Raising the bar for incumbents

Hans Gersbach, 3 January 2012



Incumbents in the US House of Representatives or US Senate are extraordinarily successful when they seek re-election. Depending on the office-holder's ability and time horizon, this can be a good or bad thing for the country’s voters. If an incumbent is a more able and far-sighted legislator than his challengers, the incumbency advantage will be welfare-improving, and elections are a good selection device.1 Challengers may even be deterred from running against incumbents in the first place.2 However, less able legislators are also re-elected, and this is detrimental. For example, an office-holder may be re-elected merely because he is better known than his challengers. Moreover, an office-holder has ample opportunity to improve his re-election chances through his policies and actions in the run-up to elections, which can help him prevail over formidable challengers who take a longer view. An incumbent may, for instance, undertake actions that are socially costly, such as contributing to pork-barrel legislation3, or supporting measures that shift the onus of financing current public expenditures onto future generations, resulting in excessive public debt accumulation.4 Although such actions are likely to be socially detrimental, they may attract the support of interest groups and may ensure the incumbent’s re-election.

Higher vote thresholds

No direct influence can be exerted on the sources of the advantages of incumbency because they are too complex. So it would be useful to have an election rule that curbs the advantages of incumbency when it would reduce welfare, and preserves them when it would be socially beneficial. To achieve this objective, we suggest setting higher re-election hurdles for incumbents.

In two-candidate races in particular, our proposal is that for incumbents to be re-elected, they must obtain a percentage of the vote that exceeds 50% by a step or two.5 If the incumbent does not reach this threshold, either his challenger is elected or a conditional run-off ballot between two other candidates will take effect. The conditional procedure would ensure that the candidates elected to public office would always receive the support of at least 50% of the citizens who cast their vote.

Vote-share thresholds for incumbents affect political outcomes in a number of ways. First, a higher threshold stimulates greater effort, as the incumbent wants to be re-elected and re-election is less certain. This is desirable.

Second, a higher threshold tends to favour able incumbents in their re-election bids, as they will be more able to garner enough votes to attain it. This is socially desirable as long as incumbents with above-average ability are re-elected. If the threshold is too high, however, even incumbents with above-average qualities as legislators would be deselected, which is socially undesirable. A socially optimal vote-share threshold that balances advantages and disadvantages is expected to be above 50% and to increase with the number of times a member of Congress stands for re-election.6

Third, when the ideological preferences of office-holders (and the citizens supporting them) are very diverse, higher vote-share thresholds may help to achieve political compromises, as the veto power of interest groups with immoderate agendas will decline. This may be particularly desirable in the context of controlling public debt, as we will demonstrate next.

Less polarisation, more reforms

Vote thresholds could be a powerful instrument for curtailing excessive public debt and reducing polarisation in Congress. Consider, for example, the recent discussions surrounding the failure of the so-called “Supercommittee”7 to propose measures for the restoration of fiscal balance. Party polarisation was so strong that members had categorically ruled out support for various policy measures designed to reduce the budget deficit. The intersection between the policy packages supported by the Democrats and the Republicans turned out to be empty.

Higher thresholds for incumbents would induce greater policy moderation, as incumbents would need to gain broader support to be re-elected. Moreover, as sitting members of the House and the Senate would not be able to easily secure re-election simply by blocking the other party’s proposals, a debt reduction plan with a suitable combination of spending cuts and increases in revenues would have a better chance of gaining approval by Congress.

Competition with higher re-election hurdles

A vote-share threshold could be set by the legislature itself. Alternatively, one could also allow candidates to compete on the basis of vote-share thresholds. They could commit to a percentage of votes they would need to obtain on re-election day in order to remain in office. Then the majority of voters would tend to elect the candidate who commits to a vote-share threshold that is closer to the one that this majority desires. True, candidates will still be moved by their ideological objectives to some extent. But higher hurdles for re-election would help limit those incumbency advantages that are socially detrimental.

If vote-share thresholds were offered competitively by the candidates at a fixed point during their campaigns, we would expect a substantial range of such thresholds across districts and states. In districts that are relatively safe for one party, re-election thresholds would typically be higher than in closely contested districts. Over time, we would also expect the typical threshold for incumbents to increase with the number of terms in office. As a consequence, the average total time in office will decline.


Higher vote thresholds for incumbents would increase the selection power of democratic elections. As they are easy to implement, it would be useful to experiment with this new instrument. Of course, institutional changes may have unintended consequences that are impossible to anticipate. Nevertheless, the potential of higher vote thresholds to improve welfare by mitigating the advantages of incumbency is such that liberal democracy can afford to explore this field – and should certainly do so.


112th United States Congress (2011), Budget Control Act of 2011, approved August 2, 2011.
Alesina, A and A Cukierman (1990), “The Politics of Ambiguity”, Quarterly Journal of Economics, 105(4):829-850.
Ashworth, S (2005), “Reputational Dynamics and Political Careers”, Journal of Law, Economics and Organization, 21(2):441-466.
Banks, JS and RK Sundaram (1998), “Optimal Retention in Agency Problems”, Journal of Economic Theory, 82:293-323.
Becker, JG, H Gersbach, and O Grimm (2010), “Debt-sensitive Majority Rules”, CEPR Discussion Paper DP7860.
Cox, GW and JN Katz (1996), “Why Did the Incumbency Advantage in US House Elections Grow?”, American Journal of Political Science, 40(2):478-497.
Cukierman, A and M Tommasi (1998), “When Does it Take a Nixon to Go to China?”, American Economic Review, 88(1):180-197.
Diermeier, D, M Keane, and A Merlo (2005), “A Political Economy Model of Congressional Careers”, American Economic Review, 95:347-373.
Gersbach, H (2007), “Vote-share Contracts and Democracy”, VoxEU.org, 26 September and CEPR Discussion Paper no. 6497.
Gersbach, H (2009), “Higher Vote Thresholds for Incumbents, Effort and Selection”, CEPR Discussion Paper DP7320.
Gersbach, H (2010), “Government Debt Threshold Contracts”, CEPR Discussion Paper DP8001.
Gordon, SC, GA Huber, and D Landa (2007), “Challenger Entry and Voter Learning”, American Political Science Review, 101(2):303-320.
Hess, GD and A Orphanides (1995), “War Politics: An Economic, Rational Voter Framework”, American Economic Review, 85(4):828-846.
Hess, GD and A Orphanides (2001), “War and Democracy”, Journal of Political Economy, 109(4):776-810.
Jacobson, GC and S Kernell (1983), Strategy and Choice in Congressional Elections, 2nd Ed., Yale University Press, New Haven, CT.
Londregan, J and T Romer (1993), “Polarization, Incumbency, and the Personal Vote”, in William A Barnett, Melvin J Hinich, and Norman J Schofield (eds.), Political Economy: Institutions, Competition, and Representation, Cambridge University Press, 355-377.
Rogoff, K and A Sibert (1988), “Elections and Macroeconomic Policy Cycles”, Review of Economic Studies, 55:1-16.
Samuelson, L (1984), “Electoral Equilibria with Restricted Strategies”, Public Choice, 43:307-327.
Stone, WJ, LS Maisel, and CD Maestas (2004), “Quality Counts: Extending the Strategic Politician Model of Incumbent Deterrence”, American Journal of Political Science, 48(3):479-495.
Zaller, J (1998), “Politicians as Prize Fighters: Electoral Selection and the Incumbency Advantage”, in John G Geer (ed.), Politicians and Party Politics, Johns Hopkins University, Baltimore, MD, 128-185.

1 See Samuelson (1984), Londregan and Romer (1993), Banks and Sundaram (1998), Zaller (1998), Ashworth (2005), and Diermeier et al. (2005).
2 See Jacobson and Kernell (1983), Cox and Katz (1996), Stone et al. (2004), and Gordon et al. (2007).
3 See e.g. Rogoff and Sibert (1988), Alesina and Cukierman (1990), Hess and Orphanides (1995, 2001), and Cukierman and Tommasi (1998).
4 See Gersbach (2010) and Becker, Gersbach and Grimm (2010) and the literature discussed there.
5 This idea has been developed in an earlier Vox column, Gersbach (2007).
6 This has been shown in various models (see Gersbach 2007 and 2009). There is no indication that higher vote thresholds induce an increase in pork barrel projects.
7 Officially known as the Joint Select Committee on Deficit Reduction and established by the Budget Control Act of 2011 of the 112th United States Congress on August 2, 2011.


Topics: Politics and economics
Tags: elections, politics, US

Professor at CER-ETH - Center of Economic Research at ETH Zurich and CEPR Research Fellow