European countries. Traffickers, and Copycats are Hijacking the 9 This raises the question as to whether regula- Global Economy (New York: Doubleday, 2005). tion of drugs would work in low-income coun- 4 For example, in Afghanistan opium was tries with weak states and poor governance, possibly leading to a different choice between roughly estimated to account for 31 percent of prohibition and legalization with regulation. total (opium-inclusive) GDP in 2005. However, law enforcement efforts also are less 5 It makes little sense to evaluate an alternative likely to be effective in weak states suffering development project positively if it merely from poor governance, with a risk of massive results in drug production shifting from the corruption and destabilization of the state. Thus, locality covered by the project to other localities whether from a cost-benefit perspective the in the country. A similar issue can be raised choice would be different for developing coun- about using the country as the unit of evalua- tries is not at all clear. tion. For example, drug control programs have 10 In addition to exploiting possibilities for been deemed successful in eliminating opium development efforts to contribute to the devel- production in Thailand over a period of opment aspects of the national counternarcotics decades, but from a global perspective this pro- strategy, it is also very important to ensure that duction shifted across the border to Myanmar. Bank-supported programs “do no harm” in rela- Similar arguments can be made about successes tion to drugs—for example, that irrigation proj- achieved against opium in Iran and Pakistan, ects do not result in expansion of opium poppy while production in Afghanistan boomed. cultivation in the irrigated area. 6 However, as the following section will discuss, 11 In the case of Afghanistan, for example, the any benefits of such drug control interventions International Monetary Fund has also become must be evaluated against their costs. concerned and has done analytical work on the 7 Although drug users may perceive that they macroeconomic implications of the opium gain some benefits from drug consumption, at economy. This note series is intended to summarize good practices and key policy findings on PREM-related topics. The views expressed in the notes are those of the authors and do not necessarily reflect those of the World Bank. PREMnotes are widely distributed to Bank staff and are also available on the PREM Web site (http://prem). If you are interested in writing a PREMnote, email your idea to Madjiguene Seck. For additional copies of this PREMnote please contact PREM Advisory Service at x87736. This PREMnote was edited and laid out by Grammarians, Inc. Prepared for World Bank Staff