Stabilization:
Lessons from the U.S. Experience in Afghanistan

During the 2009 Afghanistan strategy reviews, President Obama and his advisors set in motion a series of events that fostered unrealistic expectations of what could be achieved in a few years and ensured the U.S. government’s stabilization strategy would not succeed, first with the rapid surge and then the rapid transition.

The President and his advisors relied on poor comparisons between the success seen in Iraq at the time and what they hoped to accomplish in Afghanistan. They focused on troop numbers and their geographic priorities and mostly omitted concerns about the Afghan government’s capacity and performance. The scope of the strategy was not revised when the president shortened the surge timeline from an estimated 10 year to 18 months. As a result, the U.S. government devised a strategy far more ambitious than it originally anticipated and poorly suited to the time allotted to achieve it.

Worse, the narrowness of the surge and transition windows had a profound and harmful impact on countless subsequent decisions regarding planning, staffing, and programming.

Ultimately, the demand for fast progress meant the country had not stabilized when the transition began in the summer of 2011, and Afghans were unprepared to take the reins when the transition finished in December 2014.

To compete with the insurgents, the U.S. government assumed the Afghan government would need to out-govern the Taliban and provide services that went well beyond what the Taliban had offered. USAID and DOD programs sought to enable the Afghan government to provide health clinics, schools, retainer and flood walls, agricultural training and seed distribution, and countless other projects, while the Taliban only offered a modicum of security, and in some cases, dispute resolution.

The ambitious nature of the strategy meant that in most key districts, the coalition helped the government do many things poorly, rather than a few important things well.

Under pressure to make fast progress, the amount of money spent was often regarded as an indicator of success. The difficulty of discerning program impact only compounded the problem and made it tempting to use money as a proxy for effect. Yet, the sums were far more than Afghanistan could absorb, which fueled corruption. Moving at high speed, the coalition often exacerbated the very problems it hoped to mitigate. Powerbrokers with access to coalition projects became kings with patronage to sell, and stabilization projects created or reinvigorated conflicts between and among communities. In turn, Afghans who were marginalized in this competition for access and resources found natural allies in the Taliban, who used that support to divide and conquer communities the coalition was keen to win over.

Most of the coalition’s key districts proved so dangerous they had to be continuously cleared, which meant that many of the stabilization programs designed to jumpstart local governance and development were implemented prematurely and in communities ill-suited to host them. Coalition stakeholders at every level recognized in principle that physical security was a prerequisite for stabilization, yet in practice, stabilization usually took place in chronically insecure environments. As a result, Afghans were often afraid to serve in local government, and Afghan civilians had little faith their districts would remain in government hands. When the coalition eventually withdrew, implementing partners struggled to execute projects amid the violence, and U.S. government agencies were unable to adequately monitor and evaluate the projects that were implemented. So, rather than tackle the hardest districts first and use the momentum to stabilize less challenging districts, this prioritization meant that less was accomplished overall as the coalition was increasingly bogged down.

While poor security made it difficult to monitor and evaluate stabilization programs, the U.S. government had not put the necessary thought into how to do so, even in areas that were more accessible. Agencies typically used the amount of infrastructure built and the number of civilians employed or trained as indicators of progress. Analyzing these outputs was a poor metric for understanding whether Afghans’ support for the government was increasing and their support for the insurgency was decreasing. The United States implemented stabilization programs in Afghanistan from 2002–2017, yet only in 2012 did USAID begin to take a more thoughtful approach to measuring impact for stabilization programs, and DOD never did.

Afghan forces and civil servants were generally unwilling, unprepared, or unable to carry forward the momentum created by coalition forces and civilians, particularly on the unrealistic timeline defined by the coalition. By the time all key terrain districts had transitioned in 2014, the services and protection Afghan forces and civil servants were in a position to provide could not compete with the threats of a resurgent Taliban. Baseline capacity for Afghan institutions was so low that they could not become self-sufficient on the stabilization strategy’s timelines, particularly in the most dangerous districts in the country. Similarly, because stabilization is only a stopgap measure, gains were lost when little follow-on programming was implemented as the coalition withdrew.

Successful stabilization projects generally occurred in areas that met four conditions prior to projects beginning. First, physical security had to be sufficient to allow Afghans to feel comfortable openly aligning with the government. Second, the building blocks for rudimentary governance needed to be in place to give Afghans some form of local representation prior to the commencement of stabilization projects. Third, coalition forces and civilians had to recognize the value of methodically working and planning together to jumpstart service delivery. Fourth, the government’s engagement had to be continuous and focused on process, not product. The number and dollar value of projects implemented far exceeded the coalition’s ability to monitor and evaluate them. Opportunities for corruption abounded, making many of those projects more harmful than helpful.

For future large-scale stabilization missions, the U.S. government should set reasonable expectations for what can be accomplished. Unless the expected timeframe exceeds 10 years and includes meaningful programs and investments following the shorter-term stabilization component, such a mission will likely be ineffective. With compressed one- to three-year timelines like those seen in Afghanistan, programs will be forced to compromise quality as they ramp up and finish too quickly, and the population will not trust the staying power of the government’s improved services and legitimacy. Without the patience and political will for a planned and prolonged effort, large-scale stabilization missions are likely to fail.

As occurred after the Vietnam War, there is a tendency to believe that the U.S. government should not and will not conduct large-scale stabilization missions in the future. However, there is little reason to believe this fatigue is permanent, as there will likely be times in the future when insurgent control or influence over a particular area or population is deemed an imminent threat to U.S. interests, and it may not be possible to rely on partner forces to clear and hold the area. To ensure the U.S. government is better prepared for such a scenario, the capabilities, institutions, and cultural orientation toward counterinsurgency and stabilization that was built through trial by fire over the last 17 years should be maintained and honed, as they are likely to be critical in the future.

Every organization and agency that worked on stabilization in Afghanistan—from civil affairs and SOF to State and USAID—suffered from personnel and programming deficits borne from rapid scaling, short tours, and the pressure to make quick progress. No organization was prepared to ramp up quickly and it showed across the board. Given the nature of most contingency scenarios, rapid scaling may be unavoidable, so properly identifying and preparing the appropriate personnel ahead of time is paramount.

Unrestricted funding can exacerbate local conflicts, drive corruption, and distort local economies, particularly if compressed timelines are driving spending. The stabilization effort in Afghanistan was derailed as money spent became the metric of success and programming drove wedges within communities and between communities and the government, rather than between insurgents and the population.

Stabilization is most likely to be successful where there is basic security for implementing partners and local stakeholders on the ground. If stabilization programming is implemented in areas under incessant attack, host nation civil servants will be unlikely to volunteer and reliable information on the population will be elusive. Progress necessary to build momentum will also take far longer to achieve, and the population will be slower to trust that any gains will endure. Further, in these areas, implementing partners will have less oversight, limited freedom of movement, and reduced capacity to implement quality projects.

Attempting to simultaneously stabilize an area and build local governance structures is unlikely to be effective. Some semblance and history of local government must be in place before an area can be stabilized with robust programming. Supporting pre-existing informal governance structures (or rebuilding them) may be preferable to building formal government, which is both costly and often culturally unfamiliar. Ensuring target communities have competent, accountable, and sustainable local governance is important for effective stabilization programming.

In ungoverned or poorly governed areas, top-down efforts that attempt to stabilize diverse communities are not effective. While this was recognized in theory, too often locally tailored approaches were forced aside to accommodate changing Coalition priorities, which were imposed on stabilization programmers at the expense of addressing diverse local sources of instability.

No matter the logistical or political obstacles, impact is the primary criterion against which a project or program’s success should be judged. Unless a stabilization mission is just starting, no other metric—money spent, number of projects, people employed—should be used in determining the success or failure of a stabilization program. While the high number of variables in stabilization environments makes it difficult to discern cause and effect, programming should not take place in areas where it is impossible to monitor and evaluate. From the beginning of the mission, every agency involved in stabilization should plan to prioritize and discern program impact, and they should dedicate more staff to oversee monitoring and evaluation efforts.

The U.S. government either did not know enough about the Afghan government and population or paid insufficient attention to them during the strategy reviews. The sense of success in Iraq by 2009 lulled the U.S. government into believing the governance challenges in Afghanistan were equivalent; they were not. This disconnect made programs overly ambitious and set up to fail in the timelines provided.

On the ground in Afghanistan, DOD, State, and USAID implemented programs without sufficient knowledge of the local institutions, sociopolitical dynamics, and government structures, which often exacerbated local conflicts, empowered insurgents, and created unnecessary enmity between the population, government, and the coalition.

The kinds of services the U.S. government sought to help the Afghan government deliver were unnecessarily ambitious and not tailored to the environment. While improvements in the delivery of healthcare, rule of law, education, and agriculture likely helped many Afghans, the coalition and the Afghan government aimed to provide contested areas an array of high-quality services that went well beyond what the Taliban had provided and would have required a level of capacity and legitimacy far beyond what the government could offer, particularly in the time allotted.

For various reasons, DOD in practice made the most consequential strategic decisions regarding the implementation of the president’s strategy. However, building local institutions, changing civilian perceptions, and buttressing a government’s legitimacy are political undertakings and should be implemented with instruments suited to these tasks.

With good reason, the National Security Council has institutionalized a Fragile States and Stabilization policy coordination committee to oversee U.S. national security policy in priority conflict-affected areas. If the NSC instructs the agencies to pursue a large-scale stabilization mission, State’s broader strategy should assume a timeline of 10–15 years or longer, and explicitly outline how shorter-term stabilization efforts in targeted areas will gradually transition to longer-term development and host nation control, according to realistic benchmarks defined by host nation capacity and conditions on the ground. The stabilization component of the strategy should be integrated into larger political objectives. Stabilization is not transformative, but rather a catalyst that will burn out if subsequent investments in governance and development are not integrated into a larger, singular strategy and applied in prioritized areas.

If implemented, this recommendation would ensure that the ends, ways, and means of the stabilization effort are properly calibrated with one another and are built on a political rather than military foundation.

Most communities accustomed to insurgent rule are not ready for stabilization projects immediately after clearing. They need the space to evaluate their new security forces and governance structures before they can commit themselves, so establishing security, building governance structures, and ramping up stabilization projects should not happen simultaneously. Using Afghanistan as an example, a more methodical approach for other environments would be as follows:

• An area is cleared of insurgents. Humanitarian aid is distributed and condolence and compensation payments are made, as necessary. DOD, State, and USAID personnel take the time to understand local relationships, rivalries, and sources of instability.
• Simultaneously, pre-existing informal governance structures are identified and evaluated to ascertain their legitimacy. All other stabilization programming designed to connect the population to its government should not begin until two conditions are met: (1) the entire community has steady and legitimate formal and/or informal representation and (2) a thorough examination of the community’s interests and vulnerabilities has been completed.
• Programs like VSO should be considered at this stage to ensure the community continues to have sufficient and regular physical security so stabilization programming can progress.
• The community is told that stabilization projects will only begin after it has demonstrated its support for the government by reporting IEDs, joining local or national police forces, publicly denouncing insurgent groups, and working with their informal and formal governance representatives to address community grievances. If they are not ready to do these things, then they need more time to trust the government’s staying power, and they are neither suited nor ready for projects, despite what they may say. Implementers need to be afforded the space to say when communities are not ready for stabilization.
• Subsequently, if the stabilization cycle is successful, the stories should be broadcast and leveraged in nearby areas on the cusp of being cleared, illustrating to neighboring communities what they can expect if they help keep the insurgents out once security forces clear their communities. Without reliable security, there can be no rudimentary governance; without rudimentary governance, there should be no stabilization projects.
• In practice, this will likely mean that the transition from “clear” to “hold” will take longer, but doing so will give the community time to adjust to a new normal as a permanent sense of physical security and legitimate governance takes root.
• This sequence should and will have an impact on the criteria for selecting areas for clear-hold-build. If an area is not physically secure by civilian standards, expanding the aperture to focus on building rudimentary governance and stabilization projects is wasteful, often counterproductive, and should not be attempted.

In many ways, the process described above somewhat resembles what was intended with stabilization in Afghanistan, but the strategic and tactical timeframes made it impossible to take the time to stabilize Afghan communities with the patience, priorities, and sequencing outlined above.

CERP, or any successor program, should incorporate measures of effectiveness into its monitoring and evaluation processes, and civil affairs should be trained on these processes. DOD should modify CIDNE, the database used to track CERP spending and projects, so that civil affairs officers are required to conduct a baseline assessment that justifies every CERP project, as well as an assessment of the project’s effectiveness after project completion and before the project can be formally closed. DOD should develop criteria for every CERP project to assess its impact; this impact assessment should be the sole criterion for judging whether it is successful, even if this means fewer projects are implemented.

Evaluating stabilization programming is difficult, given the number of variables, how quickly they can change, and the biases inherent in data collection efforts. These problems should not be magnified further with inaccurate or spotty data. USAID should develop baseline indicators for targeted areas before stabilization programming begins, and then to carefully track project spending at the community level with accurate GPS coordinates for every project, which USAID should hold in a centralized database. No matter how stability and its indicators are defined, understanding impact or even outcomes is far harder without properly and systematically comparing targeted areas before and after interventions, and ideally, comparing targeted and untargeted areas. Equally important, this process cannot be done without dedicating the necessary staff to design the research and collect the data.

Stabilization requires detailed information about what are often poorly or rarely studied communities and social groups. This type of information is difficult to collect and analyze, and the process requires a special skillset. In addition to qualitative research, among the many intangible skills involved are an ability to read social cues, to ask penetrating questions without being intrusive, and to understand what makes a community tick, all within a very short period of time. While localized subject matter expertise is very important, given that the locations of future contingency operations are difficult to identify years in advance, it is more important to recruit for, build, and retain these adaptable, generalized skills than it is to build country-specific subject matter expertise. For example:

• –DOD should establish a cadre of socio-political enablers for SOF units participating in VSO-like endeavors. U.S. SOF personnel assigned to Village Stability Operations needed an intimate understanding of local socio-political conditions and well-developed cross-cultural communication skills. Civilian or military personnel who have been carefully selected and trained for the role should be integrated as enablers into every stage of VSO-like endeavors. Each enabler’s mission will be to gather, analyze, and present socio-political information to the Village Stability Platform commander, help the VSO team build effective relationships with the local community, and help discern the appropriate ways to connect the community to local governance/government structures. The selection criteria for these personnel should include experience working in in developing countries, advanced training in research methods, and a psychological evaluation. Furthermore, these enabling personnel should be integrated into the SOF training cycle to institutionalize them and ensure prospective teams appreciate and understand how to use them.

While the above example is specific to U.S. SOF, U.S. conventional forces and USAID should adopt similarly tailored efforts and determine whether and how: the need for such expertise can be met with additional training for existing personnel in existing structures; new organizations or divisions have to be established to house and grow these internally developed experts; external expertise has to be recruited and cultivated; or some combination.

In recent years, the number of active duty civil affairs personnel has been decreasing. Currently, there is one active duty civil affairs brigade assigned to U.S. Army Special Operations Command, much as there was before the surge—but the active duty CA brigade headquarters and its five CA battalions that were stood up during the surge to support conventional forces have been reduced to a single battalion in recent years. The future of any CERP-like program and its thoughtful implementation should rest in the hands of properly trained and supported civil affairs personnel who are in the best position to prioritize and measure impact. Maintaining a well-resourced, active duty civil affairs brigade that regularly partners with conventional forces and civilian partner organizations is vital to the military’s preparedness for future stabilization missions, both large and small.

Recruiting, training, and deploying the required civilian personnel for large-scale stabilization missions is difficult, particularly with short time-horizons. Most contingencies will need to ramp up quickly with experienced civilian personnel who can build coalitions and understand diverse communities. Processes and structures need to be in place to ensure the quality of personnel does not suffer simply because of the urgency of the task at hand. Previous efforts to meet this objective were costly and hindered by political infighting, among other obstacles. Furthermore, justifying large expenditures today for an unknown future contingency is difficult. With only modest investments ahead of time, the U.S. government can maximize already-available expertise and position itself to make significant strides from the beginning of a contingency, when the most important decisions are considered and made. Between contingencies, the modified civilian response corps outlined below could cost as little as $2 million per year, compared to tens of millions of dollars annually for the previous iteration of the CRC.

At a minimum, a civilian response corps should have these three components, which have been modified to account for the challenges faced by earlier efforts:

• An active corps of government professionals who focus on stabilization. State/Bureau of Conflict and Stabilization Operations and USAID/Office of Transition Initiatives staff should serve as the active component of a civilian response corps, deployable within a week. In contrast to previous efforts, participation in this active corps would be limited to State and USAID, which have the expertise and equities to justify their role. Notably, there would be minimal cost associated with this active component.
• A standby corps of government professionals with broader but relevant stabilization experience. If the contingency is large and/or many of the active corps members cannot be withdrawn from their current assignments, members of the standby corps should be activated. These officials would be pulled from the cleared-personnel pools at State and USAID, have knowledge of and experience working in conflict-affected environments, and be deployable within 90 days. The target number should be 300–500 people with diverse skillsets, including contract management, program management, and political engagement. State and USAID staff should be encouraged to participate with career-enhancing benefits and financial incentives. If activated to deploy, participants in the standby corps would be compelled to do so. Agencies and departments that lose staff to fill contingency civilian slots would be compensated for temporary backfills, under the assumption that supplemental congressional funding would be forthcoming in a contingency operation.
• A database of former State and USAID employees and other civilians with relevant experience. Depending on the size and nature of future contingencies, it is possible that the active and standby components would be insufficient in number of personnel or diversity of skillsets. In that case, it may be necessary to hire another wave of temporary employees—such as the “3161s” at State and the “Foreign Service Limiteds” and “Personal Services Contractors” at USAID. To prepare for such a situation, State and USAID should maintain profiles and contact information for their junior-, mid-, and executive-level professionals as they leave government service, in order to easily invite them back in the event of a contingency. Having an easy way to locate, recruit, and hire qualified individuals in the event of a contingency would allow State and USAID to avoid much of the chaos of the civilian surge in 2010, at minimal cost. This database should also be open to civilians without government experience but still relevant stabilization experience to ensure all capabilities and skillsets are available in this database.

In 2005, U.S. civilian agencies established a civilian response corps, which Congress authorized in 2008. However, by 2011 the CRC had lost the political and financial support necessary to continue. In the process, the CRC was discontinued, which inhibited the U.S. government’s effort to organize and deploy civilian stabilization experts from multiple civilian agencies.

SIGAR recommends that Congress revive the CRC with the necessary modifications outlined in the executive branch recommendations above . Doing so would ensure this important mechanism is fiscally sustainable and positions civilian expertise where and when it is most needed for future stabilization missions.

Specifically, Congress should provide State and USAID funding to pay for the financial incentives for members of the standby component to enlist between contingencies, to pay for backfills when members of the standby corps are activated during contingencies, and to pay for State to build and manage a regularly updated database of former State and USAID employees with relevant experience in conflict-affected environments.

At the outset of a contingency operation to stabilize a country or region, Congress should require State to take the lead in developing and implementing an overall strategy with at least a 10–15 year plan that outlines its specific stabilization components, including geographic priorities, benchmarks, how the various stabilization and development programming phases will be sequenced, and under what specific conditions control of territory will be transitioned to the host nation. Congress should require State to report regularly on the status of the strategy’s implementation.

Shortly after State has written its stabilization strategy, USAID should write and provide to Congress a robust data collection and monitoring and evaluation plan outlining how the agency will determine whether stabilization programs are pushing cleared territory closer toward the benchmarks outlined in the strategy, and in ways that improve upon the problems outlined in this report. Congress should require USAID to report regularly on the status of the plan’s implementation.

Congress should use its oversight authority to scrutinize how U.S. funds are spent and to what effect. Representing U.S. taxpayers, Congress sets the tone for a contingency operation’s expected return on investment. If Congress focuses on outcomes, the agencies will have incentive to do the same. Specifically, Congress should inquire about the agencies’ progress on implementing this report’s executive branch recommended reforms, both before and during a contingency.