Special Interest Politics: How Interest Groups Influence Policy Outcomes in the US, Study notes of Communication

Download Special Interest Politics: How Interest Groups Influence Policy Outcomes in the US and more Study notes Communication in PDF only on Docsity! Journal of Economic Literature Vol. XL (December 2002) pp. 1221–1229 Review of Grossman and Helpman’s Special Interest Politics1 DAVID P. BARON2 1221 1. Introduction INTEREST GROUPS are ubiquitous in U.S. politics and in the political systems of most nations. In some countries, interest groups are closely aligned with political par- ties and exercise influence through those parties. In some countries, members repre- sent their interest groups by holding offices in government. In the United States, interest groups typically are not identified with par- ties—although their members may predomi- nately vote for one party—and do not repre- sent their members directly in government. Instead, U.S. interest groups generally serve their members from outside government. Expenditures on lobbying in the United States reported to the federal government totaled $1.45 billion in 1999, according to the Center for Responsive Politics (www.opensecrets.com). In addition, lobby- ing is conducted by many who are not re- quired to report their activities and expendi- tures: individuals, business executives, labor leaders, and association members. Interest groups also make campaign contributions. In the year-2000 election cycle, political action committees (PACs) contributed $259.8 mil- lion. Soft-money contributions by interest groups and individuals totaled nearly $500 million. Both lobbying and election cam- paigns are amply funded by interest groups. In Special Interest Politics, Gene Gross- man and Elhanan Helpman study how spe- cial interest groups influence political out- comes to benefit their members. The title of the book is fitting, but understates the con- tribution of the authors. They take interest groups seriously by considering a range of theories and supporting evidence about interest-group activity. Their book, however, is much more than a thoughtful study of the political strategies of special interest groups (SIGs). It provides insights into how to study interest-group politics and provides a set of methods for that study. Although the authors present a number of standard models, they also present much that is new. The reader gains a multitude of results, tools, models, and new research ideas. The result is an out- standing book full of insight, useful methods, and perspective. The book is intended for political econo- mists and graduate students, as well as econ- omists working in other fields who seek to add political forces to their work. The focus is positive throughout, although some of the analysis identifies Pareto improvements from political activities such as lobbying. The authors present “tools for analyzing the 1 Special Interest Politics. By Gene M. Grossman and Elhanan Helpman. Cambridge and London: MIT Press, 2001. Pp. xv, 364. ISBN 0-262-07230-0. 2 Stanford University Graduate School of Business. interaction between voters, interest groups, and politicians.” They seek to “shed light on the mechanisms by which SIG activities af- fect policy outcomes” and to “portray the key tensions and conflicts” rather than ex- plore how behavior varies with institutional details. The theory is made accessible to those outside the field through detailed dis- cussions of the incentives facing interest groups and officeholders. The approach is theoretical but with an emphasis on intuition and understanding rather than on formali- ties and generality. The focus on how and why SIGs have in- fluence allows the authors to avoid the stum- bling blocks that occupy the attention of many political scientists and political econo- mists who study the behavior of public of- ficeholders and candidates for those offices. For example, political economists who study electoral competition often work to escape the Downsian incentives that lead to the convergence of party positions. Grossman and Helpman focus instead on interest- group influence. When there is conver- gence, they ask to what extent the conver- gent positions reflect the interests of SIGs—i.e., how the political strategies of the interest groups influence the common posi- tion. The book is not oriented toward empirical applications or testing, but the models pro- vide a wealth of falsifiable predictions. Testing predictions of models with private information can be a challenge, since the in- formation of players is generally not observ- able. In many of the models analyzed by Grossman and Helpman, however, testing can focus on the relation between prefer- ences and policies. For example, when some members of SIGs are imperfectly informed about the policies of parties, the parties cater to the preferences of the SIG with the greater number of better-informed mem- bers, and poorly informed voters are ig- nored. Consequently, those SIGs that ex- pend resources informing and mobilizing their members should, other things equal, have more influence when there is uncer- tainty about party positions. 2. Content, Organization, and Style Grossman and Helpman focus not on the magnitude of the effect of interest groups but instead on how they influence govern- ment policies. Influence can focus on gov- ernment in office or on electing a govern- ment (Grossman and Helpman 1994). A complete theory would incorporate both arenas of influence, and the authors incorpo- rate both in their theory. They accomplish this in two stages. First, they focus on a single officeholder and one activity such as infor- mational lobbying or campaign contribu- tions. In this stage reduced form preferences are used, for example, to represent subse- quent electoral incentives. In the second stage they bring together information provi- sion or campaign contributions with a formal model of elections. The book is rigorously argued but the ar- guments are made without formal proposi- tions and proofs. Equilibria are identified and their properties examined through care- ful analysis intended to help the reader un- derstand the logic of equilibrium behavior. For example, to study costless lobbying, or cheap talk signaling, they begin with a two- state model, move to a three-state model, and then extend the logic to many states and a continuum. They also use a constructive approach for identifying equilibria. The typ- ical approach is to begin with the basic in- centives for the players, examine the inter- actions among their strategies, and verify that the strategies do or do not form an equi- librium. The players are assumed to be ra- tional in the sense that they maximize an ob- jective function, but they may be myopic or rationally ignorant. The approach through- out is game theoretic. For example, social choice theory is not used nor is Arrow’s Impossibility Theorem mentioned. In some political settings, institutional pro- cedures determine the sequence of moves of 1222 Journal of Economic Literature, Vol. XL (December 2002) each agent on the dimension on which the agent’s interests are aligned with the policy- maker. With costly signaling, more information can generally be conveyed because the cost of sending a message gives it a degree of credibility it otherwise would not have. When the cost is exogenous, i.e., it depends only on whether a message was sent, the equilibrium can distinguish only between two states, but when lobbying costs vary with the message sent, it is possible to have full revelation. Lobbying costs may also allow the SIG with a large bias to convey informa- tion. With both exogenous and endogenous costs the policymaker benefits from the lob- bying, but the SIG may or may not be better off compared to tying its hands. Grossman and Helpman bring together the lobbying models and the baseline elec- tion model by considering the SIG that can educate its members, and the public more broadly, using its information about which pliable policy is better for them. To simplify the model, both the SIG and the politicians are assumed to know the information, but the politicians are assumed to be unable to communicate their information to the pub- lic. SIG members prefer that the group leaders follow a narrow mandate focusing on the pliable policies of the parties. Early com- munication by SIGs directly affects the par- ties’ choices of pliable policies, and late communication affects those policies indi- rectly. With late communication the parties cater to the SIGs by attempting to become the party that they will subsequently prefer to support. The authors also consider en- dorsements; i.e., stating a preference for one party over another, and the education of their own members. Even when the SIG communicates only with its own members, the broader public may be able to infer some of the information by observing how the parties choose their pliable policies in response to the communication. The broader public must be fully rational and so- phisticated to make such an inference. 5. Campaign Contributions In the third part of the book, Grossman and Helpman focus on buying influence with campaign contributions as the instrument. The basic framework is that of agency in which an interest group offers a contribution schedule to the policymaker, who chooses a policy based on its preferences for policy and contributions. The analysis begins with a sin- gle policymaker and one SIG, with the SIG making a take-it-or-leave-it offer. The equi- librium is efficient in the sense that the pol- icy chosen maximizes a weighted average of the policy preferences of the policymaker and the SIG. The focus is on “compensating schedules” that reflect the SIG’s valuation of policy changes whenever contributions are positive.4 When the policymaker maximizes a weighted sum of the SIG’s interests and the public’s interests, the SIG contributes the least when the weights are extreme and the most when the weights are equal. When the policymaker has strong preferences for pub- lic welfare, it is very costly to influence its policy choice, and when the policymaker has strong preferences for the SIG’s interests, there is little need for influence. The authors recognize that the power in the relationship may not rest with the SIG, and also consider Nash bargaining between the two players. The basic model is applied to the allocation of public spending and to regulation and protection. When multiple SIGs attempt to influence a single policymaker, the power in the rela- tionship can shift to the policymaker. This is studied using a common agency model in which SIGs simultaneously choose fully compensating schedules. The resulting equilibria are jointly efficient and coalition proof in the sense that no subset of SIGs can gain from coordinated, self-enforcing actions. Applied to a trade problem, such equilibria have the property that when all Baron: Review of Grossman and Helpman’s Special Interest Politics 1225 4 Douglas Bernheim and Michael Whinston (1986) refer to such schedules as “truthful.” members of the public are represented by some SIG, free trade results with the SIGs competing away the potential gains from political action. The power then rests with the policymaker, leaving the SIGs worse off than if they had not taken political action. This power can be restrained if the policy- maker’s instruments are restricted as in the case of an international trade agreement. Many if not most government policies are chosen not by a single policymaker but by a legislature operating under some form of majority rule. Grossman and Helpman be- gin with vote-buying by a single SIG that also chooses the policy on which a legislature votes. Extending this model to multiple SIGs results in considerable complexity, since no pure strategy equilibrium exists. The authors thus study this setting in a model with an agenda-setter who is selected in advance of the SIG’s offers. The SIG of- fers a contribution schedule to the agenda- setter for both her policy proposal and her vote and subsequently buys the additional votes needed for passage of the proposal. The SIG may offer contributions to an agenda-setter with opposing policy prefer- ences, and its influence is generally less than if it set the agenda. The SIG’s contributions are generally smaller and the policy more fa- vorable the stronger are the agenda-setter’s preferences for contributions. With multiple SIGs the characterization of an equilibrium becomes more complicated, since the second-stage vote-buying equilibrium involves mixed strategies. Grossman and Helpman are able to obtain a complete charac- terization of the equilibrium, however, when the SIGs use compensating contribution schedules. The authors conclude by bringing to- gether campaign contributions and the base- line elections model. Contributions affect the voting of impressionable voters who do not know the positions of the parties but re- spond to the campaign spending of the two parties. Parties thus have a derived demand for contributions. The other voters are strategic and know the positions and policies of the parties. The SIGs focus on the pliable policies and can have two motives for contri- butions. One is the influence motive of in- ducing the parties to choose more favorable policies. The other is to elect the party with the more favorable policies. The influence motive is stronger the more the parties de- mand contributions, and their demand is in- creasing in the fraction of impressionable voters and in the effectiveness of campaign spending. The contributions can also lead the parties to choose policies other than those that maximize the welfare of the aver- age voter. The two motives interact in the sense that the electoral motive is present only if the influence motive leads to contri- butions that result in different pliable poli- cies of the parties. SIGs can contribute to both parties to influence their positions but only to their favorite to influence the elec- tion outcome. With multiple SIGs the free- rider problem is present and at most one SIG contributes to each party. The electoral favorite is induced to cater more to the in- terest groups. If differentiable contribution schedules are used, the policies maximize a weighted sum of the SIGs’ welfare and that of the strategic voters. Multiple equilibria exist and pertain to which party wins the election, but if election expectations are posited, unique policies for each party are predicted. 6. Additional Perspectives In this section I identify additional per- spectives on interest-group politics with a fo- cus on richer detail, alternative perspectives on informational lobbying, and the potential significance of repetition and dynamics. Interest groups differ considerably in their objectives, membership, and activities. One addition to the study of interest groups would be a theory of their industrial organization. Self-interested groups such as labor unions and most firms serve the in- terests of their members or stakeholders, 1226 Journal of Economic Literature, Vol. XL (December 2002) respectively, whereas advocacy groups serve the interests of others. Activists might be distinguished from advocacy groups by their actions, which often involve confrontation in public arenas. Grossman and Helpman acknowledge these differences, but instead focus on influence activities and the institu- tional arenas to which those activities are di- rected. Knowing more about the interest groups would be useful for explaining which influence activities are used by which groups in which settings. The authors also do not consider how SIGs are formed. In some cases this is straightforward. Common interests provide a natural incentive to participate in SIGs, and the appropriable benefits may be suffi- cient to overcome free-rider problems. When the appropriable benefits are small relative to the cost of participating, SIGs may provide selective benefits to members to overcome the free-rider problem. To re- cruit members and raise contributions, some interest groups use high-profile strategies directed toward the news media as a means of obtaining leverage in communicating with the public. In this sense the SIG may be ad- vertising itself. The need to raise funds to support political activities is crucial for most interest groups that recruit members from among the public, which suggests that SIGs could be modeled as having a budget con- straint and a revenue-generating activity. Grossman and Helpman do not study coalitions among SIGs, but their models provide a starting point for such an analysis. On the one hand, when cheap-talk lobbying is the instrument of influence, SIGs located on the same side of the policymaker are un- able to communicate more information than what the most moderate SIG can communi- cate. In such a case, the incentives to form a coalition may be weak. On the other hand, when contributions are the instrument of in- fluence, SIGs with aligned policy interests may have an incentive to form a coalition. In a common agency model where all SIGs use compensating contribution schedules, the equilibrium is coalition proof, so no subset of interest groups can coordinate, without commitment or enforcement, their actions to their benefit. This suggests that in com- mon agency settings the formation of coali- tions turns on the ability to commit to coor- dinated actions and to enforce that commitment through direct punishments or the exclusion from future benefits. Some form of dynamics or repeated play then may be necessary to explain coalition formation. This suggests that coordination by SIGs could affect the relative power of SIGs and government in a common agency setting. As Grossman and Helpman indicate, in- terest groups may direct their communica- tion at the public to affect the election of government. In doing so, interest groups may also be attempting to motivate the pub- lic to take other influence activities. This could involve protests and demonstrations, direct communication with elected officials, and other grassroots political activity. It would be useful to unpack the preferences of government and the strategies of SIGs to accommodate such grassroots political action. Grassroot actions can be directed not only at government but at private parties with the objective of causing them to change their policies. Activist groups, for example, cam- paign to force firms to improve their human rights policies and their practices, and en- vironmental groups work to induce firms to reduce emissions below those allowed by regulations. This private politics (Baron 2002) has become a major activity of some interest groups, and the methods presented by Grossman and Helpman can be applied directly to private politics with bargaining rather than elections or legislatures resolv- ing the conflict of interests. Although institutional details are not a fo- cus of the book, it is useful to ask where and how such details might matter for the study of interest-group politics. In parliamentary systems with proportional representation electoral systems, coalition governments are Baron: Review of Grossman and Helpman’s Special Interest Politics 1227