NATIONAL BUREAU OF ECONOMIC RESEARCH
NATIONAL BUREAU OF ECONOMIC RESEARCH

NBER Reporter: Summer 2000


Special Interest Groups and Economic Policy

Gene M. Grossman * and Elhanan Helpman


In the idealized democratic society, economic policy is determined by "one man, one vote." But in all real societies, special interest groups play an important role in the process that determines economic policy. Pressure groups represent relatively narrow interests, for example of peanut farmers, auto workers, or shareholders of firms that produce semiconductors. They also represent broader interests, such as those of retired workers, capital owners, and those with special concerns for the environment.

Hardly a day passes without the media reporting on the activities and influence of special interests. Discussions of campaign finance reform in the United States have focused attention on one important way in which interest groups seek to influence policies: interest groups are contributing ever-larger sums to political campaigns and political parties, apparently to encourage politicians to take positions favorable to their causes and to aid those who do so in their bids for election. Contributions by Political Action Committees to congressional candidates, which in 1975-6 totaled less than $23 million, exceeded $430 million in the 1995-6 electoral cycle. (1) Interest groups also engage in public display, such as the highly visible protests that surrounded the World Trade Organization (WTO) meetings in Seattle and the recent International Monetary Fund and World Bank meetings in Washington, DC. These displays are meant to demonstrate to politicians the strength of the groups' convictions and to educate the public about the policy issues. Less visible but still important are the everyday activities of the legion of lobbyists in Washington, Brussels, and other capitol cities. According to the Center for Responsive Politics, the number of registered lobbyists in Washington grew from 14,946 in 1997 to 20,512 in 1999. (2) These lobbyists spend a good portion of their days meeting with elected officials in order to share their alleged expertise and to persuade the policymakers about the worthiness of their causes.

In the early 1990s, we began to study the effects that special interests have on trade policy deliberations and outcomes. Trade is an area where interest groups have been especially visible. Groups representing workers and firms figured prominently in the protection afforded the U.S. steel, auto, textile, and footwear industries in the 1970s and 1980s. Agricultural interests were active and effective in pushing the Common Agriculture Policy of the European Community (now the European Union). Groups representing labor and environmentalists were vociferous in their opposition to the North American Free Trade Agreement, while many business groups lobbied in favor of that agreement. These same groups are joined by still others in the current debate about the future of the WTO.

Although an extensive literature on the political economy of trade policy existed by the time we became interested in the subject -- with important contributions from William Brock and Stephen Magee, Robert E. Baldwin, Jagdish Bhagwati, Anne O. Krueger, and Wolfgang Mayer, among others -- the prevailing approaches took shortcuts that seemed to obscure important relationships and to limit the purview of the theory and empirics. For example, the framework used often was ill suited to making predictions about the variation in rates of protection, despite the fact that much of the empirical work focused on exactly this sort of evidence.

Our early work on special interest politics focused on campaign contributions (of time and money) as the tool by which interest groups might influence the policy process. In our first paper, we develop an analytical approach that emphasizes the strategic interaction between interest groups and policymakers on the one hand and the strategic interaction among the interest groups on the other. (3) We suggest that trade policies can be viewed as objects "for sale," with the policymaker as seller and special interest groups as buyers. This assumes that the policymaker cares about general welfare and, in this sense, represents the interest of the voters. But, in addition, the policymaker covets political contributions, which can be used to finance a bid for re-election. Our approach endows the policymakers with an objective function that is a weighted sum of aggregate welfare and contributions, thereby rendering them as "common agents" of the special interests. Special interest groups are assumed to represent industry interests in an economic model in which those with (human or physical) capital invested in an industry stand to gain from protection or subsidization of their sector.

In the model, every organized interest group makes a bid for influence. These bids take the form of "contribution schedules" that link the amount that a group will contribute to the policymaker to the collection of trade policies that the policymaker adopts. We impose no restriction on the design of these offers, except that no group can contribute negatively. (4) Each group understands that the others also will be bidding for influence and that, ultimately, the policymaker will set trade policies to maximize a political objective function that includes aggregate welfare and aggregate contributions as arguments. The groups design their contributions to maximize the welfare of their members, assuming that other groups will be behaving similarly and that the policymaker is politically motivated.

Using this approach, we can derive a formula for the equilibrium structure of protection in a small country that takes international prices as given. The rate of protection to an industry is the product of two components: the first component, common to all industries in a polity, reflects the weight that the policymaker attaches to aggregate social welfare and the fraction of the population that is represented by an organized interest group. This can be regarded as a summary of the political environment. The second component varies by industry according to its economic characteristics. It implies that more protection will be provided to politically organized industries with a high ratio of output-to-imports and to those industries for which the demand for imports is relatively inelastic.

Our formula stands up well to empirical analysis. Pinelopi K. Goldberg and Giovanni Maggi, using measures of nontariff protection of U.S. industries as their independent variable, have estimated the relationship between the structure of protection and the determinants we identify. (5) They find that the model fits the data, and that other variables usually included in cross-industry regressions of rates of protection add little to the model's explanatory power. According to their estimates, U.S. policymakers appear to place a high weight on aggregate welfare, and most voters are represented by one special interest group or another.

Next, we apply our model to the interaction between large countries. (6) The governments of these countries might either compete or cooperate in setting their trade policies. Our framework captures a "two-level game," to borrow Robert Putnam's terminology. On one level, the interest groups compete for domestic influence. On another level, the governments compete or cooperate in their international relations.

When governments do not cooperate, as arguably was the case in the early postwar years, they set trade policies to promote their domestic special and general interests. Promoting the general interest dictates policies that better the terms of trade. Promoting special interests requires protection of politically powerful industries, as in "Protection for Sale." The outcome is a blend of these two, with each country protecting its industries with inelastic import demands and inelastic foreign export supplies.

Our analysis of the noncooperative equilibrium reveals a motive for international cooperation that does not rely on the benevolence of national governments. A noncooperative equilibrium is not efficient for the various policymakers, inasmuch as each government fails to take account of the political harm it causes the other by not granting market access to the other's export industries. In an international tax negotiation, such as those that have taken place under the auspices of the General Agreement on Tariffs and Trade (now WTO), each government can benefit politically by gaining market access for its politically powerful export industries while granting access to foreign countries whose domestic import-competing interests are not so strong. As before, political strength is determined by a combination of political factors (the weight the government places on social welfare and the fraction of the population that is represented by an organized interest group) and economic factors (the ratio of domestic output-to-trade and the elasticity of import demand or export supply). In fact, the tariff formula derived in "Protection for Sale" gives an exact measure of an industry's political strength. If a group representing an export industry is politically stronger than its counterpart in the importing country, then trade will be relatively unfettered in a politically negotiated trade agreement. If the import-competing interests are stronger, then high trade barriers will persist under an international agreement.

We also study the formation of free trade areas. (7) Interest groups line up for or against such agreements depending on whether they stand to gain or lose. The organized groups that will benefit from an agreement offer contributions to the policymaker that are paid if the agreement comes to pass. Those who will be harmed make their offers contingent on the opposite outcome. In a political-economic equilibrium, a free trade agreement is viable if and only if it garners sufficient political support from those who would gain in all member countries.

One important result from our analysis is that free trade agreements are most likely to be politically viable when they might be socially harmful. A free trade agreement will be socially harmful if it induces a great deal of trade diversion; that is, if the preferential treatment afforded to members induces the countries to import goods from their partners that could be purchased more cheaply from the outside. When trade diversion occurs, the exporting interests prosper, but the import-competing industry does not suffer. Thus, the prospect of trade diversion generates political support for the agreement from some special interests and little opposition from others. In contrast, social welfare gains require that new trade be created, but that would spell income losses for workers and firms in the import-competing industries. These groups then would be inclined to oppose the agreement. Thus, the political prospects for an agreement that would create trade are surely worse than for those that would mostly divert it.

Our work on the political economy of protection leads us to examine more carefully the relationship between voters, candidates, and special interest groups. (8) In our early analyses, we did not treat voters explicitly; they were only indirectly represented by an assumed concern of the policymaker for aggregate social welfare. In subsequent work, we introduce voters explicitly and allow them to choose between two slates of candidates. (9) These slates are offered by competing political parties each aiming to maximize their share of the vote. The parties have certain policy positions that they regard as inviable, but others on which they are willing to be more flexible. The parties might choose their positions on the latter set of issues either to appeal to knowledgeable voters or to cater to the special interests that are willing to contribute to their campaigns. Campaign funds are potentially valuable to the political parties because they can be used to purchase advertising that is effective in swaying impressionable voters.

We characterize an equilibrium of a multi-stage political game. First, the many interest groups bid for influence from one or both of the political parties. The bids take the form of contribution schedules that link campaign gifts to the positions adopted by the parties. Next, the parties choose their platforms for the issues on which they can be flexible. Finally, the voters cast their ballots, with the knowledgeable ones voting for the party that offers the most attractive package of policies and the impressionable ones influenced somewhat by the parties' campaign spending.

Several interesting conclusions emerge from this analysis. First, we find that interest groups typically contribute to both political parties. This is very much in keeping with the evidence for the United States and Israel. Second, the groups rarely give more than the minimum amount needed to exert their influence. Since the offers are used to push the parties toward positions that are unpopular with the average knowledgeable voter, the gift from any one group has no marginal effect on the recipient's electoral success. In this sense, the groups' contributions can be seen as motivated only by a desire for influence and not by a desire to help their favorite parties to win more seats. Third, the positions chosen by the parties maximize a weighted average of campaign contributions and the aggregate welfare of knowledgeable voters. Importantly, the voting environment determines the relevant weights. Parties place more weight on social welfare relative to campaign contributions as the fraction of knowledgeable voters increases, as campaign spending on attracting impressionable voters becomes less effective, and as the preferences of the knowledgeable voters become less dispersed in regard to the issues on which the parties' positions are fixed. Finally, we find that the party whose fixed positions are more popular among voters is more likely than its rival to cater to the special interests.

More recently, we have been studying how interest groups might use their specialized expertise as a tool to influence policy. An interest group is likely to know quite a bit about its particular areas of interest. For example, a group representing the firms in an industry is likely to be well informed about industry conditions, while one representing environmentalists may know a great deal about ecological sciences. Interest groups are a potential source of information for voters and policymakers alike. The groups can advise the voters about the wisdom of different candidates' positions and can advise the policymakers about what policy approaches would be desirable. But the recipients of the advice must be wary because the group has a stake in the outcome. Our research focuses on what types of statements and actions by interest groups will be credible and how these actions can affect the policy outcomes.

In one paper, we study the issuance of endorsements by special interest groups. (10) Endorsements may be seen as a means by which the leaders of interest groups can communicate with their rank-and-file members and with other voters. An endorsement indicates which of the candidates has a position that better serves the common interests of the group members. This message may be useful to the voters if they are imperfectly informed about the policy issues. There are circumstances in which the ability to issue an endorsement confers a benefit to a special interest group and circumstances in which it does not. If the issue is one of pure distribution, where anything the members gain comes at the expense of voters who are not members of the group, then the endorsement will have no effect on the candidates' positions or on the outcome of the election. Nonmembers will interpret the endorsement in the same way that members do and realize that the candidate who would better serve the group's interests would be harmful to their own. These voters will be less inclined than otherwise to vote for an endorsed candidate, so the candidates will be wary of being endorsed. But if the policy issue is not one of pure distribution, an endorsement can have real effects. The endorsed candidate will capture a greater share of the votes of interest group members than otherwise, while losing relatively fewer votes among nonmembers. In this event, the candidates will "compete for the endorsement" by adopting positions that cater to the interest group.

Our forthcoming book will discuss other ways that interest groups might use their knowledge of the issues to influence the policy process. (11) Among these are attempts at persuasion by lobbyists in one-on-one meetings with policymakers, public information campaigns aimed at the general voter, and resorting to costly displays such as those that took place in Seattle. Our book will also provide an integrated treatment of campaign contributions as a means of policy influence and will incorporate numerous examples of applications to specific policy questions.


1. See B. A. Loomis and A. J. Cigler, "Introduction: The Changing Nature of Interest Group Politics," in Interest Group Politics (5th edition), A. Cigler and B. Loomis, eds. Washington, D.C.: CQ Press, 1998.

2. See the summary of Influence, Inc., 1999 edition, at http://www.opensecrets.org/pubs/lobby98 (accessed May 16, 2000).

3. G. M. Grossman and E. Helpman, "Protection for Sale," American Economic Review, 84 (September 1994), pp. 833-50.

4. Actually, all we need is a lower bound on the size of contributions. So, we can allow for negative contributions and think of these as gifts promised to the incumbent policymaker's opponent.

5. P. K. Goldberg and G. Maggi, "Protection for Sale: An Empirical Investigation," NBER Working Paper No. 5942, February 1997.

6. G. M. Grossman and E. Helpman, "Trade Wars and Trade Talks," Journal of Political Economy, 103 (August 1995), pp. 675-708.

7. G. M. Grossman and E. Helpman, "The Politics of Free Trade Agreements," American Economic Review, 85 (September 1995), pp. 667-90.

8. Our published work on the politics of protection also includes an analysis of tariff jumping by multinational corporations in G. M. Grossman and E. Helpman, "Foreign Investment with Endogenous Protection," in The Political Economy of Trade Policy, R. Feenstra, G. Grossman, and D. Irwin, eds. Cambridge, Mass.: MIT Press, 1996; and a comparison of alternative theoretical approaches to the political economy of trade policy in E. Helpman, "Politics and Trade Policy," in Advances in Economics and Econometrics: Theory and Applications, D. Kreps and K. Wallis, eds. New York: Cambridge University Press, 1997.

9. G. M. Grossman and E. Helpman, "Electoral Competition and Special Interest Politics," Review of Economic Studies, 63 (April 1996), pp. 265-86.

10. G. M. Grossman and E. Helpman, "Competing for Endorsements," American Economic Review, 89, June 1999, pp. 501-24.

11. G. M. Grossman and E. Helpman, Special Interest Politics, Cambridge, MA: MIT Press, forthcoming.


* Grossman is an NBER Research Associate in the Programs on International Trade and Investment and International Finance and Macroeconomics and the Jacob Viner Professor of International Economics at Princeton University. Helpman is an NBER Research Associate in the Program on International Finance and Macroeconomics, a professor of Economics at Harvard University, and the Archie Sherman Professor of International Economic Relations at Tel Aviv University. His Profile appears later in this issue.

 
Publications
Activities
Meetings
NBER Videos
Data
People
About

Support
National Bureau of Economic Research, 1050 Massachusetts Ave., Cambridge, MA 02138; 617-868-3900; email: info@nber.org

Contact Us