Skip to main content

Political competition in judge and prosecutor elections


The United States is unique in that important actors within the criminal justice system, namely judges and prosecutors, are selected in popular election. Several states are currently adjusting whether political party affiliation is listed on the ballot. Additionally, states differ by how easy it is for candidates from non-dominant parties to gain access to the ballot. We use a laboratory experiment to investigate how these two important policy changes to political competition affect campaign spending and outcomes. Using asymmetric contests designed to capture the institutional change, we find that subjects spend beyond both the socially optimal level and the amount predicted by theory. This over-competition is not uniform, but rather concentrated in those subjects who have the strategic advantage (either dominant-party affiliation or restricted ballot access of competitors). Opening up the election process, therefore, leads to reductions in the wasteful, rent-seeking spending (contrary to theory). Disadvantaged subjects are less likely to exit the race when the election process is opened, as well. Thus, a level playing field promotes participation. Furthermore, we explore heterogeneous treatment effects, finding less campaign spending for risk-loving, non-ambiguity averse, strategically sophisticated, and pro-social subjects. Therefore, elections disproportionately select individuals without these characteristics. Additionally, the mix of those who win the election adjust with changes in the electoral rules as well.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2


  1. 1.

  2. 2.

  3. 3.

    In fact, it is not uncommon for an individual running for a judge or prosecutor position to be unable to gain the nomination of a dominant party, but then be promoted by a third-party to the general election ballot.

  4. 4.

    North Carolina requires 2% of the registered voting population to sign a petition to grant an unaffiliated candidate ballot access. The costs are not limited to the time and effort involved in collecting signatures though. It often involves attorney fees as a dominant party often challenges the signatures in a pre-election lawsuits.

  5. 5.

    Local prosecutors are chosen by popular election in forty-six states. The states that appoint them are Alaska, Connecticut, New Jersey, and Rhode Island. District-level judges are routinely elected by their communities. Public defenders, on the other hand, are commonly appointed. Exceptions exists in four states (Florida, Nebraska, Tennessee, and California).

  6. 6.

    The retention mode (either popular election or appointment) has also been shown to correlate with sentencing (Iaryczower et al. 2013), judicial quality (Choi et al. 2010), litigation (Hanssen 1999), and influence campaign financing (Kang and Shepherd 2015).

  7. 7.

    Our framework does not include any positive externalities as would arise, for example, in R&D contests or workplace effort exertion tournaments.

  8. 8.

    See Nagel (1961) for an early example. Also, there is an active literature in political science assessing voter behavior. See Bonneau and Cann (2015) for an example within the context of judicial elections.

  9. 9.

    The experimental literature on Tullock contests is too expansive to fully document here. An interested reader is encouraged to consult Decheneaux et al. (2015) and the reference discussed therein. Stripping away the context, we provide new results when competition is asymmetric. Previous work has considered endowment asymmetries (Kimbrough et al. 2014). We are the first, to our knowledge, to explore multipliers and free entries into the contest.

  10. 10.

    Our work complements the literature using laboratory methods to study voting behavior. For example, Aimone et al. (2018) document altruistic punishment motivations for indifferent voters. In our analysis, voters are passive players and we focus on political competition between the candidates running for office.

  11. 11.

    It is more general to assume player i wins the “tie” with probability \(p_{i}\) (assuming \(\Sigma p_{i}=1\)). The Nash equilibrium is unaffected by the value of \(p_{i}\). Thus, this assumption is not crucial.

  12. 12.

    In all treatments except the Baseline, we also consider the environment where player 2 also has the advantage In the Asymmetric treatment, \(\omega _{2}=110\) as well. In the Multiplier treatment, \(\alpha _{2}=2\). Finally, in the Endowment treatment, player 2 also receives 10 free entries into the Account.

  13. 13.

    Others have incorporated similar but distinct versions similar to our Multiplier treatment. It has been justified to capture asymmetric positions in conflict (Kimbrough et al. 2014) and asymmetric productivities, such as R&D races (Anderson and Freeborn 2009). Thus, we match their arguments by thinking of partisanship as an asymmetric conflict.

  14. 14.

    On average, subjects completed all the treatments in 24.7 min with a median of 25.0 and a maximum of 36.7 min.

  15. 15.

    In the lab, we also added a fourth treatment. In it, the advantaged player is given an endowment of 110 tokens, rather than the 10 tokens added to the Account in the Endowment treatment. In our upcoming analysis, we will compare behavior in the Baseline treatment to the Endowment treatment. A concern is that choices in the latter are simply an artifact of having, in effect, a larger value of \(\omega _{i}\), and not due to the “head start” contribution directly. Adding this fourth treatment allows us to separate these effects. Theory predicts that behavior in this fourth treatment should be the same as in the Baseline treatment. This additional treatment, like the Baseline, was done over 11 rounds. Thus, the laboratory subjects engaged in 42 rounds, but the analysis we present here only considers the 31 rounds.

  16. 16.

    Nonparametric tests confirm the statistical differentiation in behavior between subjects in the Baseline treatment, when advantaged in the Multiplier treatment, and when in the disadvantaged position. A Wilcoxon Ranksum test verifies that the three samples are pairwise distinct (\(|z|>6\) and \(p<0.001\) for all three tests), t tests for difference in means with unequal variances rejects the null hypothesis that any two means are equal (\(|t|>5\) and \(p<0.001\) for all three tests), and two-sample Kolmogorov–Smirnov tests for stochastic dominance find distributional differences for pairwise comparisons (\(D>0.10\) and \(p<0.001\) for all three tests)

  17. 17.

    We observe spending of 35% of the endowment, which is 64% greater than equilibrium predictions with \(n=2\). Their design differs in that they include the option of a coin-flip determined even divide of the prize to avoid the conflict.

  18. 18.

    Table 2 presents the results from a fixed effects model with HAC robust standard errors. If a random effects model is estimated, a Hausman test rejects the estimate’s consistency. Furthermore, the Welch F test indicates that the subject-specific intercepts are jointly significant. Therefore, the fixed effects model is more appropriate. The signs, magnitudes, and statistical significance of the explanatory variables remain relatively unchanged though when the random effects or pooled OLS specifications are estimated. Additionally, if the standard errors are clustered by round, the statistical significance of the treatment variables and number of candidates persists. Thus, the results presented in Table 2 are robust to the specification used.

  19. 19.

    Nonparametric statistical tests provide similar results. Comparing the choices made when in the advantaged position sample (pooling across group sizes) to both the baseline, symmetric environment and the disadvantaged decision making, a Wilcoxon Ranksum test confirms their distinction (\(|z|=8.4\) and 10.8, respectively, with \(p<0.001\)). Relatedly, t tests for difference in means with unequal variances confirm their difference (\(|t|=7.7\) and 11.5, respectively, with \(p<0.001\)), and two-sample Kolmogorov-Smirnov tests reveal a stochastic dominance relationship (\(D=0.16\) and 0.19, respectively, with \(p<0.001\)). Comparing decision making in the Baseline treatment to being disadvantaged in the Endowment treatment is less clear. While overlapping confidence intervals arise when \(n=4\), a Wilcoxon Ranksum test suggests the two samples are drawn from different distributions when the group sizes are pooled (\(|z|=3.2\) with \(p=0.0012\)). Similarly, a t test and a Kolmogorov-Smirnov test suggest their distinction (\(|t|=4.4\) and \(D=0.05\) with \(p<0.001\) and \(p=0.017\), respectively).

  20. 20.

    If we further consider ballot access’ nonlinear effects by including interaction terms between the number of candidates and treatment, behavior in the two-candidate elections continue to differ from that in three-candidate races. Again, a significant negative coefficient exists on the interaction between having two candidates and being in the disadvantaged position in the Endowment treatment. Similar findings arise if interaction terms are added to (2) in Table 2 as well.

  21. 21.

    As mentioned in the methods section, we also included a fourth treatment with asymmetric endowments. Theory predicts behavior in it replicates the Baseline treatment and, hence, differs from the Endowment treatment. While not presented here, the behavior considering that treatment matches qualitatively those presented in Table 3. The subjects with the additional endowment behave similarly to those with the ten-token entry in the Account. Those in the disadvantaged position do not behave differently. Thus, it is having the advantage which drives over-spending.

  22. 22.

    It is straightforward to verify that there does not exist an asymmetric equilibrium, given the parameter values used, where one or more players choose a zero contribution.

  23. 23.

    We consider any subject who does not enter an amount (i.e., a blank entry) as exiting. Blank entries make up 2.7% and 3.0%, respectively, of the samples used.

  24. 24.

    Background controls we include are a constant, year in school, and indicator variables for being female, white, business major, WV resident, and a non-U.S. citizen.

  25. 25.

    Risk is the selected reservation price for a 50–50 gamble of receiving $0 or $20. Ambiguity aversion is the difference in the reservation prices in a risky and ambiguous gamble. We use the BDM mechanism developed by Halevy (2007). These two variables have means of 8.875 and 1.537 and standard deviations of 3.889 and 3.049, respectively. It is worth pointing out that it is common in the literature to see an inverse relationship between risk aversion and spending, while we report a positive relationship. If Ambiguity Aversion is replaced with the reservation price in the ambiguity decision problem, then Risk is insignificant and it is positive and statistically significant. This suggests that the relationship identified in the literature may, in fact, be capturing ambiguity preferences rather than risk preferences. Alternatively, the literature uses a price menu approach for measuring risk, while we use a BDM mechanism. The difference could be due to the risk elicitation used.

  26. 26.

    See Ho et al. (1998) for a discussion of the Beauty Contest’s use measuring sophistication. The mean guess in our sample is 33.34 with a standard deviation of 19.24.This is quite high and indicates that most subjects are level-0 rational individuals. In our sample, 13.3% are level-1 (Beauty Contest\(\in [22.28]\)) and 10.0% are level-2 (Beauty Contest\(\in []\)). This suggests, absent the use of examples and allowing questions, some subjects did not comprehend the game. Nevertheless, a correlation between choices made and behavior in the contest exists.

  27. 27.

    The mean value contributed in the Public Goods Game is 3.04 with a standard deviation of 1.65. The assessments also include a Trust Game. If choices in it replace the Public Goods Game giving, the same results arise. Choices in the two decision problems are highly correlated.

  28. 28.

    For example, Lim and Snyder (2015) provide evidence that partisan affiliation provides a valuable informational cue to voters in judicial elections.


  1. Aimone, J. A., Butera, L., & Stratmann, T. (2018). Altruistic punishment in elections. European Journal of Political Economy, 53, 149–160.

    Article  Google Scholar 

  2. Anderson, L. R., & Freeborn, B. (2009). Varying the intensity of competition in a multiple prize rent seeking experiment. Public Choice, 143(1–2), 237–254.

    Google Scholar 

  3. Ash, E., & Bentley MacLeod, W. (2015). Intrinsic motivation in public service: Theory and evidence from state supreme courts. Journal of Law and Economics, 58(4), 863–914.

    Article  Google Scholar 

  4. Bandyopadhyay, S., & McCannon, B. C. (2014). The effect of the election of prosecutors on criminal trials. Public Choice, 161(1–2), 141–156.

    Article  Google Scholar 

  5. Bandyopadhyay, S., & McCannon, B. C. (2015). Prosecutorial retention: Signaling by trial. Journal of Public Economic Theory, 17(2), 219–256.

    Article  Google Scholar 

  6. Bandyopadhyay, S., & McCannon, B. C. (2018). Queuing up for justice: Prosecutor elections and case backlogs, Working Paper.

  7. Berdejo, C., & Yuchtman, N. (2013). Crime, punishment, and politics: An analysis of political cycles in criminal sentencing. Review of Economic Statistics, 95(3), 741–756.

    Article  Google Scholar 

  8. Bonneau, C. W., & Cann, D. M. (2015). Party identification and vote choice in partisan and nonpartisan elections. Political Behavior, 37(1), 43–66.

    Article  Google Scholar 

  9. Brandts, J., & Charness, G. (2011). The strategy versus the direct-response method: A first survey of experimental comparisons. Experimental Economics, 14(3), 375–398.

    Article  Google Scholar 

  10. Caplan, B. (2000). Rational ignorance versus rational irrationality. Kyklos, 54(1), 3–26.

    Article  Google Scholar 

  11. Caplan, B. (2007). The myth of the rational voter. Princeton: Princeton University Press.

    Google Scholar 

  12. Choi, S. J., Mitu Gulati, G., & Posner, E. A. (2010). Professionals or politicians: The uncertain empirical case for an elected rather than appointed judiciary. Journal of Law, Economics, and Organization, 26(2), 290–336.

    Article  Google Scholar 

  13. Cohen, A., Klement, A., & Neeman, Z. (2015). Judicial decision making: A dynamic reputation approach. Journal of Legal Studies, 44(S1), S133–S159.

    Article  Google Scholar 

  14. DeAngelo, G., & McCannon, B. C. (2017). Judicial compensation and performance. Supreme Court Economic Review, 25(1), 129–147.

    Article  Google Scholar 

  15. DeAngelo, G., & McCannon, B. C. (2018). Using appellate decision to evaluate the impact of judicial elections, Working Paper.

  16. Decheneaux, E., Kovenock, D., & Sheremeta, R. M. (2015). A survey of experimental research on contests, all-pay auctions and tournaments. Experimental Economics, 18(4), 609–669.

    Article  Google Scholar 

  17. Downs, A. (1957). An economic theory of democracy. New York: Harper and Brothers.

    Google Scholar 

  18. Drometer, M., & Rincke, J. (2009). The impact of ballot access restrictions on electoral competition: Evidence from a natural experiment. Public Choice, 138(3), 461–474.

    Article  Google Scholar 

  19. Drometer, M., & Rincke, J. (2014). Electoral competition and endogenous barriers to entry. European Journal of Political Economy, 34, 253–262.

    Article  Google Scholar 

  20. Dyke, A. (2007). Electoral cycles in the administration of criminal justice. Public Choice, 133(3/4), 417–437.

    Article  Google Scholar 

  21. Engel, C., & Reuben, A. (2015). The people’s hired gun? Experimentally testing the motivating force of legal frame. International Review of Law and Economics, 43, 67–82.

    Article  Google Scholar 

  22. Engel, C., & Zhurakhovska, L. (2017). You are in charge: Experimentally testing the motivating power of holding judicial office. Journal of Legal Studies, 46(1), 1–50.

    Article  Google Scholar 

  23. Gordon, S. C., & Huber, G. (2002). Citizen oversight and electoral incentives of criminal prosecutors. American Journal of Political Science, 46(2), 334–351.

    Article  Google Scholar 

  24. Gordon, S. C., Huber, G. A., & Landa, D. (2007). Challenger entry and voter learning. American Political Science Review, 101(2), 303–320.

    Article  Google Scholar 

  25. Halevy, Y. (2007). Ellsberg revisited: An experimental study. Econometrica, 75(2), 503–536.

    Article  Google Scholar 

  26. Hanssen, F. A. (1999). The effect of judicial institutions on uncertainty and the rate of litigation: The election versus appointment of state judges. Journal of Legal Studies, 28(1), 205–232.

    Article  Google Scholar 

  27. Ho, T.-H., Camerer, C., & Weigelt, K. (1998). Iterated dominance and iterated best responses in p-beauty contests. American Economic Review, 88(4), 947–969.

    Google Scholar 

  28. Iaryczower, M., Lewis, G., & Shrum, M. (2013). To elect or to appoint? Bias, information, and responsiveness of bureaucrats and politicians. Journal of Public Economics, 97, 203–244.

    Article  Google Scholar 

  29. Kang, M. S., & Shepherd, J. M. (2011). The partisan price of justice: An empirical analysis of campaign contributions and judicial decisions. New York University Law Review, 86(1), 69–130.

    Google Scholar 

  30. Kang, M. S., & Shepherd, J. M. (2015). Partisanship in state supreme courts: The empirical relationship between party campaign contributions and judicial decision making. Journal of Legal Studies, 44(S1), S161–S185.

    Article  Google Scholar 

  31. Kimbrough, E. O., Sheremeta, R. M., & Shields, T. (2014). When parity promotes peace: Resolving conflict between asymmetric agents. Journal of Economic Behavior and Organization, 99, 96–108.

    Article  Google Scholar 

  32. Kräkel, M., Nieken, P., & Przemeck, J. (2012). Risk taking and investing in electoral competition, Working Paper.

  33. Lim, C. S. H. (2013). Preferences and incentives of appointed versus elected public officials: Evidence from state trial court judges. American Economic Review, 103(4), 1360–1397.

    Article  Google Scholar 

  34. Lim, C. S. H., & Snyder, J. M, Jr. (2015). Is more information always better? Party cues and candidate quality in U.S. Judicial Elections. Journal of Public Economics, 128, 107–123.

    Article  Google Scholar 

  35. Lim, C. S. H., Snyder, J. M, Jr., & Strömberg, D. (2015). The judge, the politician, and the press: Newspaper coverage and criminal sentencing across electoral systems. American Economic Journal: Applied Economics, 7(3), 103–135.

    Google Scholar 

  36. McCannon, B. C. (2013). Prosecutor elections, mistakes, and appeals. Journal of Empirical Legal Studies, 10(4), 697–715.

    Article  Google Scholar 

  37. McCannon, B. C., & Pruitt, J. (2018). Taking on the boss: Informative contests in prosecutor elections. Journal of Public Economic Theory, 20(5), 657–671.

    Article  Google Scholar 

  38. Nagel, S. S. (1961). Political party affiliation and judges’ decisions. American Political Science Review, 55(4), 843–850.

    Article  Google Scholar 

  39. Öncüler, A., & Croson, R. (2005). Rent-seeking for a risky rent—A model and experimental investigation. Journal of Theoretical Politics, 17(4), 403–429.

    Article  Google Scholar 

  40. Posner, R. A. (1993). What do judges and justices maximize? (The same thing everybody else does). Supreme Court Economic Review, 3(1), 1–41.

    Article  Google Scholar 

  41. Shepherd, J. M. (2009). The influence of retention politics on judges’ voting. Journal of Legal Studies, 38(1), 169–206.

    Article  Google Scholar 

  42. Sheremeta, R. M. (2010). Expenditures and information disclosure in two-stage political contests. Journal of Conflict Resolution, 54(5), 771–798.

    Article  Google Scholar 

  43. Sheremeta, R. M. (2013). Overbidding and heterogeneous behavior in contest experiments. Journal of Economic Surveys, 27(3), 491–514.

    Article  Google Scholar 

  44. Sonnemans, J., van Dijk, F., Donders, B., & Bauw, E. (2018). An economic approach on countering the misuse of the right to challenge judges: An experiment. European Journal of Law and Economics, 45, 29–57.

    Article  Google Scholar 

  45. Spamann, H., & Klohn, L. (2016). Justice is less blind, and less legalistic, than we thought: Evidence from an experiment with real judges. Journal of Legal Studies, 45(2), 255–280.

    Article  Google Scholar 

  46. Stratmann, T. (2005). Ballot access restrictions and candidate entry in elections. European Journal of Political Economy, 21(1), 59–71.

    Article  Google Scholar 

  47. Tullock, G. (1971). Public decisions as public goods. Journal of Political Economy, 79(4), 913–915.

    Article  Google Scholar 

  48. Tullock, G. (1975). On the Efficient organization of trials. Kyklos, 28(4), 745–762.

    Article  Google Scholar 

  49. Tullock, G. (1980). Efficient rent seeking. In J. M. Buchanan, R. D. Tollison, & G. Tullock (Eds.), Toward a theory of the rent-seeking society (pp. 97–112). College Station, TX: Texas A&M University Press.

    Google Scholar 

  50. Wright, R. F. (2009). How prosecutor elections fail US, Ohio State. Journal of Criminal Law, 6(2), 581–610.

    Google Scholar 

Download references


We thank Cristina Bicchieri, John Dinan, Dan Houser, Kai Ou, Roman Sheremeta, Thomas Stratmann, Rick Wilson, and participants of the Economic Science Association meetings and Experimental Public Choice Workshop at Lille Catholic University. We appreciate financial support from the Center for Free Enterprise at WVU.

Author information



Corresponding author

Correspondence to Bryan C. McCannon.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.



Here is a screen shot of the instructions used in the online program. Figure 3 depicts the instructions from the Multiplier treatment. The bolded words BLUE and RED are colored blue and red, respectively, on the screen (as is the player labels ORANGE and GREEN).An example is provided for each treatment. Figure 4 presents the example used in the Multiplier treatment. Figure 5 depicts a screen shot of the decision problem a subject faces in each round of each treatment. Slider bars are used for subjects to make their choices. For two treatments (all other than the Baseline treatment), two choices are made—one if selected for the advantaged position and one if selected for the disadvantaged position. In the Baseline treatment only one slider bar is displayed.

Fig. 3

Instructions screen shot

Fig. 4

Example screen shot

Fig. 5

Decision problem screen shot

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

DeAngelo, G., McCannon, B.C. Political competition in judge and prosecutor elections. Eur J Law Econ 48, 167–193 (2019).

Download citation


  • Asymmetric contest
  • Ballot access
  • Campaign spending
  • Election
  • Judge
  • Partisan affiliation
  • Prosecutor
  • Tullock contest

JEL Classification

  • H11
  • D01
  • H40
  • C91