13

Assessments and Risks

Financial columnist Gillian Tett reported extensively on the 2008 financial crisis. The more she learned about organizational behavior, the more she wondered why smart and capable people working in modern institutions collectively act in ways that sometimes seem stupid. The answers inspired her groundbreaking book The Silo Effect. Nearly everywhere she looked, tunnel vision and tribalism appeared. “People were trapped inside their little specialist departments, social groups, teams, or pockets of knowledge.”1 Silos in big banks, for instance, often created perverse incentives that encouraged employees to do things that made sense at a micro level but look very foolish from a macro perspective.²

The silo effect undermined management’s ability to assess these problems and understand the risks. When viewed within silos, performance may be satisfactory, and dangers may be manageable. The macro perspective must deal with issues such as compounding risks. These occur when actions in one silo affect other silos. Simply presenting individual assessments of risk and performance from each silo to create an overall evaluation is flawed because it ignores the effects of interaction and masks compounding risks. The Union Bank of Switzerland made this error during the financial crisis.³ Robert W. Komer’s classic study Bureaucracy Does Its Thing outlines how on-the-ground turf battles, bureaucratic infighting, institutional inertia, and bureaucratic silos undermined US efforts in Vietnam.⁴ Such challenges persistently undermined the quality of the US government assessments in Afghanistan and Iraq, too.

The administration’s new strategy required an annual assessment, which offered the opportunity to assess the strategic direction of the war and the likelihood of success. This process began in October and lasted until December 2010. Violence hit record highs that year. The Afghan security forces were growing in size and capability but had serious leadership and corruption problems and were progressing more slowly than forecasted toward being able to operate independently. The Kabul Bank collapsed late that same year, after over $900 million had been looted by corrupt officials, warlords, and other power brokers. Mahmoud Karzai, the Afghan president’s brother, and Qaseem Fahim, the vice president’s son, were implicated.5 Afghanistan remained at the very top of the world’s most corrupt governments.⁶ These problems led Under Secretary of Defense for Policy Michèle A. Flournoy to conclude that the Afghan government was not winning the battle of legitimacy.⁷ Diplomatic efforts to induce Pakistan to “change its strategic calculus” and to turn on the Afghan Taliban, meanwhile, had yet to achieve any tangible results.⁸

Interagency battle lines formed quickly. The intelligence community assessed that the Taliban had strengthened; the military command countered that the increases in violence were due to ISAF taking the fight into more Taliban-controlled areas.⁹ The Defense Department insisted the counterinsurgency campaign was on track and simply needed more time to work.¹⁰ After all, the surge forces had only been entirely on the ground for a few months. Gates famously noted, “The sense of progress among those closest to the fight is palpable.”¹¹ Holbrooke, who died suddenly of a heart attack during the process, was convinced that COIN (shorthand for counterinsurgency) had already failed and wanted a significant push on reconciliation. Members of the White House staff, skeptical of the surge in the first place, were convinced the military campaign was unlikely to produce results. Some reportedly pushed for a quicker drawdown.¹² The Departments of Defense and State both accused the White House staff of being policy advocates rather than honest brokers and of placing the highest weight on the most cynical interpretation of events.¹³

Confirmation bias was damaging the objectivity of assessments. Individual agencies and the White House were making selective use of information to bolster their cases. The intelligence community assessments about the strength of the Taliban, endemic corruption in the Afghan government, and sanctuaries in Pakistan competed with optimistic claims of progress from the field.¹⁴ The conflicting assessments had plenty of facts on their side. What the NSC did not adequately consider was that the positive results that were happening on the ground were not nearly sufficient to alter the factors that were undermining strategic success. A dispassionate view of the evidence suggested that a successful transition in 2014 was highly unlikely. Interagency disagreement and an inability to develop and assess strategically relevant metrics, however, reinforced a bias toward maintaining the status quo.

Overall, the contentious 2010 Afghanistan-Pakistan Annual Review refined the war aims and specified five lines of effort: a civil-military campaign to degrade the Taliban and build Afghan capacity; strategic partnership; transition to full Afghan sovereignty (security, economic, political); regional diplomacy; and reconciliation.15 The lines of effort were coequal and unprioritized but assumed to be mutually reinforcing toward a successful outcome. The updated approach did not address the unity of effort problems, so each agency continued to optimize its silo. The NSC handled disagreements by watering down issues toward consensus.¹⁶ Reconciliation became a friction point, as we will discuss in the next part.

The United States had a final opportunity at the end of 2011 to assess whether the transition strategy remained viable, the last chance to alter course before the 2012 US presidential elections and the troop drawdowns in 2013. Administration officials and the military command in Kabul continued highlighting progress while noting that government corruption and insurgent safe havens in Pakistan were critical risks to success. At no point, however, did any senior official testify that these risks were insurmountable or that success required a significant course correction.¹⁷ The October 2011 semiannual report from the Department of Defense to Congress, for instance, summarized the situation in optimistic terms:

[T]he International Security Assistance Force (ISAF) and its Afghan partners have made important security gains, reversing violence trends in much of the country (except along the border with Pakistan), and beginning transition to Afghan security lead in seven areas. . . . Although security continues to improve, the insurgency’s safe havens in Pakistan, as well as the limited capacity of the Afghan government, remain the biggest risks to the process of turning security gains into a durable, stable Afghanistan. The insurgency remains resilient, benefitting from safe havens inside Pakistan. . . . Nevertheless, sustained progress has provided increased security and stability for the Afghan population and enabled the beginning of transition in July of security responsibilities to Afghan forces in seven areas, comprising 25 percent of the Afghan population.¹⁸

The military believed its campaign plan would be successful in its specified tasks. However, the military component, although dwarfing other agencies in resources, comprised only a small part of the administration’s five lines of effort. The command was responsible for the military portion of the civil-military campaign to degrade the Taliban and build Afghan capacity (in this case, the ANSF). The governance and corruption problems, which belonged to State, were outside their authorities.19 For strategic partnerships, the military played a supporting role in negotiations with the Afghan government. On transition, the military oversaw the security transition, but not the political and economic ones. The military likewise played only supporting roles in regional diplomacy and reconciliation. Chairman of the Joint Chiefs of Staff Admiral Michael Mullen’s congressional testimony in September 2011 captures the matter: “The military component of our strategy, to the extent it can be separated from the strategy as a whole, is meeting our objectives. Afghan and ISAF forces have wrested the initiative and the momentum from the Taliban in several key areas. The number of insurgent-initiated attacks has for several months been the same or lower than it was at the same time last year. And we are on a pace and even slightly ahead of our end strength goals for the Afghan national security forces.”²⁰

Senator Carl Levin, the chairman of the Senate Armed Services Committee, asked Mullen if the effort was on course to meet the Obama timetable. “As far as I can see, yes, sir . . . while the risk is up, I think it’s manageable and that there’s no question that we can get there and sustain the military success and the military component of the campaign.”²¹

America’s bureaucratic way of war encouraged military officials to stay in their bureaucratic lane and focus on military progress and risks.²² Officials from other agencies reported metrics from their silos. What no one questioned was whether a successful military campaign could still result in strategic failure. After all, the military could not make a dent in the Taliban’s sustainability and the government legitimacy problems.²³ No one addressed those damaging cross-cutting issues, because no person below the president was accountable for strategic success and thus forced to confront them.

Despite the security gains, the insurgency remained sustainable. Violence levels in the second half of 2011 were indeed lower than in the second half of 2010, but characterizing that as a trend was misleading, as 2011 showed a higher level of violence in the first half of the year (see figure 6). The 2010 parliamentary elections counted for much of the spike in attacks that summer, as the Taliban aimed to discredit the election and candidates used violence to suppress voting in areas where people supported their opponents, so the year-on-year comparison was misleading. Compared to 2009, the number of violent incidents in 2011 was three times higher. The military was supposed to hand over a residual insurgency to the ANSF by 2014, but there were no standards that defined what residual meant or even if insurgent strength in 2009 was the benchmark.24

Figure 6. Enemy-Initiated Attacks, Afghanistan, 2009–2013

Source: US Department of Defense, Report on Progress Toward Security and Stability in Afghanistan, July 2013, A-3 (circle and annotation added).

The military offered no reason in its 2011 semiannual reports to Congress, or in future assessments or testimonies, to be overly concerned whether the ANSF could handle the insurgency by the end of 2014. “The ANSF are on track to assume full security responsibility by the end of 2014,” the Department of Defense assessed, “after successfully securing the presidential and provincial council elections and performing well during the fighting season.”²⁵ Although the Afghan security forces looked strong on paper, corruption and poor leadership were sapping their strength. Patron-client problems such as predation, sale of fuel, food, and equipment on the black market, and ghost soldiers damaged readiness faster than the coalition could build it. “How long would you stand and fight if your commander is stealing your food and equipment?” asked a former Afghan military officer.²⁶ The Afghan local police, despite some positive examples, was particularly corrupt and predatory.²⁷ The fielding of western systems and equipment, meanwhile, was increasing ANSF dependency on western advisors. A June 2011 Senate Foreign Relations Committee report warned of Afghanistan’s growing dependence and corruption.28 ANSF readiness to assume security responsibility was eroding even as their numbers and resources were increasing.

Adding to the problems, the insurgent sanctuary in Pakistan remained intact. The efforts to build a strategic partnership with Pakistan that would change the latter’s strategic calculus were ineffective. In May 2011, Obama approved the raid into Abbottabad, Pakistan, that killed al Qaeda leader Osama bin Laden. Pakistan was outraged by the breach of its sovereignty.²⁹ A month earlier, a US contractor had killed two Pakistani citizens in Lahore and was detained by Pakistani authorities despite American objections about diplomatic immunity.³⁰ A final blow came in a border incident in which US aircraft killed 25 Pakistan Frontier Corps soldiers who had fired on a nearby ISAF-ANSF patrol.³¹ At that point, Pakistan cut the ISAF logistics line leading from Karachi to Afghanistan, forcing the coalition to move supplies through Russia and Central Asia instead. In his final testimony to the Senate Armed Service Committee, Admiral Michael Mullen, who worked diligently to build a productive relationship with Pakistan’s chief of army staff, General Kayani, decried Pakistan’s use of insurgents as a “veritable arm” of foreign policy.³² A frustrated Senator Carl Levin, the chairman of the Senate Armed Services Committee, reportedly asked the Pakistani prime minister, Yousaf Raza Gilani, why Pakistan had not condemned attacks on US forces by groups operating from Pakistani soil. Gilani did not answer him.³³

Finally, the Afghan government was not showing any evidence of reform. According to Transparency International’s Corruption Perceptions Index, Afghanistan was tied with Myanmar as the third most corrupt country in the world, behind North Korea and Somalia.³⁴ Karzai resisted efforts to address corruption, faulting the United States for the problem.³⁵ Mullen carefully explained, “If we continue to draw down forces apace while such public and systemic corruption is left unchecked, I believe we risk leaving behind a government in which we cannot reasonably expect Afghans to have faith.”³⁶ Some Defense officials pressed hard for the US government to develop a credible anticorruption strategy, mainly Mullen’s senior advisor Sarah Chayes. Still, State was unable to put one together and feared the risks of breaking the relationship with Karzai. The Department of Defense eventually dropped the matter.³⁷

The White House was unwilling to address these problems or examine their implications. Part of the issue was bandwidth. Senior officials complained that getting any time on the national security advisor’s or president’s calendar was nearly impossible.38 By 2011, although deputies meetings were frequent, NSC meetings on Afghanistan were rare.³⁹ Events such as the Arab Spring, Iraq withdrawal, bin Laden raid, Libya intervention, and domestic challenges competed for NSC attention. When the war in Afghanistan did come up, it generally involved crisis management.⁴⁰ Little time or energy was available for considering highly complex strategic issues.

The reconciliation effort, as we will discuss in the next part, was in disarray by the fall of 2011 and was not available to serve as a credible alternative to transition. Believing that the “occupation narrative” was the driving force behind the Taliban’s ability to recruit, some senior White House and State officials clung to the argument that the drawdown would reduce the strength of the insurgency on its own.⁴¹ In this view, winding down the war would lower the threat of the Taliban and ease the ANSF’s job of securing the country. That Afghans had been fighting other Afghans from the Soviet withdrawal in 1988 to the Taliban overthrow in 2001 did not factor into this fanciful theory.

In some ways, Defense’s assurances about the military campaign plan were self-fulfilling for drawdown advocates in the White House. As long as the administration believed that ANSF development was on track to secure the country by the end of 2014, there was no reason to take the political risks of shifting priority to reconciliation, tackling corruption, or reexamining the withdrawal timeline. Confirmation bias, reinforced by evidence of progress within bureaucratic silos, kept the withdrawal timeline off-limits to serious examination.

The administration decided not to conduct a thorough review in 2011 as it did at the end of 2010. Although some Defense officials believed they could convince the president to “put more time on the clock” (delay the drawdown), Obama held firm. He announced in June 2011 that the withdrawal had begun and that the combat mission would end in 2014, at which time the Afghans would be responsible for their security.⁴² The advocates of drawdown had won the argument, and there was no reason, from their standpoint, to open an interagency process that would invariably try to rehash the issue.

In the event, the security situation declined as international forces handed over security responsibility to the ANSF in mid-2013.⁴³ Although the surge achieved temporary effects where foreign forces concentrated, the Afghan government proved incapable of winning the battle for legitimacy in contested areas.⁴⁴ The Afghan National Security Forces, ISAF asserted, was to be the “defeat mechanism” of the Taliban.⁴⁵ Problems, however, continued to materialize. Although rated by coalition advisors at the end of 2014 as capable of independent operations to secure the province, nearly the entire Helmand-based 215th Afghan Army Corps collapsed a year later.46 The 215th was the newest corps in the Afghan Army, but the post-2014 advisory mission did not cover them. The British refused to take on the task, and the Obama administration declined to raise its troop presence to make up the difference. An Afghan military officer reported finding a Taliban warehouse of abandoned corps equipment during a 2015 operation in Marjah.⁴⁷ By September of that year, the Taliban temporarily captured Kunduz, a provincial capital, and the first major city they had controlled since 2001.⁴⁸ Al Qaeda’s presence reportedly began to grow in 2015, and an Islamic State affiliate established a presence in the strategic border province of Nangarhar.⁴⁹ In early 2017, more than 30,000 suspected ghost soldiers were dropped from the rolls—nearly 10 percent of the ANSF’s reported strength.⁵⁰ Instead of fighting a residual Taliban insurgency, as the US policy had forecasted in 2009, the ANSF were facing the most muscular Taliban force since 2001 and a growing terrorist presence.⁵¹ ISAF handed off a strengthening insurgency to an ANSF that was severely compromised by corruption, poor leadership, and political pressures.