Loading content


When things fall apart

In this blog the author reflects on the deterioration in donor expectations since the beginning of Lessons for Peace, and on the choices donors now face in Afghanistan.

Published: 03 August 2021

Image: Funeral in Kabul, CC Koldo

About the author: Nigel Roberts has been the Lead Consultant for the Lessons for Peace Project (L4P) since its inception. Over the past 40 years he has worked as an economist in many countries undergoing violent transitions, including Ethiopia, Lebanon, Nepal, Palestine, the Philippines, Somalia and Ukraine. He was the co-director of the World Bank’s 2011 World Development Report on Conflict, Security and Development.

A darkening prospect

ODI’s website tells its readers that Lessons for Peace (L4P) was launched in May 2019 because “international donors… felt an urgency to be ready to support any emerging peace process….particularly in the likely scenario of a government-Taliban power sharing arrangement”. Much of this donor urgency centered on the Taliban. What were their true intentions? Would they prove more flexible than the 1996-2001 Taliban government? Would Afghanistan’s need for continued external funding translate into continuing donor leverage? ODI believed that donors could benefit from a project that shed some light on these questions. There is also a rich literature on Afghanistan’s development experience, and we also felt it would be useful if L4P could bring the more important lessons of this experience into donor planning processes.

At L4Ps’ inception, various government and donor documents anticipating a peace agreement were in circulation. This agreement, it was generally assumed, would involve the Taliban in the nation’s governance, implicitly in a ‘manageable’ role, and could open the door to a brighter economic future.

Two years later, with the last US and NATO troops leaving amid widespread Taliban battlefield gains, those documents have little apparent relevance.

While much valuable work has been done by L4P, our team – along with most other analysts, commentators and aid practitioners - remains unable to answer the question that has preoccupied donors from the outset of the project: what can we expect from a government in which the Taliban exercise significant power? Or, more realistically now, how will the Taliban govern if, as seems increasingly likely, they come to dominate the apparatus of state? Definitive answers remain elusive. In part this is because Taliban actions in areas they have occupied does not necessarily predict their behavior as national rulers. Recent Taliban announcements could be interpreted as relatively encouraging in relation to women’s and girls’ rights, but they still suggest a darker future than the mid-2019 planning documents were hoping for.

It is clear from the Afghanistan Partnership Framework and from recent public statements that today’s key donors are allergic to supporting any government that rolls back gains in democratic expression, women’s rights and the creation of responsible state institutions. What is less clear is how these principled views will play out when they are tested. Afghanistan’s extreme donor dependency is well-known. Up to now, donors have aspired to moderate Taliban policies, believing the Taliban understand the importance of aid and hoping it provides some leverage. How much erosion of fundamental rights can be tolerated before donors withdraw their support, though? And what more should donors do to try and develop a meaningful socio-economic policy dialogue with the Taliban?

Asking such questions means accepting a real possibility of Taliban state capture. Difficult as this still is to admit to at a political level, donor planners need to consider this outcome. The cost of failing to do so in time will be a last-minute mess: hasty decisions that cater to domestic public opinion, rather than more considered plans that balance the defense of human rights and the needs of all of Afghanistan’s vulnerable citizens. Helping donors work through these questions is arguably the most important task for L4P in the coming months.

The donor record

Much of L4P’s efforts were devoted to examining donor experience over the past two decades, and I will briefly recall some of those findings.

Unprecedented amounts of foreign aid from 2001 onwards helped drive impressive socio-economic achievements, including a 75 percent increase in average real per capita incomes, the emergence of modern civil society organizations and unprecedented advances in the status of women and girls’ education – yet Afghanistan finds itself in the lowest ten percent of countries under five of the six World Governance Indicators. The tremendous gains in welfare have been undercut by regional powerbrokers appropriating public and private resources and preying on citizens through national security and justice institutions. With the decline in aid flows since 2012, economic growth had stalled and poverty rates had surged even before Covid-19 hit.

Rather than counteracting political dysfunction, aid has exacerbated it. The US-brokered 2001 Bonn Agreement reinforced centuries of center-regional conflict by creating a Pashtun-dominated state and simultaneously empowering militia commanders who had helped oust the Taliban, despite their unpopularity with the general population. At the same time, Bonn excluded the Taliban, and denied them a reintegration pathway into Afghan society - a policy of exclusion and missed opportunity repeated until the Taliban could negotiate from a position of strength. Tragically, the predation and corruption enabled by external assistance subverted the ideological legitimacy of the post-Taliban order and facilitated the Taliban’s resurgence. This allowed them to capitalize on a powerful Afghan historical trope: resistance to a corrupt puppet government created by outsiders intent on subverting Afghan culture and Islam.

These problems were amplified by the way donors managed aid to Afghanistan. A confused, disoriented rush to deliver ‘results’ channeled funds to the unscrupulous, raided the government for Afghan talent, created redundant systems, distributed resources inequitably and short-changed sustainability. The widespread use of off-budget channels, particularly for short-term security objectives, severely impeded the influence of the national government. Some 85 percent of all grant support since 2001 has bypassed government systems, the lion’s share through Provincial Reconstruction Teams and civilian contractors pursuing ‘stabilization’ objectives; much of this was badly supervised and poorly coordinated, stimulating the emergence of a contracting and services economy dominated by local strongmen with preferential access. In addition, excessive numbers of civilian projects created a fragmented, wasteful aid environment featuring a proliferation of parallel project implementation structures and inflated pay scales that drew many of the most talented away from the civil service. These acknowledged failures have contributed to donor fatigue and the possibility of crippling reductions in aid levels, particularly if the Taliban seize power.

What useful lessons can be deduced from this? Afghanistan’s extensive aid evaluation literature suggests several starting points.

  • The most obvious is the need for a clear-eyed understanding of Afghan power dynamics, and of why donors have resisted tailoring their programs to these realities; this also implies a better understanding of aid as a source of political ‘rent’.
  • It is also evident (reflecting experience in other countries) that the most consequential and sustained development enterprises, for all their well-known flaws, are the on-budget ‘national’ programs financed through the Afghanistan Reconstruction Trust Fund, and through multilateral channels.
  • On-budget contributions, however, have failed to counteract the impacts of perverse governance. Part of the reason for this is that donors have worked at cross-purposes, and much off-budget funding has fuelled abusive behavior; at the same time, donors have often been reluctant to use their leverage to demand better governance – for all the rhetoric of four successive government-donor compacts. This unwillingness to sanction the government reflects the hazards of confronting an ally in a war on a common enemy, and needs careful rethinking in today’s changed context.

What should donors now expect?

L4P and the planning documents prepared in 2019 were premised on a peace process that would lead to a power-sharing arrangement with the Taliban – implicitly, one that would not seriously compromise the development aspirations of the past two decades. Thus in the Joint Coordination and Monitoring Board’s July 2019 Afghanistan – Day After Peace Program, the government declared that Afghanistan is closer than ever to achieving a historic peace agreement which would put an end to the decades of fighting and loss of life”, while the World Bank’s 2019 Financing Peace opened with “The international community is currently planning for a potential political settlement in Afghanistan….. This note provides an analysis of overall medium-term financing needs and identifies implications for ongoing grant support and post-settlement programming”.

Current planning documents are much less optimistic. Recent World Bank work with which I have been associated, for example, identifies three possible political scenarios for the coming 1-2 years. None of these are encouraging. The first two differ in the speed and extent of political deterioration: one envisages a Taliban military ‘victory’ which would usher in a period of governance turmoil, accompanied by significant armed resistance in various parts of the country, and by rapid economic decay (steep declines in aid transfers and private investment, and further increases in poverty and hardship).

The second posits a government still able to hold onto key cities and trade routes, but unable or unwilling to pursue substantive reform; declining levels of aid and investment and intensifying hardship also accompany this scenario. Common to both are continued shrinkage of space for civil society, further erosion of democratic norms and women’s/girls’ rights, and an accelerated migration of educated Afghans.

The third scenario, now seen as the least likely of the three, assumes a negotiated peace and a power-sharing arrangement between the government and the Taliban; as a result, overall security would improve, though some armed resistance and criminal violence would persist. Medium-term economic prospects should improve if the settlement lasts, particularly if better regional cooperation emerges: but immediate economic relief could be slow in coming - government decision-making is unlikely to proceed smoothly, rent-seeking would persist, and donors and investors would initially hold back.

These scenarios are typical of what donors are discussing internally, and their implications are very different from those underpinning the 2019 planning documents:

  • The struggle for control over territory, resources and trade routes is likely to continue, whether or not a peace agreement is concluded, with violence remaining the default mechanism for sorting out disputes. Conventional notions of ‘good governance’ are a largely irrelevant consideration in such a landscape; clientelist/private priorities will continue to trump concerns about public welfare.
  • The security conditions facing donors and their NGO/commercial project contractors will not improve soon, if at all.
  • Aid management and coordination will remain messy; future governments will continue to struggle with donors’ preference for off-budget aid, while today’s donors – their contributions in decline, their influence reduced - will find it harder than ever to agree common priorities, negotiate reforms or impose credible conditionalities. These difficulties will be compounded if regional neighbors become more involved in a post-US settlement, either financially or by sponsoring anti-government militias.
  • Efforts to encourage private investment will remain arduous until and unless a stable, recognized regime emerges.

What then should donors do?

Unstable governance, poor security, a fragmented international policy agenda and low private investment levels will undercut comprehensive, ‘progressive’ development agendas of the type articulated in the Afghanistan National Peace and Development Framework II 2021-2025. The directions in which Afghanistan is likely headed mean that donors should place a higher premium on flexibility and opportunism than on conventionally balanced programming.

  • If the Taliban capture the apparatus of state, donors should do what they can to help preserve basic service delivery and the core fiscal functions of government, while stressing the importance of these building blocks to any version of a credible state. Donors will want to help protect an increasingly vulnerable population; realistic approaches might include combinations of cash transfers to impoverished families, short-term employment schemes and conventional humanitarian relief. A tempting default is to transfer development funds en bloc to humanitarian agencies – but, as the experience of the 1990s shows, this should not be done lightly as it is likely to result in a withering away of essential services.
  • A politico-military stalemate offers more scope to sustain or adjust ongoing initiatives in support of agricultural production and small and medium enterprises (SME), particularly in and around the more secure urban areas; this might help retain some investor confidence and slow the pace of capital flight. Donors’ main focus should still be on maintaining critical state functionality at manageable cost, though. It will also make little sense to continue pushing reforms that have met serious resistance under far more propitious circumstances – such as professionalizing the civil service, creating an accessible and fair justice system, proper oversight of the minerals sector, or systemic efforts to eradicate public sector corruption.
  • If a political agreement emerges, there could be scope for a more transformational donor response. It would be vital for donors to focus their contributions better, and to reinforce what has worked well in Afghanistan and in other countries attempting recovery from bitter, protracted civil wars. Counterintuitively for development technocrats, part of the challenge is to ensure that sufficient economic dividends are made available to powerbrokers and their followers, and are distributed in ways that keep key spoilers inside the new political settlement – in contrast to the lopsided and unconditional rent distributions of the past. In addition to sustaining the integrity of the main national service delivery programs and extending them into areas currently under Taliban control, a combination of short-term income generation opportunities and medium-term investments in the productive sectors (in particular, agriculture and SMEs) would be needed, as would measures to support financial inclusion and draw new private capital (including foreign direct investment) into areas of potential. A stable peace could lead to better Afghanistan integration with regional markets, and could make private investment in trade-related infrastructure attractive.

The difficulties that all these scenarios embody and the lessons from the past show how important it is that donors look at how they will operate as much as what they will finance.

  • Operations should be designed with a better understanding of the evolving political context – not only the extent to which they can help address grievances and lessen vulnerability, but also the manner in which they are likely to benefit rival political actors and/or be subject to informal capture. To assist technical staff in making these complex determinations, and to adjust their operations in response to power realignments, donors can make use of simple governance and conflict checklists. These can force staff and management to focus on political economy ‘entry conditions’, and make timely adjustments during implementation. Future L4P work on ‘adaptive programming’ will aim to provide further guidance on this.
  • While the pursuit of major reforms seems inadvisable under current and projected realities, donors should be much firmer at holding the line on project-level agreements and conditionalities. Clear standards should be established during project preparation, and agreements should be adhered to – examples include insisting that senior positions are staffed by qualified appointees, that project enabling conditions are sustained (staffing, counterpart budget allocations, land acquisition, community consultations, etc), that regular and reliable auditing takes place, and that instances of project-level corruption have specific consequences. This is particularly important in the key service delivery and fiscal management programs supported under the ARTF, but will require a greater willingness by ARTF donors to suspend operations and/or halt new commitments while egregious breaches of agreements are being remedied.
  • Conversely, though, donors also need to be clear about the degrees of imperfection they are prepared to accept in pursuit of continued service provision. How much ‘taxation’ of project activities by different armed groups should be tolerated? To what extent are donors willing to overlook conservative local adjustments to school curricula, or the exclusion of girls and women from educational and economic opportunities?
  • Donors must decide how far they can stretch their fiduciary oversight standards with little or no on-the-ground-presence - without making unacceptable demands on the NGOs on which they are so dependent. Donors should normalize Covid-19 related remote supervision and monitoring arrangements, and should expand the use of remote verification technologies and third-party proxies - but if access by NGOs, private contractors and the ARTF Third-Party Monitoring Agent becomes too restricted or too risky, and if communities are unable to exercise control over project resources, donors will presumably be obliged to suspend conventional operations. One variant of this dilemma may arise if the central government begins to disintegrate, and operations through government systems become impossible. In anticipation of this, donors should establish contingency plans that allow for the delivery of some residual services through humanitarian actors, assuming that these actors and their NGO proxies are themselves still able to operate.
  • Donor impact on policy and service delivery in an increasingly challenging environment will depend even more than ever on effective coordination. This will not be achieved through rhetorical declarations or aspirational documents like the current Afghanistan Partnership Framework: it requires collaboration at an operational level. In Afghanistan, pooled, on-budget assistance from the EU and under the ARTF has proven its worth despite difficult governance and security conditions, and the ARTF’s achievements need better promotion to counter flawed western public perceptions of the pros and cons of working through imperfect government systems. The US and its allies also need to put their diplomatic weight behind constructive regional efforts to help Afghanistan, and can reinforce these by investing in regional energy and trade connectivity.

Beyond development programming: what we still owe Afghanistan

President Biden’s withdrawal announcement, along with recent Taliban battlefield gains and an increasing fear of a government defeat, has sparked a renewed public focus on Afghanistan. Part of the discussion focuses on the importance of continuing support to the post-withdrawal government, in particular the Afghan military. There is also debate on whether or not a Taliban-dominated Afghanistan increases terrorist risks to the US and Europe.

Importantly, others are focusing on what can and should be done for those Afghans who have led the attempt to create a pluralistic, accountable state if the Taliban take over. Several governments are facilitating the provision of visas to those who have worked directly with coalition armed forces and embassies. The list of those at threat from a Taliban takeover, though, is a much longer one, and includes those in leadership positions in the Afghan government, police and the armed forces as well as aid project officers, civil society activists and journalists – quite apart from those many Afghans who have little desire to live under a potentially oppressive regime. Refugee agencies are predicting a significant increase in the numbers of Afghans fleeing to neighboring countries, many of whom will be seeking asylum in Europe and the US. These are, in many cases, people who have responded to the democratic, rights-based agenda that donors have championed in Afghanistan over the past two decades.

The Afghanistan Partnership Framework strongly underlines the importance to donors of any Afghan government upholding and deepening these rights. If the Taliban come to power and do not stick to this script, donors will no doubt sanction the government by reducing aid contributions. The need for a careful appraisal of how much policy imperfection should be accepted in the name of general public welfare has already been mentioned. Beyond this, though, is the need for donors to provide adequate support for UN refugee operations in neighboring countries, as well as to take in a sizeable number of those whose embrace of international values is leading to their exile. This is a moral issue, and cuts against a politicized resistance to immigration in many coalition countries; it is also a pragmatic issue, since washing one’s hands of former allies gives a chilling message to other countries combating extremist ideologies.

How to deal with the collateral human damage from the Afghan wars is perhaps the greatest of the tests that donors now face. When they drafted the Afghanistan Partnership Framework, they were criticized for placing an excessive weight of obligation on the government. A Taliban takeover would place the weight of obligation back onto the donors.