Agency problems and interest representation: An empirical analysis of the costs of lobbying


Most recent work on the politics of interest representation assumes that lobbyists represent their clients’ interests with fidelity. We test the validity of this assumption. Relying on a principal-agent framework developed by Stephenson and Jackson, Kersh and Lowery and Marchetti, we first discuss the nature of agency problems in lobbying and the utility of several potential solutions for those problems. We next develop and test two sets of hypotheses on how agency problems might influence one form of lobbying behavior – the prices contract lobby firms charge their interest organization principals. The hypotheses are tested with a purposive sample of clients employing major contract lobby firms operating in Washington DC in 2012.

This is a preview of subscription content, log in to check access.

Figure 1
Figure 2
Figure 3
Figure 4


  1. 1.

    Instead, the unit of analysis we now use in studies of mobilization is typically the lobby organization itself, not its lobbyists (for example, Schlozman, 1984; Gray and Lowery, 1996). Similarly, studies of tactics and strategies now typically focus on issues as units of analysis (for example, Baumgartner et al, 2009).

  2. 2.

    Even Olson’s (1965) by-product theory blames any gap between the interests of members and their representatives on a lack of concern for lobbying on the part of the former given selective inducements for joining. In effect, by-product theory makes any gaps between the interests of lobbyists and their clients uninteresting. If there are further doubts in the literature about the fidelity of lobbyists, it is almost entirely focused on the legislator–lobbyist relationship where trust is thought to be vital (Kersh, 2002).

  3. 3.

    Such problems are not unique to lobbying. For example, Arrow’s (1963, p. 951), analysis of health care highlighted the importance of the ambiguity of outcome criteria.

  4. 4.

    But Arrow (1969) and Williamson (1993) have noted that social norms based on trust and professional ethics constitute an alternative by which to suppress agency problems. Such reliance, however, readily devolves to letting the market police the relationships between lobby agents and their organizational principals, a policing power that we have already seen is suspect (Lowery and Marchetti, 2012).

  5. 5.

    There are, however, several qualitative studies of lobbying that provide some support for expectation that agency problems bedevil lobbying (Kersh, 2000; Leech, 2013). But while useful in terms of establishing construct validity, neither provides a systematic test of agency hypotheses.

  6. 6.

    There is, however, a plausible alternative interpretation of this variable usefully suggested by one of the reviewers. That is, the presence of an in-house lobbying shop means that a principal may not need to pay a lobbying firm as a ‘general contractor’ to identify, design and implement large-scale lobbying strategies. The company’s in-house lobbyists generally may do at least some of the strategic design and organization. This means that when a principal does need to hire external lobbying firms, it may need to pay less overall, instead paying the contract lobbyists for only specific lobbying tasks. While the models presented here control for the total amount spent lobbying by principals, including both own spending and spending on agents, that control may not capture all of the spending done by in-house lobbying on overhead, salaries, planning and strategic planning. That is, a principal with its own in-house lobbying shop may not have to pay for the same breadth of services from a contract firm as when the principal uses a contract lobby firm to generate an entire lobbying effort. We will assess this alternative explanation later in the analysis.

  7. 7.

    Consumer sovereignty is also used in a normative sense. As Rothenberg (1962, pp. 269–270) noted, ‘In its normative sense consumers' sovereignty asserted that the performance of an economy should be evaluated in terms of the degree to which it fulfills the wants of consumers’.

  8. 8.

    The assumption of consumer sovereignty implies, for example, that if we knew what beer consumers wanted in terms of the cost, quality, and variety of beer, we would not need additional information about brewers per se to predict what those brewers produce and sell to consumers. This does not imply that producers would sell a uniform product. Rather, brewers might develop a rich ecology ranging from microbrewers to mass producers of Swill LiteTM (Hannan and Carroll, 1992). But it would be beer consumers who would sculpt the structure of this production ecology and determine how it operates. There would be no further independent information to be provided by observing the brewers themselves. And, if such information were useful in understanding the variety and quantities of beer that appeared on the market, this would constitute evidence of excessive market power on the part of producers.

  9. 9.

    Still, even Scitovsky (1962, p. 266) was in the end at least somewhat skeptical about whether the requirements of consumer sovereignty can be met.

  10. 10.

    Within the Senate's Lobbying Disclosure Act Database, we searched by filing year and client name. We recorded data for the clients' quarterly reports with each firm they hired. For some clients this included firms within the original list of the top 20 firms as well as smaller firms. Moreover, some clients lobbied on their own behalf. After eliminating the termination and registration reports, this resulted in a total of 722 observations. For each of the four quarters in 2012, we consolidated the multiple issue areas by firm and also multiple institutions lobbied (for example, Department of Defense, Senate, White House). For instance, a firm may have lobbied on a client’s behalf multiple times in a particular quarter on multiple items across multiple institutions.

  11. 11.

    When the models in table 1 were rerun on only the original sample, the results did not change in any meaningful way. Therefore, we report results for the amended sample so as to take advantage of the larger n.

  12. 12.

    Additional variables were coded but not used in the analysis: the level of mobilization, organizational structure and type of organization coded from the websites of the principals. In order to ensure inter-coder reliability in the coding of these variables, we double coded all of the principals and met to discuss any differences between the double entries. Where differences existed, the three coders made a reasoned decision about the final coding based on maintaining consistency across the firms. We also separately identified the three-digit NAIC code for each principal from These data were not used in the analysis because they proved to be largely redundant with the industry and issue codes.

  13. 13.

    The empirical analyses to follow work with the actual spending data.

  14. 14.

    These amounts, per se, are not especially large. Google, Lockheed, Comcast and the American Hospital Association each spent at least five times as much on lobbying. The difference is that the Blackstone Group and the Gila River Indian Community relied on only one and two lobbying firms, respectively, while the others employed 25, 26, 37 and 7 lobbying firms, respectively, and lobbied on their own behalf as well.

  15. 15.

    Measurement of this variable via a dummy is appropriate given that, theoretically, the importance of lobbying on a principal’s own behalf is not in terms of how much it is done but that it is done at all. Doing any lobbying of any kind is a signal that the principal retains the capacity to leave the lobby firm and work on its own. That said, the data in our sample indicate that those principals that do lobby on their own typically spend a great deal on it – on average, 65.13 per cent of their total lobby spending.

  16. 16.

    The extreme case of 37 firms is one principal – Comcast Corporation – which was included four times in our extended sample for lobbying via four of our top 19 lobby firms in addition to lobby on its own.

  17. 17.

    Somewhat surprisingly, however, the simple correlation between the amount spend with the top lobbying firm and the proportion of total expenditures allocated to that firm was negative and very weak (P=−0.03) even without considering our control for total lobbying expenditures.

  18. 18.

    In two of these cases, the principal filed a report but reported zero expenditures even though they employed lobby firms that reported spending a great deal. In these two cases, we coded the principal as not lobbying on their own behalf despite the filing of an independent lobby disclosure report.

  19. 19.

    With ‘other’ excluded as a reference category.

  20. 20.

    The two most extreme spending cases are included in the second and fourth highest average spending firms in Figure 4, which in part explains their high ranking in terms of spending.

  21. 21.

    The models were also, however, estimated with clustering on the client given that several were included in multiple dyads or employed several of lobby firms. The substantive results did not change. Indeed, they did not change in any meaningful way when using default OLS estimation of standard errors, although the precision of the estimates was weakened somewhat in several cases.

  22. 22.

    In Model 3, for example, the bank, defense, communication, finance, and telecommunication issue dummies were significant at the 0.10 level, as were the manufacturing, defense, service, medical, research and transportation industry dummies. Similarly, 10 of the 19 lobby firm dummies were significant at at least the 0.10 level in Model 3.

  23. 23.

    We noted earlier in footnote 6, however, that there was a plausible alternative account of the negative estimate for the dummy measuring whether the principal lobbied on its own. That is, our control for total lobby spending may not fully account for unreported overhead expenses of those who lobby on their own. This potentially unmeasured overhead spending would reduce the need to spend on lobby firms as agents. If true, this suggests that those who lobby on their own will spend less overall on agent lobbying even after controlling for total lobby spending. We assessed this alternative interpretation by assuming that such unmeasured overhead spending is proportional to the reported lobby spending by a principal; those who engaged in more reported lobbying would also have higher overhead expenses. If this assumption is true, then the alternative explanation would suggest that spending by the principal for their own direct lobbying should be negatively related to their total spending on lobbying agents. We test this explanation using negative binomial estimation with – given non-independent observations – clustering on the lobbying principal where the dependent variable was total lobby spending on agents (total lobbying spending minus the principal’s own spending). All of the independent variables were the same as those reported in Model 3 of Table 1 except that total lobby spending and its squared value were necessarily excluded so as to avoid a near tautology. The key independent variable – the principal’s own lobby spending – generated a negative estimate, which is consistent with the overhead interpretation of the own lobbying variable. However, the estimate was not discernibly different from zero (z=−0.26, P=0.794). This result suggests that how much the principal spends on reported direct lobbying – and, by assumption, how large their unobserved overhead expenditures are – has little bearing on their overall spending on agents. This implies that our controls for total lobby spending in the models reported in Table 1 are not failing to account for unmeasured overhead spending on the part those principals lobbying on their own in a manner that would undermine our interpretation of the results for the own lobbying variable. One additional implication of this finding is that our decision to measure own lobbying in our substantive models as a dummy was appropriate. It is not how much principals spend that matters. It is, instead, whether they preserve the option of going it alone as they bargain with agents.

  24. 24.

    In one further robustness check not reported in Table 2, we considered whether the inclusion of cases where lobbying might be limited to monitoring via a retainer with no active ‘lobbying’ might influence our results. To assess this possibility, Model 4 in Table 2 was reestimated dropping the 17 cases where spending by a principal on a given agent was less than $100 000. The results did not differ in a meaningful way.

  25. 25.

    There were, however, some minor differences in the results for several control variables. The lobbied Congress variable was significant in Models 2 and 3, as was number of issues lobbied in Model 3.

  26. 26.

    We thank an anonymous reviewer for this suggestion.

  27. 27.

    We must thank an anonymous reader for this point and this language.

  28. 28.

    On the basis of an OLS version of Model 3 in Table 1.

  29. 29.

    It may also be worth extending the specifications reported here by considering other determinants of spending that we could not readily address with the kinds of data used here. For example, one such determinant is coalition activity, which might reduce the need for lobby spending.


  1. Ainsworth, S. and Sened, I. (1993) The role of lobbyists: Entrepreneurs with two audiences. American Journal of Political Science 37 (3): 834–866.

    Article  Google Scholar 

  2. Alchian, A.A. and Demsetz, H. (1972) Production, information costs, and economic organization. American Economic Review 62 (5): 777–795.

    Google Scholar 

  3. Anderson, J., Newmark, A., Gray, V. and Lowery, D. (2004) Mayflies and old bulls: Demographic volatility and experience in state interest communities. State Politics and Policy Quarterly 4 (2): 140–160.

    Article  Google Scholar 

  4. Arrow, K.J. (1963) Uncertainty and the welfare economics of medical care. American Economic Review 53 (5): 941–973.

    Google Scholar 

  5. Arrow, K.J. (1969) The organization of economic activity: Issues pertinent to the choice of market versus non-market allocation. Washington DC: Paper Published by the Joint Economic Committee of Congress.

  6. Baum, K. (1995) Political control versus expertise: Congressional choices about administrative procedures. American Political Science Review 89 (1): 62–73.

    Article  Google Scholar 

  7. Baumgartner, F.R. and Leech, B.L. (2001) Interest niches and policy bandwagons. Journal of Politics 63 (4): 1191–1213.

    Article  Google Scholar 

  8. Baumgartner, F.R., Berry, J.M., Hojnacki, M., Kimball, D.C. and Leech, B.L. (2009) Lobbying and Policy Change. Chicago, IL: University of Chicago Press.

    Google Scholar 

  9. Berkhout, J. and Lowery, D. (2011) Short-term volatility in the EU interest community. Journal of European Public Policy 18 (1): 1–16.

    Article  Google Scholar 

  10. Bhuyan, S. and Lopez, R.A. (1997) Oligopoly in the food and tobacco industries. American Journal of Agricultural Economics 79 (3): 1035–1043.

    Article  Google Scholar 

  11. Coase, R.H. (1937) The nature of the firm. Economica 4 (16): 386–405.

    Article  Google Scholar 

  12. Dahl, R.A. (1982) Dilemmas of Pluralist Democracy. New Haven, CT: Yale University Press.

    Google Scholar 

  13. Demsetz, Harold (1988) The theory of the firm revisited. Journal of Law, Economics, and Organization 4 (1): 141–161.

    Google Scholar 

  14. Furobotn, E. and Richter, R. (1997) Institutions and Economic Theory. Ann Arbor, MI: University of Michigan Press.

    Google Scholar 

  15. Godwin, E.K., Godwin, R.K. and Ainsworth, S. (2007) Is corporate lobbying rational or just a waste of money? In: A.J. Cigler and B.A. Loomis (eds.) Interest Group Politics, 7th edn. Washington DC: CQ Press, pp. 256–278.

    Google Scholar 

  16. Gray, V. and Lowery, D. (1996) The Population Ecology of Interest Representation. Ann Arbor, MI: University of Michigan Press.

    Google Scholar 

  17. Hannan, M.T. and Carroll, G.R. (1992) The Dynamics of Organizational Populations. New York: Oxford University Press.

    Google Scholar 

  18. Heinz, J.P., Laumann, E.O., Nelson, R.L. and Salisbury, R.H. (1993) The Hollow Core. Cambridge, MA: Harvard University Press.

    Google Scholar 

  19. Hillman, A.J., Keim, G.D. and Schuler, D. (2004) Corporate political activity: A review and research agenda. Journal of Management 30 (6): 837–857.

    Google Scholar 

  20. Hirschman, A.O. (1970) Exit, Voice and Loyalty. Cambridge, MA: Harvard University Press.

    Google Scholar 

  21. Huber, J.D. (2000) Delegation to civil servants in parliamentary democracies. European Journal of Political Research 57 (3): 397–413.

    Google Scholar 

  22. Huber, J.D., Shipan, C.R. and Pfahler, M. (2001) Legislatures and statutory control of bureaucracy. American Journal of Political Science 45 (2): 330–345.

    Article  Google Scholar 

  23. Keim, G. and Baysinger, B. (1988) The efficacy of business political activity: Competitive considerations in a principal-agent context. Journal of Management 14 (2): 163–180.

    Article  Google Scholar 

  24. Kersh, R. (2000) State autonomy and civil society: The lobbyist connection. Critical Review 14 (2-3): 237–258.

    Article  Google Scholar 

  25. Kersh, R. (2002) The well-informed lobbyist: Information and interest group lobbying. In: A. Cigler and B. Loomis (eds.) Interest Group Politics, 7th edn. Washington DC: Congressional Quarterly Press, pp. 389–411.

    Google Scholar 

  26. LaPira, T.M. and Thomas, III. H.F. (2014) Revolving door lobbyists and interest representation. Interest Groups and Advocacy 3 (1): 4–29.

    Article  Google Scholar 

  27. Leech, B.L. (2013) Lobbyists at Work. New York: Apress.

    Google Scholar 

  28. Loomis, Jr. B. (2007) Does K Street run through Capitol Hill? Lobbying congress in the republican era. In: A. Cigler and B. Loomis (eds.) Interest Group Politics, 7th edn. Washington DC: Congressional Quarterly Press, pp. 412–432.

    Google Scholar 

  29. Lowery, D. (2007) Why do organized interests lobby? A multi-goal, multi-context theory of lobbying. Polity 39 (1): 29–54.

    Article  Google Scholar 

  30. Lowery, D. (2013) Lobbying influence: Meaning, measurement, and missing. Interest Groups and Advocacy 2 (1): 1–26.

    Article  Google Scholar 

  31. Lowery, D. and Marchetti, K. (2012) You don’t know Jack: Principals, agents, and lobbying. Interest Groups and Advocacy 1 (2): 139–170.

    Article  Google Scholar 

  32. Miller, G.J. (1992) Managerial Dilemmas. Cambridge, UK: Cambridge University Press.

    Google Scholar 

  33. Miller, G.J. (2005) The political evolution of principal-agent models. Annual Review of Political Science 8 (2): 203–225.

    Article  Google Scholar 

  34. Moe, T.M. (1984) The new economics of organization. American Journal of Political Science 28 (4): 739–777.

    Article  Google Scholar 

  35. Moe, T.M. (1987) An assessment of the positive theory of ‘congressional dominance. Legislative Studies Quarterly 12 (4): 475–520.

    Article  Google Scholar 

  36. Olson, M. (1965) The Logic of Collective Action. Cambridge, MA: Harvard University Press.

    Google Scholar 

  37. Rosenthal, A. (2001) The Third House. Washington DC: CQ Press.

    Google Scholar 

  38. Rothenberg, J. (1962) Consumers’ sovereignty revisited and the hospitability of freedom of choice. American Economic Review 52 (2): 269–83.

    Google Scholar 

  39. Schlozman, K.L. (1984) What accent the heavenly chorus? Political equality and the american pressure system. Journal of Politics 46 (4): 1009–1032.

    Article  Google Scholar 

  40. Schlozman, K.L., Verba, S. and Brady, H.E. (2012) The Unheavenly Chorus. Princeton, NJ: Princeton University Press.

    Google Scholar 

  41. Schmidt, S. (2004) Insiders worked both sides of the gaming issue. Washington Post. Page A01.

  42. Scitovsky, T. (1962) On the principle of consumers' sovereignty. American Economic Review 52: 262–268.

    Google Scholar 

  43. Stephenson, M. and Jackson, H.E. (2010) Lobbyists as imperfect agents: Implications for public policy in a pluralist system. Harvard Journal on Legislation 47 (1): 1–20.

    Google Scholar 

  44. Teske, P., Schneider, M., Mintrom, M. and Best, S. (1993) Establishing the micro foundations of a macro theory: Information, movers, and the competitive local market for public goods. American Political Science Review 87 (3): 702–713.

    Article  Google Scholar 

  45. Weingast, B.R. and Moran, M.J. (1983) Bureaucrat discretion or congressional control? Regulatory policymaking by the federal trade commission. Journal of Political Economy 91 (5): 765–800.

    Article  Google Scholar 

  46. Williamson, O.E. (1983) Markets and Hierarchies. New York: Free Press.

    Google Scholar 

  47. Williamson, O.E. (1979) Transaction-cost economics: The governance of contractual relations. Journal of Law and Economics 22: 233–261.

    Article  Google Scholar 

  48. Williamson, O.E. (1981) The economics of organization: The transaction cost approach. American Journal of Sociology 87: 548–577.

    Article  Google Scholar 

  49. Williamson, O.E. (1993) Calculativeness, trust, and economic organization. Journal of Law and Economics 36 (1): 453–486.

    Article  Google Scholar 

  50. Wood, B.D. and Waterman, R.W. (1991) The dynamics of political control of the bureaucracy. American Political Science Review 85 (3): 801–828.

    Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to David Lowery.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Schiff, E., Seufer, K., Whitesell, A. et al. Agency problems and interest representation: An empirical analysis of the costs of lobbying. Int Groups Adv 4, 225–248 (2015).

Download citation


  • lobbying
  • interest organizations
  • representation