VoxEU Column Politics and economics

Institutional design and the quality and behaviour of government officials: Variation in selection and retention methods

There are many ways to select and retain public officials in representative democracies. This column discusses the literature on selection and retention procedures for low-information public offices. The authors first overview the historical origins of the variation. Then, they discuss conceptual frameworks used to assess the functioning of direct elections and present empirical regularities.  Finally, the authors suggest avenues for future research aided by digitisation and improved textual analysis of media coverage and government data.

In 2015, a sexual assault case shook one of the world’s top universities. Brock Turner, a student at Stanford University, was caught sexually assaulting an intoxicated and unconscious woman on campus. The resulting legal case, People v. Turner (2015), caused a more complex controversy than the offence itself due to the lenient sentence handed out by the judge. The offences Turner was convicted of carried a statutory minimum sentence of two years in prison. The prosecutor recommended a prison sentence of six years. The judge sentenced Turner to six months in county jail followed by three years of probation. Turner served three months and was released early. The leniency of this sentencing decision eventually led to a recall election in which voters ousted the judge from the bench (the first such recall in nearly a century). 

The political system that allows voters to elect and remove judges is unique to the US. However, it is not unique to the judiciary that US state governments let voters elect public officials who perform highly bureaucratic and technical tasks. For example, public utility regulators and school board superintendents are elected in many states. More interestingly, for a given office, there is much cross-state and cross-time variation in whether officials are elected by the voters or appointed by the governor, and which electoral processes are employed. Moreover, there is considerable variation around the world in the selection and retention rules for executives of local governments. 

This institutional variation in the selection of public officials leads us to the following questions:

  • What is the history behind this institutional variation?
  • What conceptual framework can we use to understand its influence on government actions and voter welfare?
  • What empirical understanding does the existing research provide?
  • What are the fruitful directions for future research in this area? 

Our paper (Lim and Snyder forthcoming) provides an overview of the recent advances in this literature, focusing on the four questions above. 

History and overview

The principal variation in selection rules we study is whether the officials are elected directly by the voters, or appointed by another official, who is often the head of the executive branch. The second variation is in the election rule – whether it is partisan or nonpartisan. In the US, the movement to elect state officials was largely associated with 19th-century Jacksonian democracy and the early 20th-century populist and progressive movements. The selection rules in a state often vary according to which system was popular in the era when the state joined (or re-joined) the Union. This history led to geographic clustering in selection systems. 

Outside the US, variation in the selection rules for the executives of local governments evolved over the second half of the 20th century. Examples we discuss in the paper are Germany, the UK, Italy, Indonesia, and China.

Conceptual framework

The basic framework economists use to analyse the benefits of direct election is the principal-agent model, wherein voters are the principal and government officials are the agents. The basic argument is that government officials are likely to serve voters’ interests better when voters can elect them directly or force them out. In practice, however, such an argument is based on a set of strong assumptions, including an expectation that voters are relatively well-informed about the candidates. A large body of work specifies institutional and non-institutional factors that undermine the functioning of direct elections, which we discuss below. 

One key institutional factor is the role of party cue on ballots. In many elections – even for government officials who perform tasks that are not highly partisan –a candidate’s party affiliation is placed on the ballot (‘partisan elections’). Voters in such elections tend to vote strongly along party lines, reducing the influence that candidates’ characteristics and behaviour have on election outcomes. Another institutional factor is the timing of elections. Elections for ‘low-information offices’ – such as judges, regulators, and school boards – often exhibit low turnout when not combined with major elections. When turnout is very low, special interest groups like public sector unions can have disproportionate influence over election outcomes. A third institutional factor is the collective decision-making process. Many government decisions are made collectively. In such cases, a system in which voters hold individual officials accountable may not be an ideal institutional design. 

One important non-institutional factor that undermines the desirability of direct elections is the incentive for officials to pander. When government officials have private information that influences the relative desirability of different policy options, they have an incentive to ignore such information to appease voters who do not have the same information. The second non-institutional factor is potential adverse selection. If officials must routinely alter their decisions in order to please voters, or else face a high probability of losing re-election, then holding public office will be less valuable. This, in turn, might discourage those with high human capital and good outside options from pursuing public office.

Other factors have ambiguous effects on the desirability of direct elections. For instance, heterogeneity of voter preferences across localities can render markedly different policy outcomes under election and appointment. In the appointment regime, the selection is centralised in the hands of the appointer, which should tend to reflect the preferences of a larger polity. In contrast, localities choose their own officials in the election regime, which will tend to produce greater spatial variation in government policies. 

Empirical findings 

Empirical findings regarding the benefit of direct elections have been as mixed as theoretical predictions. Besley and Coate (2003) and Gordon and Huber (2007) suggest that regulators’ and judges’ decisions are more aligned with voter preferences under direct election, respectively. Huber and Gordon (2004), Berdejo and Yuchtman (2013), and Hessami (2018) document the influence of election cycles to demonstrate judges’ and mayors’ responsiveness to direct elections not observed among appointments. In contrast, Choi et al. (2010), Iaryczower et al. (2013), Ash and MacLeod (2015), and Ash and MacLeod (2019) document the negative influence of direct elections on the quality of judges’ work. On the other hand, Martinez-Bravo et al. (2017) and Lim et al. (2013) focus on the role of direct elections in decentralisation, and argue that it serves local voters’ preferences better. 

Empirical studies find that the role of party cue, discussed above, is significant in determining the functioning of direct elections. Canes-Wrone et al. (2014) and Lim and Snyder (2015) document that outcomes in nonpartisan elections tend to better reflect voter preferences and candidate quality, respectively, compared with partisan elections. Relatedly, Lim et al. (2015) demonstrate that active media coverage significantly influences judicial decisions, but only in the nonpartisan election regime.

The effect of elections also depends on the pecuniary and non-pecuniary benefits from the office. For many public offices, voters do not have much information about candidates at the initial selection, making re-election incentives the primary mechanism for congruence between officials’ behaviour and voter preferences. However, re-election incentives can only function well if payoffs from the office are significant enough. In contrast, the appointer often has detailed information about the candidate, leading to a strong selection on policy preferences at the initial stage. These mechanisms are demonstrated in Lim 2013, which estimates a structural model, specifying judges’ policy preferences and payoffs from the office. Lim’s estimates show significant non-pecuniary payoffs (e.g. prestige) for the judiciary – which plays an important role in incentives for elected judges – and argues that the effect of selection systems on policy can differ substantially across offices with different payoffs. 

Directions for future research 

Despite the richness of the literature, there is much room for innovative research in this area. On the theoretical front, we need more advances in the framework for welfare evaluation. Existing studies have focused primarily on proposing positive theories, predicting how different systems would render different policy outcomes. The discussion above suggests that the desirability of direct elections depends very much on the specific nature of the tasks (e.g. information asymmetry) and job design (e.g. compensation), as well as characteristics of surrounding political environments (e.g. preference heterogeneity and media environments). Frameworks that map such conditioning factors to optimal institutional design would be a valuable contribution. 

On the empirical front, digitisation of government records, improvements in computing power, and technical advances in the analysis of text (text-as-data) have opened avenues for new research in several directions. First, now we can easily digitise media and government documents, and use them to measure various aspects of governance (e.g. corruption, quality of public service). Second, it has become easier to construct large-scale data on government officials and candidates. This can significantly improve our understanding of the role that institutions play in political selection. Third, over the past decade we have seen significant advances in the textual analysis of media coverage. We can use such advances to understand the information voters receive about government officials under different selection systems. 


Ash, E and W B MacLeod (2015), “Intrinsic motivation in public service: theory and evidence from state supreme courts”, The Journal of Law and Economics 58: 863–913.

Ash, E and W B MacLeod (2019), “Selection and incentive effects of elections: evidence from state supreme courts”, NBER Working Paper 22071.

Berdejó, C and N Yuchtman (2013), “Crime, punishment, and politics: an analysis of political cycles in criminal sentencing”, The Review of Economics and Statistics 95: 741–56.

Besley, T and S Coate (2003), “Elected versus appointed regulators: theory and evidence”, Journal of the European Economic Association 1: 1176–206.

Canes-Wrone B, T S Clark and J P Kelly (2014), “Judicial selection and death penalty decisions”, American Political Science Review 108: 23–39.

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

Gordon, S C and G A Huber (2007), “The effect of electoral competitiveness on incumbent behavior”, Quarterly Journal of Political Science 2: 107–38.

Hessami, Z (2018), “Accountability and incentives of appointed and elected public officials”, The Review of Economics and Statistics 100: 51–64.

Huber, G A and S C Gordon (2004), “Accountability and coercion: Is justice blind when it runs for office?”, American Journal of Political Science 48: 247–63.

Iaryczower M, G Lewis and M Shum (2013), “To elect or to appoint? Bias, information, and responsiveness of bureaucrats and politicians”, Journal of Public Economics 97: 230–44.

Lim, C S H (2013), “Preferences and incentives of appointed and elected public officials: evidence from state trial court judges”, American Economic Review 103: 1360–97.

Lim, C S H and J M Snyder Jr (2015), “Is more information always better? Party cues and candidate quality in U.S. judicial elections”, Journal of Public Economics 128: 107–23.

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

Lim, C S H and J M Snyder Jr (forthcoming), “What Shapes the Quality and Behavior of Government Officials? Institutional Variation in Selection and Retention Methods”, Annual Review of Economics 13.

Martinez-Bravo, M (2014), “The role of local officials in new democracies: evidence from Indonesia”, American Economic Review 104: 1244–87.

315 Reads