Conditions, Coalitions, & Pathways

Jennifer Widner Princeton University

In 1990, Steph Haggard published a book called Pathways from the Periphery. The agenda was to compare and contrast the routes to export-led growth in Korea, Taiwan, Singapore, Hong Kong—the NICs—versus Mexico and Brazil. In intellectual ambition, the project has much in common with the conversation at this meeting. Haggard’s focus was on the interplay between conditions and coalition building opportunities, as is ours today.

The work we do at Princeton’s Innovations for Successful Societies is related, although our focus is on institution building. In part, we want to understand the types of challenges reformers face in building more effective and accountable core government institutions and the kinds of coalitions they must assemble to build support for their work.

This note sketches some of our thinking about coalitions.

Coalition Building

Haggard begins his book with an epigram from Machiavelli’s The Prince.

It must be considered that there is nothing more difficult to carry out, nor more doubtful of success, nor more dangerous to handle, than to initiate a new order of things. For the reformer has enemies in all those who profit by the old order, and only lukewarm defenders in all those who would profit from the new order, this lukewarmness arising partly from fear of their adversaries, who have the laws in their favour; and partly from the incredulity of mankind, who do not truly believe in anything new until that have had actual experience of it. Thus it arises that on every opportunity fro attacking the reformer, his opponents do so with the zeal of partisans, the others only defend him half-heartedly, so that between them he runs great danger.

In Machiavelli’s time, the reformer needed to win acceptance—a will to move forward with changes and well as restraint from violent repudiation. Today’s reformers need these things, but they may also require a majority or even a super-majority at the ballot box.

To attain these outcomes—toleration, cooperation, votes—reformers enter into bargains with those who may disrupt or drag their feet or those who control vote blocs. These partners may be the heads of government institutions, political party leaders, the people who control interest groups such as unions or business, the big men who manage patronage networks, religious or ethnic authorities, crime bosses, armed factions, or civic organizations. Persuasion may play a role in the formation of these bargains, but so may payoffs or various forms of subtle coercion.

The degree of difficulty attached to coalition building depends on a variety of things. The elements of the coalition-builder’s calculus typically include:

Number of players and relative size: The costs of negotiating with many small groups are greater than the costs of negotiating with larger blocs. The size of the minimum winning coalition may be important in this calculus. If there are many groups but only two are needed to win legislative passage of a reform,

then the task may be more manageable than if the reformer needs the support of five or six groups to win the vote. Where the aim is not a majority vote share but restraint from violence (tolerance) or active support, even small groups can be disruptive and may have to have a place in the coalition.

Elite coherence: The sense of shared mission or status among the leaders of the groups whose support or restraint is important to the reformer may matter—and whether the reformer is on the side or outside of that group, or has a champion in that group. Shared status or a sense of shared mission reduces persuasion costs.

Ability of groups to bring followers along: In coalition building, the aim is to win the support of a group. Where the bargaining agent, the group representative, cannot reliably win the compliance of his or her base with agreements negotiated, the costs of effective coalition building rise. For example, during the South African transition one of the concerns was whether the ANC was losing the capacity to win the support of youth and others. If it lost the ability to persuade its followers to sign on to an agreement, then negotiations would have to include the leaders of many new groups representing a wider range of ideological perspectives.

Whether the basis of power of key coalition partners is patronage, identity, occupation, or something else (the distribution of preferences in the population): The basis of a coalition partner’s power shapes the currency that a reformer can use to forge a deal. If all that matters to constituents is a concrete private benefit now—patronage—then a reformer either has to provide such resources or provide these in a form that satisfies the demand but may reshape incentives over time. Coalition building requires different currencies in ethnically divided societies, in patronage-based systems, or in occupation-based systems. Depending on how the form the exchange assumes, a reformer may sharpen these bases or weaken them over time. A shrewd reformer may find inconspicuous strategies for weakening divisive claims.

Whether a player is armed: Where there is no government monopoly on the means of violence, the need to bring an armed fringe actor into the coalition may rise.

Costs of communication (information and coordination costs): All of the preceding factors affect the costs of communication, but other things may be important as well—whether language, cultural norms, past history of conflict makes. Or technology make communication more difficult or easier. The higher the costs of communication, the less likely it is that the reformer will successfully draw key partners into an agreement or the less likely that partners will win the support of their followers.

Whether deal has to be worked out and maintained in public: Where transparency is essential—usually where supporters do not trust their leaders—the risk of grandstanding increases. Compromises become harder to achieve.

Whether there is an internal enforcement mechanism to respect deal terms: If there is no punishment or cost for defecting, the coalition is not likely to endure over time in most settings. Loss of the benefits of partnership must have real meaning. Legal recourse and the ability to shape the flow of discretionary funds are among the things that shape the costs of defection.

Whether there is a realistic prospect of success and how large the fruits of cooperation will be. Reformers (or other leaders) form coalitions to accomplish specific ends. Paul Collier has noted in his book Wars, Guns and Votes: Democracy in Dangerous Places that the likelihood that a political party leader will be

able to improve public services is often low. Institutional change is hard to effect. If the candidate cannot deliver improvement, then why run the risks of holding higher office? The rent-collectors—the patronage politicians—are the people who are more likely to accept the risks of running for office than those who want to generate public goods. (ISS works mainly on providing information that can increase the probability of success.)

Whether elite perceptions are likely to change over time—and in which direction. Sometimes the reformers who have to build coalitions also have to contend with likely changes in preferences or perceptions over time. On the positive side, often economic growth shifts incentives and makes partners more willing to accept measures they may previously have rejected. Generational change within leadership can have similar consequences. However, it is also possible that differences will harden or that competition on ethnic or religious grounds will grow sharper as groups compete.

The coalition builder’s calculus may also include whether an alliance or pact contains rivals--on the theory that many potential partners don’t want to do something that would benefit a rival. Certainly in the business arena, companies sometimes refuse to participate if coalition membership brings a benefit to a rival or if it requires sharing of information. In the political world, coalitions that bring rivals together may be more unstable than others (or suffer immobilism), but the inclusion of the rival may also have advantages.

Dichotomies v. Pathways

Adrian Leftwich, my colleague Atul Kohli, and other scholars have tried to decipher why predatory coalitions emerge in some settings, while developmental coalitions appear in others. To be developmental, a coalition has to pursue public goods provision (as well as efficient provision of private benefits—divisible benefits—that promote growth), be able to take the long view (fairly stable winning coalition), and face pressure for performance.

Our work at ISS has some similar concerns but it makes two important points.

My first observation is that there is a spectrum along a predatory-developmental axis, and most coalitions fall in between. The conditions that impinge on the coalition builder’s calculus shape the kinds of challenges that a reform-minded politician or public servant will face. Because the calculus has many elements, there are clearly many “challenge contexts.” It might be more profitable, in analytical terms and in policy, to disaggregate these and then to define strategies and tactics—pathways—that can help produce more developmental outcomes (better public goods provision, for example) in each.

The second observation is that a good politician can break free of conditions others find binding. Social scientists must be cautious about assuming on overly deterministic relationship between conditions, strategic options, and outcomes. There is an element of art. Some people are better at setting a tone, remonstrating with others, etc. than others.

It is helpful to think about coalition building contexts in terms of decision trees—rather complicated decision trees, unfortunately. Each of the elements of the decision calculus constitutes a node. The pathways are the kinds of strategies that work within the contexts the decision trees define.

For example, based on this sort of categorization, one distinctive bargaining context would display high fragmentation among identity-based groups, little elite coherence, and substantial costs of communication or coordination because of linguistic diversity or limited infrastructure. This context resembles the conditions present in many clan-based or segmentary societies. The factions are often armed. Under these conditions, inequality in the distribution of benefits and sacrifices for longer-run gain rarely win acceptance. The best strategy for “good governance” might well be to keep the size and scope of the public sector very limited, to practice strict equality, and to keep programs limited to short-term projects that gradually build trust. Where identity is less important, these same kinds of conditions might present different strategic opportunities. The ISS case studies of Qena, Egypt and Somaliland match these types of bargaining contexts.

A developmental coalition is more likely to emerge in contexts with less fragmentation, where identity plays a more limited role as the basis of group formation and leaders command support from their followers, elites know one another socially, and the costs of communication are low because everyone speaks the same language or there are civic groups that translate and convey ideas. Under these conditions, it may be possible to win support for policies that impose short-term inequality or loss in the pursuit of longer-term gain for the society as a whole. However, if leaders believe they cannot make government deliver the benefits demanded—that is, if people generally believe prospects of success are low—the likelihood of securing a developmental outcome goes down.

Even where fragmentation is limited, if potential coalition partners are patronage-based, the costs of communication are high, and politicians feel they aren’t likely to get government to deliver what people want, it can be very difficult to form a developmental coalition. Under these conditions, it may make sense to exploit opportunities at the sub-national level—usually the city level—to try to nudge citizens away from patronage based politics by transforming incentives or setting examples, and to provide models of success that may give leaders a sense that they can make government deliver some of what followers want.

These partial examples simply suggest that there may be many different pathways for achieving developmental outcomes, depending on bargaining conditions and leadership skills. Elaborating these pathways may be one of the more valuable functions we can serve.