Unexploited potential? What role can international financial institutions play in drugs and development?

This paper assesses the potential for and obstacles to engagement on drugs issues from a development perspective by the World Bank and other International Financial Institutions, based on a review of the World Bank’s experience in Afghanistan as well as more general institutional aspects. Afghanistan demonstrates both the possibilities and benefits as well as the limitations of World Bank/IFI engagement on drugs and development. Prospects for duplicating, let alone exceeding, the level of Bank engagement on drugs seen in Afghanistan would not appear to be good. With pro-active encouragement and support by member countries of the World Bank/IFIs, however, there may be some entry points in major drug-producing countries and relating to global themes such as money-laundering or illicit-economic activities.


Introduction
As the premier development actors in the world, the World Bank and other International Financial Institutions (IFIs)-mainly the regional development banks-bring together sizable human and financial resources along with broad development mandates and their nearly comprehensive country membership including powerful shareholders. They still provide substantial (though relatively declining) financial support and technical assistance for developing countries. The question therefore naturally arises as to whether there are unexploited opportunities for the Bank and other IFIs to engage more on drugs and counter-narcotics issues from a development perspective, and if so how?
This paper approaches this question mainly based on a review of the World Bank's experience in Afghanistan, where the World Bank meaningfully engaged on drugs with both significant accomplishments and some disappointments. The paper first briefly lays out the main interfaces between drugs and development, both more generally and in the case of Afghanistan. It then discusses the main obstacles to the Bank and other IFIs' engagement on drugs and development. The Bank's experience in Afghanistan is then summarized, including country-specific enabling factors, initial progress, and subsequent loss of momentum and reversals. The final section looks at prospects and possible entry points for Bank and IFIs' engagement on drugs and development in the future, before concluding on a realistic note of pessimism.

Drugs and Development Interfaces
This section discusses the multiple impacts of drugs on development (there is also reverse causation from development to drugs-for example, sustained economic growth may provide numerous attractive alternatives to drug crop cultivation, but these tend to be more general and the effects are felt over much longer periods of time). Table 1 presents a stylized summary, dividing the drug industry into five main vertical stages: drug crop cultivation, processing and refining (or production in the case artificial drugs with no crop base), long-distance transport, wholesale and retail trade, and consumption. These stages interact with development in differing ways, some more important than others.
When a country is a major drug crop cultivator, this has a wide range of development impacts, as shown in the table, though it must be remembered that only a tiny handful of countries in the world cultivate large amounts of illicit narcotic drugs. Processing and refining has narrower, but still potentially important impacts on development, which are very much affected by the size of these activities in the country concerned. The long-distance trade-i.e.'transit countries'-primarily impacts development in proportion to the amount of money involved, which in some countries such as Tajikistan may amount to a non-negligible share of total economic activity. There also may be a tendency for transit countries to suffer from higher rates of problem drug use as a spillover from the flow-through of narcotics. The wholesale and retail illegal drugs trade in consuming countries is typically associated with organized crime and with corresponding development-related impacts to varying degrees. And finally, high rates of addiction to illegal drugs, and inappropriate policy responses, are associated with a range of human development issues particularly in the sphere of public health. In the case of Afghanistan, all the interfaces outlined in Table 1 are operative (see also  :  Table 7, p. 37), but the most striking feature is the country's role as the global near-monopoly producer of illegal opiates. This, combined with the processing of most opium into heroin or morphine within Afghanistan, means that the development implications of drugs in Afghanistan have a macroeconomic dimension as well as a profound interface with rural development. The large amount of money involved-well into the billions of dollars-increases the impact of the drug industry on governance, corruption, political economy aspects and conflict. At the demand end, despite the fact that the vast bulk of opiates produced in Afghanistan are exported, rising problem drug use (not confined to opiates), reaching what many term ' drug epidemic' levels, has generated major public health issues as well as other adverse development impacts.
In a nutshell, Afghanistan provides a unique microcosm of drugs and development, with all of the different interfaces prominently exhibited. There has been variation over time, for example the earlier trend of increasing processing of opium within Afghanistan, the progressive rise in problem drug use and the size of drug-related financial flows-not only in absolute terms but also in relation to other inflows, most notably the enormous international military expenditures within the country and flood of aid during the 'surge' period.

Obstacles To Development Agencies' And IFIs' Engagement
Given the multifaceted linkages between drugs and development outlined above, why have IFIs generally shied away from fulsome engagement let alone direct involvement on drugs issues? There have been several main constraints.
The "non-political" mindset of development agencies constitutes an underlying obstacle to meaningful engagement with drugs issues. The World Bank's Articles of Agreement are explicit: The Bank and its officers shall not interfere in the political affairs of any member; nor shall they be influenced in their decisions by the political character of the member or members concerned. Only economic considerations shall be relevant to their decisions, and these considerations shall be weighed impartially in order to achieve the purposes stated in Article I (International Bank for Reconstruction and Development 1944: Article IV, Section 10).
The regional development banks followed suit; the African Development Bank (1963: Article 38), Asian Development Bank (1965: Article 36) and Inter-American Development Bank (1959: Section 5f) have the same political prohibitions written into their charters, with virtually identical language. Even the Asian Infrastructure Investment Bank, recently established under Chinese auspices, contains a similar political prohibition in its Articles of Agreement (Asian Infrastructure Investment Bank 2015: Article 31). Thus, the approaches of other development banks appear to be broadly consistent with that of the World Bank with regard to the political prohibition, and the discussion below focuses on the World Bank. This prohibition has been interpreted fairly expansively over the years, though with clarifications that allowed the World Bank to engage with developing countries on issues of governance and corruption, for example (World Bank 1991 and1992). Nevertheless, there are strict limits on what kinds of activities the World Bank can finance through its loans and grants. Another longstanding limitation is that the World Bank cannot freely comment on member countries' defense budgets, except to the extent of assessing any adverse impact the level of defense spending may have on the adequacy of expenditures on development, though it may support developing countries' efforts to achieve greater budget transparency overall-including in the security budget (World Bank 1992: 46).
In this context, law enforcement, particularly the military and police, have constituted 'red lines,' as they are seen as not only political but also carrying reputational risks due to their mandate to use violent methods, potential for human rights abuses, etc. 1 Therefore, funding of police salaries and other non-lethal police expenditures has been prohibited, sometimes to the detriment of sound public financial management in the concerned developing countries. The Afghanistan Reconstruction Trust Fund (ARTF) is a notable example. With the World Bank as its administrator, the ARTF could not directly fund any police costs (nor defense). Therefore, a separate UNDP-managed trust fund, the Law and Order Trust Fund for Afghanistan (LOTFA), Byrd: Unexploited Potential? What Role Can International Financial Institutions Play in Drugs and Development? 110 was established to pay for police salaries and some other security sector costs. The performance of LOTFA, unfortunately, has been poor. LOTFA faced problems of reporting, waste and accountability, as well as a corruption scandal (see Byrd & Smith 2015). The kinds of financial management, accounting and financial reporting skills required were not a core competency of UNDP (unlike in the case of the World Bank), and nor did UNDP, for a number of years, bring in the requisite skills from outside. The obstacles to engagement on law enforcement issues naturally carry over to the law enforcement aspects of illicit narcotics. Beyond this, the Bank has not until recent years focused on the development side of illicit economic activities more generally, either. Related, more narrowly-focused risk aversion on the part of IFIs also may have been a factor. Illicit narcotic drugs are a complex, often highly politicized topic, understandably leading to a tendency to avoid this issue.
Perceived lack of expertise on drugs was another, self-reinforcing obstacle. Whether accurate or not, there was a perception that other agencies whose mandate explicitly focuses on drugs, most notably the United Nations Office on Drugs and Crime (UNODC), are better equipped to deal with the issue. Moreover, some bilateral development agencies, such as the UK's Department for International Development (DFID) especially during the period when the UK was designated as the 'lead nation' for counter-narcotics in Afghanistan, did deploy development expertise on drugs from a development perspective, especially in the early post-2001 period.
Yet another obstacle, in some cases, may have been that key stakeholders such as the government of a developing country, and/or a major foreign donor or ally of the country and shareholder of the World Bank, did not want the World Bank to engage on the drugs issue in that country. The former effectively blocks the Bank from acting, as it is mandated to work with the national government (represented on the World Bank's Board of Governors, normally by the Finance Minister, and on its Executive Board of Directors). It has also been difficult for the World Bank to engage on drug issues in a country if a large shareholder of the Bank opposed such engagement. It may be that past opposition by the US government, and probably also by governments of regional countries, may at least in part explain the lack of earlier World Bank engagement on drugs issues in Latin America, in particular Colombia.
Only a handful of countries in the world are sizable producers of illicit narcotic drug crops. Others are located along major drug transit routes and are affected by the associated criminality and large financial flows carrying the risk of grand corruption. But in most countries, drugs are not an important part of the rural economy, or of the economy in general, and to the extent those countries are affected, it may be primarily by problem drug use, which varies across countries. This militates against organizations like the World Bank taking a comprehensive approach to drugs and development across the many developing countries they work with, since for the large majority of them the drug trade is not a major issue except perhaps in terms of consumption and public health issues, as well as the criminality associated with trade and distribution.
A final factor that may partly explain the Bank's lack of engagement with the drug industry is what might be called 'safeguards aversion.' When viewed holistically, the drug problem is a multifaceted cross-cutting issue that affects development in a number of ways (see Table 1). At the operational level, the World Bank (or other IFIs) typically might handle such an issue through a 'safeguard' mechanism. Projects would be assessed with regard to their interrelationships with and potential impacts on the issue concerned (positive or negative), with the objective of safeguarding that the project does not adversely affect the cross-cutting issue but on the contrary, is at least consistent with or even promotes progress. Environmental and social assessments, and a number of specific issues including natural habitats, pest management, indigenous peoples, physical cultural resources, involuntary resettlement, safety of dams, projects in disputed areas and projects on international waterways are the subject of safeguards policies applied to World Bank-supported projects (World Bank 2017a).
Safeguards impose a substantial burden on preparation and appraisal of projects, so there would be reluctance to carve out drugs as a new area for safeguards that would need to be assessed project-by-project. Moreover, the benefits of a global safeguard approach to drugs and development in respect of Banksupported projects would be very low compared to the costs in the large majority of countries that are not major drug crop producers or on main transit routes. On the contrary, in most countries it would be more convenient to compartmentalize and treat the drug industry's interface with development as a sectoral issue. For example, in a country whose primary interface is illegal drug consumption (accompanied by wholesale and retail trade), it may make sense for IFIs and other development agencies to view drugs as primarily a public health issue, possibly along with to gang criminality and violence.
Emblematic of the Bank's limited engagement on issues around drugs and development is that a search of publicly available World Bank documents reveals that the vast majority of those which have illegal drugs or narcotics in their title are focused on Afghanistan and were published in the first decade of the 21 st century. 2 This paper now turns to a consideration of the World Bank's engagement on the drugs issue in that country.

Enabling factors
There was a resurgence of poppy cultivation in Afghanistan following the Taliban ban of 2000-2001 and the international intervention in the latter year that overthrew the Taliban regime. As a result, opium became too important to ignore from a development perspective, particularly in terms of macroeconomic and rural development impacts. Aside from this general recognition, extensive dialogue with David Mansfield, who has conducted extensive fieldwork on opium in Afghanistan from a development and livelihoods perspective, as well as receptivity from other Kabul-based development agencies, were important contributing factors leading up to the Bank's initial engagement in 2003. Moreover, the US government never opposed Bank engagement on drugs and development issues in the case of Afghanistan. Some US agencies were positively supportive, while the UK, the lead nation for counter-narcotics in Afghanistan, proactively encouraged such engagement in variety of fora and through direct lobbying.
Encouragement of Bank collaboration with UNODC on counter-narcotics issues by UNODC Executive Director Antonio Maria Costa was another strand in the early post-2001 years. Costa had several meetings with World Bank President James Wolfensohn and with Bank staff in Kabul. Although Costa's initiative ultimately failed to bring about broad-based Bank engagement on counter-narcotics issues in general, it was agreed that an on exceptional basis the Bank could be pro-actively involved in this area in the Afghanistan context, particularly on the analytical side. Though less important than the substantive aspects and partnerships with other agencies in Kabul, this dialogue provided high-level ' cover' from the Bank's top management for Bank engagement on drugs in Afghanistan to move forward.
A final enabling factor was the openness of the Afghan government, or at least important elements of it, to Bank engagement on drugs issues as well as other controversial subjects, such as conducting a security sector public expenditure review and work on vulnerabilities to corruption. The Ministry of Finance, led by then-Finance Minister (and current President) Ashraf Ghani, as well as some other Afghan government agencies and other stakeholders, saw the Bank as a disinterested and objective (if, at least initially, unknowledgeable) party that could bring to bear a development perspective undistorted by political, security or law enforcement mandates.

Early evolution, achievements and disappointments
There were three distinct, yet intertwined, strands of World Bank engagement on drugs in Afghanistan: (1) analytical and policy work and associated publications; (2) collaborating with other donors and convening events; and (3) efforts to factor the drugs issue into World Bank-supported projects. These different strands were characterized by differing inputs, longevity, and degrees of success. The discussion below is largely chronological, in the interest of outlining the evolution of the Bank's role across the three interrelated strands.
Analytical work quickly began near the end of 2003, starting with a familiarization mission for the Bank to learn about the issue, which resulted in an internal report that later spawned two publications ). These publications comprehensively explained the initial understanding gained by the World Bank on drugs in Afghanistan, building on the in-depth research of others. The Bank's first country economic report on Afghanistan since the 1970s incorporated this early analysis of the opium economy in a chapter on the subject (World Bank 2005: Chapter 7).
In 2004 the Bank also convened two events that brought together Afghan government officials, interested donors and outside experts: (1) a small informal brainstorming meeting on drugs in Afghanistan, held in Dubai, United Arab Emirates during 16-17 September, 2004;and (2) a larger workshop on Rural Finance in Afghanistan and the Challenge of the Opium Economy, held in Kabul, Afghanistan during 13-14 December, 2004 (see Zia et al. 2005).
The Bank's early work also involved developing an initial approach to 'mainstreaming' of the illicit narcotics dimension in development operations in Afghanistan (Ward & Byrd 2004: Annex 8). A 'screening' process was put forward for prospective World Bank supported projects that would assess their interfaces with the opium economy, identify associated risks and opportunities, and put forward recommendations for inclusion in the design of projects and in implementation arrangements to exploit potential opportunities and mitigate risks. The mainstreaming initiative was devised along the lines of familiar 'safeguards' provisions applied to World Bank projects with respect to environmental and social issues. Building on this work, the Asian Development Bank and European Commission forayed into mainstreaming, relying on expert analysis and advice from David Mansfield (see, for example, Mansfield 2006). After a pilot application of the mainstreaming approach to several World Bank supported projects, an effort was later made to formalize the mainstreaming process and apply it routinely for projects under preparation and appraisal, but the initiative did not move beyond this stage and was not integrated into the implementation phase of projects.
The Bank's analytical work on drugs in Afghanistan expanded and deepened during the next several years, in collaboration with other agencies. One highlight was a major World Bank-UNODC report published in 2006, which covered the full range of activities of the drug industry in Afghanistan (including trading, organized crime and money-laundering aspects, among others), from macroeconomic, development, and counter-narcotics policy perspectives (Buddenberg & Byrd 2006). Another highlight, where DFID took the lead, was a joint report focusing on economic and development-based approaches to reduce opium production in Afghanistan over time (Ward et al. 2008). In addition, several individually authored publications conveyed this analytical work and policy recommendations in differing formats, generally advocating for a comprehensive perspective and multi-faceted development-oriented approach (see, inter alia, Byrd 2008;Byrd 2010).
There was also an effort to link the Afghanistan-specific work with drugs and development more broadly. One example was a World Bank Poverty Reduction and Economic Management (PREM) Conference session on "Drugs, Security, and Development" (on 26 April, 2006), at which panelists included experts on drugs in other countries and general illicit economic activities and development (see Byrd 2006). Another was inputs and a contribution to a Bank publication that attempted to deal with drugs and development issues more broadly across countries and at a global level (Keefer & Loyaza 2010). However, despite some success in helping bring attention to drugs and development, little progress was ultimately made in expanding the Bank's engagement, in contrast to its active role on this issue in the Afghanistan context.

Loss of momentum and reversals
Despite the promise of its (for the World Bank) pathbreaking analytical work on drugs in the Afghanistan context and the fruitful collaboration with other agencies and serious efforts to develop the mainstreaming initiative, there were signs of loss of momentum by the second decade of the 21st century. The Bank has written and published less on drugs in Afghanistan, engaged less pro-actively with partners on this issue, did not implement counter-narcotics mainstreaming in any of its operations beyond the early pilot initiative, and may have suffered from loss of institutional learning and memory. The only notable exception, which however may prove the rule, is an extended background paper on the opium economy (Byrd & Mansfield 2014), prepared for the World Bank's Afghanistan Agricultural Strategy Review (World Bank 2014), which incorporated some of the background paper's main findings. While still in a better position regarding drugs in Afghanistan than when the Bank started out in 2003, there has been not only a loss of forward momentum but also backsliding from the peak level of engagement in the second half of the previous decade.
Why this change? Some erosion of momentum and forward progress is fully understandable, particularly on the analytical side-after several major reports and a number of publications, repetition of such initiatives would not have been warranted, and could have been replaced by lower-profile maintenance and updating of the knowledge base on drugs in Afghanistan. It also seemed clear by the end of the previous decade that the mainstreaming initiative had become stalled after the initial progress achieved, and despite occasional efforts to revive it, that has remained the situation in recent years. However, the reduced engagement with partners and the apparent partial loss of institutional knowledge need explanation.
Beyond the daunting pre-existing obstacles to Bank engagement on drugs and development outlined earlier, two additional factors likely contributed to such an outcome. First, changes in staff and management meant that those engaged on the drugs issue in Afghanistan at both levels moved on to other positions. While such turnover is quite normal, the lack of expertise in the World Bank on drugs and development more generally meant that, inevitably, replacements had to start from scratch on the subject matter even if they had some background on Afghanistan (which often was not the case). And the ending of the high-level World Bank-UNODC dialogue when Wolfensohn and later Costa completed their terms, may have weakened over time the top-level interest and support for continuing pro-active Bank engagement on the drug issue in Afghanistan.
Second, the extremely rapid, near double-digit growth of the Afghan economy during the dozen years after the 2001 international intervention, combined with modest reductions in poppy cultivation and opium production following the 2007 peak, meant that the share of the opium economy in overall economic activity was progressively declining over time-the opium economy (at border prices) shrank to the equivalent of 10-15 percent of GDP, as compared to 25-40 percent in the early years after 2001. This reduced its relative impact compared to other factors, especially the enormous inflow of international military expenditures and aid during the "surge" period. Thus, from a macroeconomic perspective, the opium economy was becoming less important, with the corollary that it became lower-priority for economic analysis. These trends have reversed with the decline in the GDP growth rate since 2012 and the resurgence of opium cultivation and production.
With regard to mainstreaming, risk aversion and inertia in development agencies understandably may have played a role in the lack of progress. In the face of the efforts of counter-narcotics agencies (both Afghan government and international partners such as UNODC and USAID) to carve out space and funding for ' alternative livelihoods' projects (see EastWest Institute, 2016), the World Bank and other development agencies may have found it easiest to cede this space to the counter-narcotics community. The initial design of counter-narcotics mainstreaming, which focused on individual projects and on 'screening,' may have given rise to a perception that mainstreaming was a kind of 'safeguards' exercise, adding a further burden to that of the other safeguards that development agencies already had to deal with. Beyond the initial analytical work and project reviews to develop the mainstreaming concept, lack of expertise and resources for mainstreaming work on a continuing basis also would have hindered mainstreaming from being taken forward. Sector and project staff could not be expected to have the necessary familiarity with the opium economy and its linkages to development, so without outside expertise it would have been very difficult for them to carry mainstreaming forward.
And finally, as aid focused on short-run stabilization during the surge period and counter-narcotics efforts were incorporated into the counter-insurgency campaign, the demand for economic analysis of the drugs issue, let alone receptivity to well-founded cautionary messages from a development perspective that progress takes time-measured in decades rather than years-became much attenuated. Hence there was little external pressure, or even necessarily strong support, for the World Bank to continue in the relatively proactive roles it had taken on earlier with general encouragement from partners.

Possible Ways Forward
The World Bank's experience in Afghanistan provides both a demonstration of the potential for proactive engagement by IFIs on drugs and development and a sobering reminder of the limitations such efforts encounter. Looking to the future, what are the prospects and options? What is the potential for the World Bank and other IFIs to engage on the drugs issue in other countries? The discussion below first looks at possible entry points for engagement on drugs from a development perspective, both longstanding entry points and potential new ones including at the global level. Then, after revisiting the mainstreaming issue, the paper concludes with some final thoughts.

'Older' Entry Points
The public health dimension is a longstanding potential entry point, which extends to the implications of drug use for HIV/AIDS, a global theme for the World Bank. This entry point is available in any country where there is relatively high problem drug use and the Bank provides support to or is otherwise engaged with the public health sector. So, at the country level public health constitutes by far the most common possible entry point for IFI engagement on drugs and development, albeit compartmentalized.
Drug crop cultivation and its nexus with agriculture and rural development constitutes a broader entry point for World Bank and IFI engagement. This was the main entry point in the case of Afghanistan, and does potentially allow a more comprehensive approach, including macroeconomic, agricultural, rural development and other impacts. However, this entry point is relevant only for the tiny handful of developing countries that are major producers of illicit narcotic drug crops. Byrd: Unexploited Potential? What Role Can International Financial Institutions Play in Drugs and Development? 114 The governance impact of drugs is another entry point, though challenging to utilize unless in combination with other dimensions. For example, the Bank's 2017 World Development Report on governance includes a number of references to drug trafficking (World Bank 2017b).
Related, in the small number of countries that are large producers of illegal drugs and/or lie on major transit routes, macroeconomic linkages may provide, together with others, an additional entry point. Limited data and poor data quality on illicit activities (including drugs) constrain the ability of economists to incorporate them in quantitative analysis of countries' economies. However, when the illicit economy of a country is large enough, not incorporating it can result in distorted findings and recommendations, so at least crude efforts to do so are warranted. Moreover, in the case of Afghanistan, data on the area of opium poppy cultivation, though far from precise, arguably is more reliable than some of the standard macroeconomic indicators.

Possible 'Newer' Entry Points
Political economy is a relatively recent area of Bank emphasis, including at the individual project level as well as at the broader level of country assistance strategies. Examples of the latter are the World Bank's most recent Systematic Country Diagnostic and Country Partnership Framework for Afghanistan (World Bank 2016a; World Bank 2016b), both of which explicitly include mention of the importance of the political economy dimension. Where drugs, their economic impact, and especially the associated financial flows are important enough to have significant political economy implications, this could be at least an indirect entry point for engagement on drugs and development.
Money laundering and illicit financial flows are a serious concern for the World Bank, which is heavily engaged in the anti-money laundering/countering terrorist financing (AML/CFT) agenda (see, for example, World Bank 2016c, and on the issue of illicit financial flows World Bank 2017b: Spotlight 13). Wherever the drug industry generates substantial financial flows in-country and outward to money laundering havens etc., this could be an entry point for the Bank and other IFIs on drugs and development.
More generally, illicit economic activities and their development impacts are of increasing interest, and drugs comprise one of the most important components both globally and in some countries.
The conflict dimension provides yet another potential entry point. For example, a recent World Bank -UN publication on conflict prevention refers to drug trafficking in several places (see World Bank Group and United Nations 2017). Specific interfaces may include violent conflict over control of drug activities and profits, narco-financing of insurgencies, organized crime, violence associated with distribution and consumption of illegal drugs, and not least the adverse development and human impacts of ill-considered counter-narcotics actions such as violent eradication of drug crops or chemical spraying. Related, in some countries the drug industry plays an important role in state fragility, which may provide an entry point as well.
Beyond their applicability in particular countries, some of these 'newer' possible entry points may also be relevant for Bank engagement on drugs at the global level, to a much greater extent than the ' older' entry points.

Better Mainstreaming
The attempt to articulate and implement counter-narcotics mainstreaming in Afghanistan did not take off, but it yielded lessons that are relevant for developing an improved approach to mainstreaming. First and foremost, integration of the illicit narcotics dimension in development initiatives must occur at the strategic level. If the drug economy is not integrated in broader development strategy and relevant sector strategies (such as agriculture and public health), trying to mainstream counter-narcotics in individual programs and investments would not work. However, such a strategic approach on the part of the World Bank might well be controversial, and could be contested by various other actors that have explicit counter-narcotics mandates.
In assessing whether and how to try mainstreaming in a given country, the first step would be to identify the salient interfaces between drugs and development in that country and the most promising entry points for engagement. In the overwhelming majority of developing countries, the drugs issue is not likely to be of sufficient importance as a development issue to justify mainstreaming, and if it is of major importance in only one dimension, for example, public health, it would make sense to employ a compartmented sectoral approach rather than attempting to engage in broader mainstreaming. On the negative side, it would also be important to assess the receptivity of the government of the country concerned to Bank/IFI engagement on drugs as a development issue, and in some cases that of other key partner countries as well.
In the small number of countries where the drugs-development nexus is of great importance and the government is receptive to Bank/IFI engagement, the drug economy needs to be integrated in country macroeconomic and development analysis, as well as in work on political economy, governance and corruption. It would also be necessary to incorporate drugs in sector strategies for the key sectors where important entry points have been identified (e.g. agriculture and rural development in major drug crop cultivating countries).
Starting from a broader strategic and sectoral level, mainstreaming at the program and project level could use a screening approach similar to that in the Afghanistan counter-narcotics mainstreaming approach, but emphasizing opportunities as well as risks. As noted, counter-narcotics mainstreaming is unlikely to work if it is seen solely as a 'safeguard' exercise, imposing yet another burden on already overburdened project staff, and running the risk of degenerating into a bureaucratic ' checking-the-box' process. Any future mainstreaming initiatives should move away from a solely safeguards-like approach, and emphasize positive synergies, bring different forms of expertise to bear, and take a practical, solution-oriented approach.
It will be important that, for the countries where mainstreaming is attempted, relevant expertise is deployed, since project staff of the IFIs cannot be expected to have or quickly build up expertise on the drugs-development nexus, both in the country concerned and more generally. Therefore, outside expertise (of the right kind) will be required, that can help both in understanding the country context and its drugs dimension, and how to implement effective mainstreaming. This should be adequately resourced.
Mainstreaming should not become a justification for excessive caution and risk avoidance. Drugs are a complex and often politically sensitive problem, but there are also risks associated with not approaching drugs as a development issue, in the small number of countries where it is of high importance. For example, the risk that newly irrigated land will be allocated by Afghan farmers in part to opium poppy cultivation is real, but it does not mean that no irrigation projects should be undertaken. On the contrary, irrigation is a prerequisite for moving Afghanistan away from dependence on poppy cultivation over the longer term.
Finally, mainstreaming drugs as a development issue could be considered at the global level, not just at the country level where few countries are suitable candidates. This means identifying key interfaces at that level, and entry points that are consistent with World Bank/IFI global themes (such as illicit financial flows, conflict dimension, etc.). Such an approach would require a favorable ' authorizing environment' at the level of top institutional leadership and shareholders, which may not be present or easy to develop.

Conclusions
Afghanistan demonstrates the potential and benefits as well as the limitations of World Bank/IFI engagement on the drugs issue in a country context. The Bank did meaningfully engage in several ways, and its involvement was well-received by at least parts of the Afghan government and donor community. In a modest way, it made a difference. However, the Bank's engagement, after peaking in the second half of the previous decade, subsequently lost momentum and suffered reversals for reasons discussed in this paper.
Given the constellation of positive factors that came together and supported Bank engagement on drugs in Afghanistan and helped enhance its impacts, it is hard to envision the World Bank or other IFIs advancing farther in other countries, or even in Afghanistan in the future, in terms of the breadth, depth, and longevity of its engagement in the drugs issue. Realistically, therefore, expectations for Bank/IFI engagement on drugs need to be modest. There may be opportunities here and there to exploit, but no sign that a sea-change would be possible.
Whether progress may be possible in terms of engagement on drugs and development at the global level, where the World Bank is emphasizing several global themes that are related to drug trafficking, is an interesting question. There may be some opportunities, especially in the areas of illicit financial flows and violent conflict. However, past experience does not provide much ground for optimism that the Bank could take on a proactive role with regard to drugs at the global level. It would seem more likely that given so many other priorities and challenges the Bank faces, as well as shareholder sensitivities, the Bank and other IFIs will not lead the global drugs and development agenda but may be brought in over time by other stakeholders, particularly member countries concerned about drugs and drugs policy as a development issue.