The Predictioneer’s Game: Using the Logic of Brazen Self-Interest to See and Shape the Future
Page 23
Okay, let’s see what they came up with, remembering that the first class began in January 2008 and had its last meeting on May 5, 2008, and that the second started in late January 2009 and ended in the first week of May 2009. Everything reported here was worked out during those months. No information has been updated or altered to take account of later developments. The students had no prior experience with my old forecasting model or my entirely different and more sophisticated new model. We met for two and a half hours each week in class. They made weekly presentations, got lots of feedback, and spent a fair amount of additional time with me in my office learning how to interpret the new model’s results. They also put in lots of additional time figuring out what questions to ask and how to frame them, assembling the data, and preparing their weekly presentations and final papers. Let’s have a look at what they found out.
PAKISTAN: WHERE HAVE ALL
THE SOLDIERS GONE?
The group that decided to work on Pakistan in 2008 was intrigued by three policy questions. They wanted to know how willing the Pakistani government was going to be to pursue militant groups operating in and around Pakistan, including al-Qaeda, the Pakistani Taliban, and the Afghan Taliban. They also wanted to investigate whether the Pakistani government would allow U.S. military forces to use Pakistani territory to launch efforts to track down militants. Finally, they wanted to forecast the level of future U.S. foreign aid to Pakistan and whether a higher or lower amount of aid was likely to change the Pakistani leadership’s approach to pursuing militants.
These are big questions that go to the heart of U.S. interests in Pakistan. While answering these questions, the students also uncovered answers to a bunch of other important and compelling issues.
By way of background, it is important to remember that when the students started their project, Pakistan was in the midst of a crisis. Benazir Bhutto, Pakistan’s former prime minister, had returned from exile in late 2007 as part of a deal she negotiated with Pakistani president Pervez Musharraf. She was expected to become the next prime minister following the general elections scheduled for January 8, 2008. Instead, she was assassinated on December 27, 2007.
The elections were postponed to February 18, 2008. Musharraf’s party was routed, while the parties of former prime ministers Bhutto and Nawaz Sharif, also recently returned from exile, won control of the national assembly (Pakistan’s parliament). Musharraf continued as president. Mrs. Bhutto’s husband, Asif Ali Zardari, became the new head of her party, the Pakistan Peoples Party (PPP), while the party of former prime minister Nawaz Sharif, the Pakistan Muslim League-Nawaz (PML-N), joined the PPP to form a coalition government.
Neither of the two victorious parties was a friend to Musharraf. He was in the awkward position of having to certify the new government, expecting that it was likely to impeach him as soon as it was certified. He had earlier fired the chief justice of Pakistan’s supreme court to prevent him from ruling on the legitimacy of Musharraf’s own reelection. The new government was expected to restore the chief justice and had declared its intention to depose Musharraf as soon as the PPP and PML-N came to power. It did not do so. Failing to get PPP support on this important issue, Sharif and the PML-N withdrew from the coalition on August 25, 2008. With pressure mounting from the United States to do something about the use of Pakistani territory as a base of operations for al-Qaeda and the Taliban, and with the Pakistani government itself deeply divided on how to move forward, the country was in turmoil and its future direction was extremely uncertain.
The situation seemed dire not only for Musharraf, but from the American perspective as well. For all of his limitations, Musharraf was an important ally in the war on terror. He had literally put his life on the line by siding with the United States against the Taliban government in Afghanistan after 9/11. By 2007, however, his support seemed to waver. He turned greater authority over the pursuit of militants to local tribal officials along the Afghan-Pakistani border, reducing the role of the Pakistani army. From the American perspective, this hurt the prospects for continued success against the terrorists. Musharraf argued that it would prove beneficial because the locals knew the situation on the ground infinitely better than any outsiders and they had the local clout to get things done. (My own view was that this was a move by Musharraf to extract more economic and military aid from the United States by threatening to allow the situation to deteriorate if the aid was not forthcoming. It’s important to note that my students knew nothing about my personal view.) Although very much his own man and hardly perfect in American eyes, Musharraf was nevertheless the United States’ best source of help in the effort to defeat al-Qaeda and the Taliban. The new government in parliament, in contrast, spoke openly of finding a way to negotiate with the groups identified by the United States government as terrorists.
So what did my students find out? Their analysis showed that the PPP would have even greater policy influence relative to Sharif’s PML-N than the PPP’s advantage in National Assembly seats implied. This was not particularly surprising, but then, if a model only produces the unexpected, we should be suspicious of it. Why wasn’t it surprising? To start with, Sharif did not personally compete in the February election. That left his party without strong leadership in the National Assembly. When he finally decided to run in a by-election, the courts ruled that he was not eligible because of his earlier conviction on corruption charges when he was prime minister. It was, in fact, his earlier corruption case that had sent him into exile in the first place. That aspect of my students’ analysis merely confirmed what any Pakistan watcher already knew. Sharif was not as popular as Bhutto, and neither was his political party as popular as hers. With her assassination the PPP gained even greater influence, riding the crest of an upsurge in sympathy for her, her political movement, and her vision for the future.
What was surprising, and distressing, was the pattern of evolving power among the national leadership that emerged from the analysis. While capturing the conventional wisdom about the relative power of the PPP and the PML-N, they also found a solid answer to a pressing question. Many Pakistan watchers speculated on whether the new leadership in the National Assembly would make a deal with hard-liners. None, as far as I know, had dared to quantify what that might mean in terms of the future distribution of political power in Pakistan and its implications for shaping policy.
The predictions my students made, based on game-theory logic and the data they amassed to seed the model, can be seen in the figures that follow. The first addresses Musharraf’s potential to survive the election’s outcome and the conditions that ultimately would lead to his ouster. At the time their study was done, I think it is safe to say, most people believed Musharraf was finished. A few speculated on whether the United States would somehow save him, but most thought he would be political history right after the February 2008 election. “Not so fast,” said my students’ results.
Figure 10.1, on the next page, tells the story of what could have been and of what was to be. If the two parties in the government, Zardari’s PPP and Sharif’s PML-N, had been willing to work together, then figure 10.1 shows us that Musharraf could indeed have been ousted in March or April 2008, just as the pundits expected. The government parties’ combined power—the heavy solid line in figure 10.1—overtakes Musharraf in the period between March and April. That would have been the opportunity to kick him out, as expected by most Pakistan watchers. But the model tells us that the PPP and PML-N were not willing to work together at that time. The model shows that Sharif believed throughout this period that he could pressure Zardari and his party into doing just what Sharif wanted. The model also tells us that Sharif was wrong. According to the model’s results, Zardari saw no reason to listen to Sharif since the PPP had substantially more clout than Sharif’s PML-N. As we know now, rather than work together, Sharif threatened to withdraw his support for the government in May 2008, because Zardari was unwilling to commit to deposing Musharraf.
FIG. 10.
1. How Long Could Musharraf Be Tolerated?
But figure 10.1 tells a more complete story than just that. We can also see that the model projected that Zardari’s PPP (the dashed line in the figure) on its own, without help from Sharif, would surpass the declining Musharraf in power by June or July 2008. At that point, the PPP didn’t need anyone’s help to dump Musharraf. They had the clout to do it on their own. (We now know that they in fact did push him out in August 2008 and that Zardari assumed Musharraf’s role as Pakistan’s president.)
So while the world’s media were counting Musharraf out in February, the students successfully forecast that the divide between the PML-N and the PPP would allow Musharraf to hang on for about six months past when they began their study. But even this is but a small part of the big emerging story played out in advance by modeling key Pakistani policy issues. Figure 10.1 compares the power of only three of many players in Pakistan’s political game. Let’s see what the picture looks like when we throw in the main potential threats to Pakistan’s civil, secular government. I have in mind al-Qaeda, the Pakistani and Afghani Taliban, and even Pakistan’s military, with its long history of coups against civilian governments.
Figure 10.2 tells an incredibly distressing story for any who hold out hope for stable democracy in Pakistan. Pakistan’s Taliban and their Afghan compatriots work together as one, so I present them as if they are one. Looked at this way, they are far and away the most powerful force within Pakistan. And al-Qaeda is next in line according to the model, at least after April 2008, when their power is projected to surpass the government’s. Al-Qaeda just continues to grow and grow. Together with the Taliban they constitute the emerging dominant source of political influence in Pakistan, with only outside influencers like the United States or the Europeans being possible counterweights. Remember, we are plotting power—political influence weighted by salience—based on information known (or at least estimated by my students) back in January 2008 and not after. Yet here is the headline from the New York Times lead story on June 30, 2008, months after the analysis was done: “Amid U.S. Policy Disputes, Qaeda Grows in Pakistan.” The story goes on to say “it is increasingly clear that the Bush administration will leave office with Al Qaeda having successfully relocated its base from Afghanistan to Pakistan’s tribal areas, where it has rebuilt much of its ability to attack from the region and broadcast its messages to militants across the world.” My students were able to foresee this troubling prospect half a year ahead of the New York Times. Maybe they could have done so even earlier; remember, they only assembled their data in January 2008, when the course, Solving Foreign Crises, began. They saw this result immediately.
FIG. 10.2. Who Will Have the Clout in Pakistan?
There is one other troubling feature to figure 10.2. Other parts of the model’s output tell us that al-Qaeda and the Taliban will try to negotiate an arrangement with the PPP and the PML-N. Sharif’s PML-N is modestly more open to such talks than is Zardari’s PPP. Both prefer to live with the existing status quo vis-à-vis the militant groups while trying to consolidate their own hold on power. In the meantime, the Pakistani military sees itself slowly but steadily losing influence. Such a circumstance raises the prospect that they will try to stem the tide against them by launching a coup to take control of the government. The optimal period for them to take such a step is projected to be between February 2009 and July 2009. Earlier than that they see no need, and later may be too late for them. Pakistan’s fragile democracy appears likely to be under assault from the militants who would establish a nondemocratic fundamentalist regime on one side, and from the army that would establish a military government on the other.
What does this mean about Pakistan’s contribution to the war on terror? Will they make a more vigorous effort to pursue militants and stamp them out, or will the Pakistani government succumb to the projected growing influence of al-Qaeda and the Taliban? I think you can guess the answer. But just in case you can’t, figure 10.3 tells that story.
The status quo commitment to go after al-Qaeda and the Taliban back when my students began their project was at 40 on their issue scale. A value of 40 meant some real efforts to contain the militants but falling well short of trying to stamp them out as the United States wanted. That was equivalent to a score of 100 on the scale. The status quo, with some erosion, was close to the policy predicted to hold, more or less, until the summer of 2008. A position of 0—al-Qaeda’s position (not shown)—meant “Do nothing against the militants.” With that in mind, let’s see what we predict for the future.
FIG. 10.3. Who Will Urge Pakistan’s Pursuit of Internal Militant Groups?
The dashed line and the dotted line in figure 10.3 show the predicted positions of Zardari’s PPP and Sharif’s PML-N, respectively. After June 2008, their approach is projected to be little more than rhetorical opposition to the militants with almost no serious commitment to go after them. Talk about a balloon bursting. The air just pours out of the antimilitant effort. That puts responsibility for going after the militants squarely in the United States’ corner.
Throughout the remainder of his term, President Bush is adamant (but ineffective) in his commitment to persuade the Pakistani government to go after the hard-liners. After the summer of 2008 even he pretty much gives up on this strategy. Instead, my students found that the U.S. approach will shift ground. The United States’ two-pronged strategy of clandestine American pursuit and open Pakistani pursuit of militants will be replaced by a much greater emphasis on the (perhaps clandestine) use of the American military directly within Pakistan. Even that commitment, the student projections indicate, will collapse shortly after the American presidential election. The new president is not likely to do much of anything about the rise of terrorist influence within Pakistan at least through the end of 2009, when the student projection ends. It seems that after a new president is inaugurated—of course, we now know that the president is Barack Obama, but back in the spring of 2008, neither my students nor the model made assumptions about the American election—the war on terror will not be effectively focused within Pakistan.
We just had a brief sketch of the major findings regarding Pakistan. There is much more detail that could be discussed about how and why these results arise, but the more interesting question is “What is to be done?” Recall that my students also studied U.S. foreign aid to Pakistan and its likely impact on Pakistani policy. They estimated the current level of U.S. economic assistance to Pakistan in fiscal 2008 as $700 million. That was an approximation, but probably a reasonable one. (The true amount is surely a mix of public and secret information.) They then looked at how congressional and presidential support for that number would change over time, taking into account domestic pressures and the pressures within Pakistan. Here is what they found.
The analysis shows that substantial domestic political pressure is likely to push for cuts in American aid to Pakistan. President Bush and the Democratic Congress are predicted to move apart at least through the summer of 2008 (remember, the data are from January of that year). Indeed, by early summer, the projection was that the president would be pressing for Congress to increase annual aid to Pakistan from $700 million to around $900 million to $1 billion. Did he? Bush proposed shifting an additional $230 million in counterterrorism funds during the summer of 2008. The model predicted that Congress would hold the line on aid during that same period, and they did. Congress has complained in actuality about the amount of aid the United States is giving to Pakistan, but more on that in a moment. After the summer of 2008 the analysis reports that while the president continues to advocate greater aid than Congress supports, the two begin to converge slowly. Both conclude that aid just isn’t buying the policy compliance the U.S. government wants.
Put bluntly, American foreign aid is supposed to pay the Pakistani government to go after the militants. It is failing. Indeed, by June 2008 the public discussion coming out of Congress alleged that Pakistan was misusing U.S. aid funds. Money wa
s being spent on items like air defense with Bush’s support even though al-Qaeda and its allies in Pakistan are not known to have any air capability. Air defense might be helpful for Pakistan against a threat from, say, India. So, while administration officials countered that the impact of aid on Pakistan’s antiterrorist efforts was being underestimated, Congress made the case that the money was being thrown down a rathole.
It is not hard to see that, as projected by my students’ analysis, President Obama will face real pressure to support less aid for Pakistan. My students worked through the implications of their assessment and they despaired. Their analysis convinced them that al-Qaeda and the Taliban were getting stronger and that U.S. reluctance to increase aid was more likely to reinforce that trend than reverse it. It was clear from their investigation that the United States cared deeply about getting Pakistan’s help in tracking down and neutralizing militants and terrorists operating within Pakistan’s borders. It was equally clear that Pakistan’s government leaders (the PPP, Sharif and his PML-N, and Musharraf’s backers in the military) wanted much more U.S. aid than they were receiving. They could see that the then current U.S. policy did not provide either a sufficient carrot or a painful enough stick to convince Pakistani leaders to put themselves at risk by going after the militants.