The Predictioneer’s Game: Using the Logic of Brazen Self-Interest to See and Shape the Future

by Bruce Bueno De Mesquita

  With these observations in hand, my students began to think about how they might go beyond predicting developments to trying to shape them (or at least simulate doing so). And so they initiated a search for a strategy that might get Pakistan’s leaders to make a more serious effort to rein in the militants. They looked at the possibility of trading aid dollars for policy concessions. Seeing that the Pakistanis want more money and the United States wants greater efforts to track down militants, they wondered whether an aid-for-pursuit deal might not improve the situation from the American and the Pakistani perspectives.

  Using foreign aid to secure policy compliance is a time-honored use of such funds even if, at the assumed $700 million in economic aid, it did not seem to be working in Pakistan. Looking at their assessments, my students could see why Pakistan’s leaders were not aggressively pursuing militants despite the then U.S. aid program for Pakistan. They could see that the leadership (Musharraf, the PPP, and the PML-N) expected to take too much political heat from al-Qaeda and the Taliban for it to be in their interest at the prevailing foreign aid level. And so my students set out to analyze how that might change if the United States gave significantly more aid than their analysis indicated was going to be the case. Figures 10.4A and 10.4B show the same projections through the end of President Bush’s term, but then they diverge. Figure 10.4A continues to forecast the relative influence of the U.S. government and the militants in Pakistan through the end of 2009 if President Obama follows the foreign aid course pursued by the Bush administration. Figure 10.4B assumes that Obama follows the course recommended by my students (and also by his now vice president, Joseph Biden), equal essentially to doubling U.S. aid.

  As is evident from the figures, continuing the current aid policy is a losing proposition. With 2008 aid levels, the United States maintains a small power advantage over the militants during Obama’s first year in office. That advantage virtually disappears by the start of 2010. The picture is entirely different if the United States and Pakistan strike a deal that trades dollars for aggressive pursuit of al-Qaeda and the Taliban. Rather than coming out of the chute looking to make a deal with the militants, the National Assembly leaders confront the Taliban and al-Qaeda. They impose heavy political and material costs on them and bear heavy costs in return. With the Pakistani government motivated by a doubling of aid dollars, the Obama administration increases U.S. clout in Pakistan at the direct expense of the militant groups (including al-Qaeda, the Taliban, and elements sympathetic to them within Pakistan’s intelligence service, the ISI). According to the model’s logic and evidence, we can fundamentally change the lay of the land in Pakistan, but to do so, we need to be responsive to the interest Zardari’s government has in getting its hands on more money. They won’t take the heat against the militants without it. No doubt some of that money will be stolen by corrupt officials, but that’s the point. They will want to continue the flow of dollars, and the only way they’ll succeed at that is by helping the United States against al-Qaeda and the Taliban.

  It is evident from my students’ analysis that a promise of greater efforts to go after militants will accompany an increase in U.S. aid. This, however, leaves two questions to be answered: How did we arrive at the idea that doubling aid is the optimal aid-for-pursuit deal, and will each side to such a deal follow through and show that they are really committed to it? My students answered the first question but didn’t have enough time to analyze the second, so I will do so here. But first, how do we work out what the optimal aid package looks like?

  FIG. 10.4A & B. Reining in Pakistan’s Militants by Doubling Foreign Aid

  Back in Game Theory 101, I introduced a way to look at how players resolve trade-offs across issues. In that chapter we were examining the notion of the national interest and saw that there were lots of ways to put a winning electoral coalition together, some of which supported freer trade and others fairer trade, some of which endorsed increased defense spending and others reduced spending. Now, using my students’ analysis, we can build on their solutions to the game and apply that methodology—known as win-sets—to their simulated results to evaluate the maximum pursuit that the United States can get from the Pakistani government and the cost in aid dollars to get that maximum pursuit.

  Figure 10.5 plots the preferred policies on aid and pursuit of militants held by Pakistan’s leadership and the average of the preferred foreign aid package supported by Congress and the president (and their shared view on pursuit of militants) around July 2008 against the predicted status quo on these two dimensions at that time. July 2008—roughly when Musharraf was predicted to be deposed and a new, PPP-dominated regime was expected to take over—is chosen as a prominent early opportunity to strike a deal that trades aid dollars for pursuit of the militants. The solid gray petal-shaped area in the figure shows the range of aid packages and levels of pursuit that are improvements over the status quo from the U.S. government’s and the Pakistani government’s perspectives. Different points within this gray area show different mutually acceptable trades between aid money and efforts to take down the militants. The optimal deal is in the top right corner of the petal. At that point the United States extracts the maximum effort by Pakistan’s leaders against the militants, and the Pakistanis in turn extract the maximum amount of money they can get. More money buys no additional commitment to go after the militants, and greater effort against the militants extracts no additional dollars. That is true because policies outside the petal are not mutually beneficial relative to the status quo, since policies outside the petal are farther from what one or the other set of players wants relative to the fallback position that is the status quo.

  So what is the optimal deal? The horizontal arrow in figure 10.5 shows the amount of foreign aid that secures the greatest effort to pursue militants by Pakistan’s leaders. That amount is $1.5 billion for 2009. The vertical arrow identifies the maximum effort in pursuing militants that secures $1.5 billion in U.S. aid. That level of pursuit is equivalent to about 80 on the pursuit scale. That is, my students found that for $1.5 billion in aid it is very likely that Pakistan’s national assembly, its president, and other important players would really go after the terrorist threat emanating from their country. That is a bit more than double the aid estimated for 2008 and many times the amount advocated by the president or Congress at the outset of the new president’s term (according to the model-based analysis). It requires a marked change in U.S. policy.

  FIG. 10.5. Aid Dollars Can Buy Great Pursuit of Militants

  What would it buy? Put in terms of the issue scale, if the Pakistani leaders get $1.5 billion in U. S. aid, their pursuit of militants will be far above the status quo score of 40 as of January 2008, but it will not equal the intense pursuit identified with a score of 100. That was the level the Bush administration and Congress wanted and surely is the level any American president will desire. So we won’t get 100, but we certainly will get a lot more than what is projected without this infusion of money. But please don’t get me, my students, or the analysis wrong: the Pakistani government is unlikely to completely quash the terrorist threat just for money. They are no fools. They know that the money will dry up if al-Qaeda and the Taliban are destroyed. So for money they will rein the threat in and reduce it (that’s 80 on the scale) but not utterly destroy it (that would be 100 on the scale). For their own political survival they will do whatever it takes. They will try to wipe out the militants if that is their best political path, but if making a deal with the hard-liners looks politically best for Pakistan’s leaders, then that is what they will do. A billion and a half dollars would go a long way to convincing them that they are better off going after the militants, insurgents, and terrorists than accommodating them.

  Of course, if an aid-for-pursuit deal is struck, each side will have to feel confident that the other side will not renege. Pakistan’s leaders must believe that aid dollars will continue to flow, and Congress must believe that, having received the money, Pakist
an’s leaders will not turn around and still make a deal with the militants. The latter may be of especially great concern because even with an aid deal, al-Qaeda’s clout, although diminished, still continues to be substantial, as projected in figure 10.4B. What the model shows is that the leadership in the national assembly will sustain pursuit of militants at around 75—80 on the scale, round after round after round. With the aid deal buying so much greater an effort to combat terrorists within Pakistan, Congress and the president are expected to remain steadfast in their commitment to the proposed aid agreement. All parties to the deal show a real commitment to sustain it, each for their own political benefit.

  Without the huge infusion of aid funds from the United States, I’m afraid the projected future is that Pakistan’s leaders will make a deal with the militants, who will become a legitimate part of policy making in Pakistan, or that country’s government will face another military coup. In all likelihood, American interests will decline and be thwarted. Too bad that this is what’s likely to happen. A billion five is a cheap price to pay to stabilize a civil, secular Pakistani government.

  IRAN AND IRAQ: IS THERE A MARRIAGE MADE IN HEAVEN?

  The future my students projected for Pakistan back in 2008 was grim. The reality a year later is at least as grim. Looking back over their predictions—some about things that have now happened and others about events yet to come—it seems that their play of the predictioneer’s game has proven depressingly accurate. Sometimes it really would be better to get things wrong.

  Now, in the spring of 2009, we have another chance to test the game. Let’s grab that chance and see where it takes us. Just a hop, skip, and a jump away from Pakistan and all its troubles, Iran and Iraq are facing an uncertain future. That uncertainty provides a golden opportunity to dare again to be embarrassed. You see, as I write this in early April 2009, I am again teaching the course on solving foreign crises that was used in 2008 to predict Pakistan’s stability (or, more precisely, its instability). Two excellent students in my 2009 course have put my model to good use, looking into the future relations between Iran and Iraq. In doing so, they have uncovered some pretty important insights about likely political developments in each of those countries. Their starting point—Will Iran and Iraq forge a strategic partnership?—illuminates a debate at the heart of two entirely different perspectives on American foreign policy: Should the U.S. government keep troops in Iraq, as President Obama proposes to do, or should we pull out altogether? The predictioneer’s game can help answer that question.

  Let’s review a few critical facts before plunging into the analysis. Back before the November 2008 presidential election, candidate Barack Obama promised to withdraw American troops from Iraq within sixteen months. On February 27, 2009, ensconced in the White House, Obama stretched his pre-election withdrawal timetable a little bit, to August 2010. That did not seem to have elicited any great controversy either among “pro-war” Republicans or “anti-war” Democrats. But when he announced the timing of the U.S. withdrawal, he also declared his intention to keep fifty thousand American troops in Iraq. That’s no small, token force. It is, in fact, 36 percent of the total number of U.S. soldiers in Iraq at the time he announced his policy. Not too surprisingly, he was subjected to plenty of complaints within the ranks of the Democratic Party for moving too slowly on pulling U.S. forces out of Iraq altogether. Predictably, he also got scant praise in return from the Republicans. Politics is not a warm and cuddly business. Obama took additional heat because these combat-ready troops are slated to stay in Iraq at least until 2011, when a pre-existing agreement with the Iraqi government calls for a full withdrawal. Of course, the possibility remains that the 2011 deadline could be extended indefinitely.

  The decision President Obama made in February 2009 and the reality he will face in August 2010 may look alike, and they may not. Pressure within his own party and changing circumstances on the ground might result in a decision to keep far fewer troops in Iraq. But of course it is also possible that President Obama will stick to his guns (fifty thousand of them). I am not going to try to resolve here which he will do, but I am going to use the predictioneer’s game to resolve which he ought to do. The answer will not depend on my personal inclinations or those of my students; I certainly don’t know what they favor. I had barely given this question any thought myself before doing the analysis.

  Of course, this investigation touches on only a few aspects of the policy implications of keeping U.S. forces in Iraq or withdrawing them. Facets of American security not examined here may also be influenced by the U.S. decision to pull out of or stay in Iraq. For instance, the troop decision also might make a difference in which way Iran heads in its pursuit of a nuclear capability. But I do not tackle that issue here. I will just say that the prospects of resolving that country’s nuclear threat are sufficiently good (based on earlier analyses I have done on Iran) that I do not believe a continued, greatly reduced American military presence will materially tip the resolution of the nuclear issue one way or the other.1

  WHY MIGHT IRAN AND IRAQ WANT TO BE PARTNERS?

  Pulling American troops out of Iraq is predicated on the idea that by the summer of 2010 Iraq will be able to defend itself against internal and external threats to its security. The Iraqi leadership must, of course, be mindful of the giant white elephant on its border as well as the potential of resurgent insurgents at home. One way to cope with its giant neighbor, Iran, is to forge close ties between the two countries. With that possibility in mind, let’s think about the range of deals Iran’s Shi’ite theocracy might strike with Iraq’s secular but Shi’ite-dominated government. As we contemplate a possible Iraqi-Iranian partnership, we must keep in mind that relations between Sunni and Shia Muslims are often extremely fractious, and more so in countries like Iraq, where both groups make up a substantial segment of the population.

  Iraq’s population is divided roughly 65 percent to 35 percent between Shia and Sunni Muslims, and many followers of the two factions hate one another. That divide certainly was a major factor that gave rise to Iraq’s insurgencies and the U.S. creation of those CLCs we talked about in Game Theory 101. During the insurgency, many Shia residents in Sunni areas were driven from their homes, and sometimes murdered on sight, by local Sunni militias. Likewise, Sunni residents in Shia-dominated communities were driven out or murdered. Although things are calmer now and some people have returned to their homes, many have not and animosities linger just beneath the surface, ever ready to explode at the first sign of provocation.

  Unlike Iraq, Iran does not have much of a domestic Shia-Sunni problem. That’s not so surprising. After all, Sunni Muslims are in scarce supply in Iran. There are about ten Shia for every one Sunni in that country. That is, however, not to suggest that Iranians are warmly disposed or even indifferent to the Sunni branch of Islam. Iran has certainly had more than its share of contentious relations with Sunni-dominated governments in the Middle East and in the wider Islamic world. Most notably, Iran had terrible relations with Iraq during the long years in which the latter was run by Saddam Hussein. Iran and Iraq fought an eight-year war that killed more than a million people and saw the extensive use of chemical warfare. Few in Iraq or Iran have forgotten, and fewer still are likely to forgive, so building bridges between these two countries will not be an easy matter. Staying apart, however, carries its own considerable share of risks.

  The Shia-dominated Iraqi government headed by Prime Minister Nuri al-Maliki sees Iran as a potentially sympathetic and like-minded ally. In contrast, he and his closest Iraqi followers may consider their own Sunni brethren a threat to their regime and their vision for Iraq’s future. Maliki surely wants to shore up Iraq’s security and, according to the data my students assembled, he sees forming a strategic alliance with Iran as the way to do so once the U.S. reduces its military presence or pulls out altogether. His starting position on the partnering issue described in the table on the next page is at 80. That means getting Iran to guarante
e Iraq’s security. Such an assurance would provide a credible military threat from Iran against any anti-Shia rebellion in Iraq. That could be just the insurance the Maliki government needs.

  THE IRAN-IRAQ PARTNER’S GAME

  Position

  Meaning

  Detailed Implications for Iran-Iraq Relations

  100

  Full Strategic Partnership

  Free flow of arms and military technology; a mutual defense alliance; joint intelligence operations

  80

  Concentrated Partnership

  Restricted flow of arms and technology; some intelligence sharing; an alliance in which each guarantees to defend the other

  50

  Restrictive Partnership

  Limited arms flow; no technology transfer; no shared intelligence; each promises not to use force against the other

  20

  Minimal Partnership

  Considerable restrictions on arms flow; no signed alliance agreement at all

  0

  No Strategic Partnership

  No flow of arms or technology; the two governments reaffirm their commitment to the Algiers Accord

  2

  Putting such a partnership together, however, will not be easy. Besides the usual complexities behind any international negotiation, it is likely that the U.S. government will present stiff diplomatic opposition to such a move by Iraq. Besides pressure from Obama, we can be confident that those who represent Iraq’s Sunni interests will also strenuously oppose any deal with Iran. As for Iran, a deal with Iraq would advance Iran’s ambition to become the dominant regional power, but the Iranian government will have to ponder the risks of associating closely with a regime that could fall into Sunni hands. The partnership issue seems especially well suited to evaluating whether the United States is better off keeping some troops in Iraq or removing all of them. Iran, after all, is hardly the state Obama would like to see exercise real influence over Iraqi policy, and a partnership between the two countries could have exactly that consequence.