Measuring constitutional loyalty

Gutmann, Jerg; Sarel, Roee; Voigt, Stefan Article — Published Version Measuring constitutional loyalty Public Choice Provided in Cooperation with: Springer Nature Suggested Citation: Gutmann, Jerg; Sarel, Roee; Voigt, Stefan (2025) : Measuring constitutional loyalty, Public Choice, ISSN 1573-7101, Springer US, New York, NY, Vol. 205, Iss. 1-2, pp. 1-18, https://doi.org/10.1007/s11127-025-01271-8 This Version is available at: https://hdl.handle.net/10419/330403 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Received: 11 September 2024 / Accepted: 20 February 2025 / Published online: 8 March 2025 © The Author(s) 2025 Jerg Gutmann [email protected] Roee Sarel [email protected] Stefan Voigt [email protected] 1 Institute of Law and Economics, University of Hamburg, Alsterterrasse 1, D-20354 Hamburg, Germany 2 CESifo, Munich, Germany Measuring constitutional loyalty JergGutmann1,2 · RoeeSarel1· StefanVoigt1,2 Public Choice (2025) 205:1–18 https://doi.org/10.1007/s11127-025-01271-8 Abstract We introduce a new concept, constitutional loyalty, which we define as the importance citizens ascribe to their government’s compliance with constitutional rules. Its measurement across countries is challenging due to differences in context, history, and culture. We overcome this challenge by exploiting the COVID-19 pandemic as a setting in which societies around the world face unfamiliar and similar public health challenges, inducing governments to adopt comparable, yet legally untested policies. Based on novel data from a global online survey collected between 2021 and 2022, we show that citizens’ support for COVID-19 mitigation policies declines if a court signals doubts about their constitutionality. We further demonstrate that this effect of constitutional loyalty depends on citizens’ characteristics, such as their confidence in the courts and their moral convictions. Keywords Constitutional loyalty · COVID-19 · Judicial power · Moral foundations · Trust in the judiciary JEL classification D02 · H12 · I18 · K40 · P48 1 Introduction Constitutions are legal devices that are meant to constrain governments. Yet, the degree to which governments comply with constitutional constraints varies significantly across countries and over time (Gutmann et al., 2024b). One factor that could determine the size of this 1 3 Public Choice (2025) 205:1–18 “de jure-de facto gap” is the extent to which citizens expect politicians to comply with their constitution.1 If citizens perceive the constitution as the legitimate foundation of their political and legal system, they are unlikely to tolerate government actions that violate constitutional rules. Where this “constitutional loyalty” is more pronounced, the government should be less inclined to overstep constitutional constraints, because it would have to expect costly opposition. The courts play an important role in this process by dispersing information on governments’ compliance with the constitution. By declaring that a constitutional rule has (or might have) been breached, courts can help citizens in coordinating their opposition (see, e.g., Weingast, 1997). An empirical evaluation of this mechanism is fraught with challenges. One important challenge is that constitutional loyalty is needed precisely when governments are most tempted to transgress constitutional constraints and the timing and nature of such episodes of constitutional stress vary across countries, making them difficult to pinpoint and compare.2 The COVID-19 pandemic, however, is an exception. It hit most countries at about the same time and gave rise to very similar challenges for policymakers all over the world. In this article, we leverage the pandemic as an exogenous event that facilitates the measurement of constitutional loyalty, operationalized as the propensity of citizens to reject a policy for the simple reason that it most likely violates the country’s constitution. Specifically, we devised a global survey, which elicits support for three typical COVID19 mitigation policies. We then asked respondents whether their support would change if the highest court in their country had signaled—with different levels of certainty—that the respective policies violate the constitution. Furthermore, the survey elicits three key attributes of each respondent: (1) their degree of confidence in the courts, (2) whether they have a law degree, and (3) their fairness concerns. This study design enables us to analyze whether respondents’ support for a policy is changed by a signal from a high court indicating that a policy is probably (or clearly) unconstitutional. Moreover, we are able to evaluate whether individual respondent characteristics affect how responsive subjects are to the court signal. Considering such conditional effects is important for understanding the mechanisms behind individuals’ response to court signals. Therefore, we use a within-subject design in the survey, where each respondent is consecutively exposed to a baseline and different signals (see Czibor et al., 2019). Our survey respondents were recruited via MTurk between March 2021 and April 2022. The resulting nonrepresentative global sample includes 1,080 respondents, distributed across 58 democratic countries. Our study contributes to a recent and still rather small stream of literature that asks how public support for policies is affected by their constitutionality (see Cope & Crabtree, 2022 for an overview).3 This literature yields mixed evidence. For instance, Chilton and Versteeg (2016) find a sizable, but statistically insignificant, decline in support for the use of torture among participants in the US when they are told that torture is banned by constitutional law. This finding is not too surprising, given that most US residents probably know that the US 1 See, e.g., Gutmann et al., 2024b); Metelska-Szaniawska (2021); Voigt (2021) for a discussion of the concept of constitutional compliance and its measurement. 2 Bjørnskov and Gutmann (2024); Choutagunta et al. (2024); Gutmann et al. (2024a) demonstrate that such episodes might be triggered, e.g., by the occurrence of extreme events or by specific types of political leaders. 3 A somewhat related literature deals with the same question regarding international law (see, e.g., Cope & Crabtree, 2020, 2022). 1 3 2 Public Choice (2025) 205:1–18 constitution prohibits torture. In another study, Chilton and Versteeg (2020) tell Turkish participants that constitutional law experts consider a Wikipedia ban to be in violation of Turkey’s constitution. They do not find a decline in support for that policy – supporters of the ruling party even increase their support. Cope and Crabtree (2022) measure how public support for an immigration policy in the US shifts when participants learn that experts consider it a violation of constitutional due process rights. Their study was conducted twice: First, in 2018, respondents treated with this information showed increased support for the policy, but this was driven by supporters of President Trump. Second, a replication in November 2020 finds no significant effect. Interestingly, Cope and Crabtree show that the treatment leads only 3% of respondents to update their belief that the law is unconstitutional. This casts doubt on the idea that public opinion can be swayed by referring to the beliefs of an unspecified group of experts. Closely related to our study, Chilton et al. (2025) measure support for COVID-19 mitigation policies in the US, Japan, Israel, South Korea, Taiwan, and China during the early days of the pandemic. Using a between-subject experimental treatment, they find that some survey participants decrease their support for some policies if they are told that many legal experts consider these policies unconstitutional. However, in China, policy support tends to increase. In sum, this literature finds that individuals often do not reduce their support for a policy after being told that experts consider the policy unconstitutional. Supporters of the government, in particular, sometimes even increase their support for the policy. Cope (2023) argues that these weak treatment effects and potential ‘backfire effects’ are due to subjects’ entrenched priors about the constitutionality of the policy in question. In other words, the problem is either that subjects have already formed a strong opinion regarding the studied policy’s constitutionality or that the information treatment is not compelling enough for them to change their opinion. Either way, subjects then do not change their belief about the policy’s constitutionality, giving them no reason not to support it anymore. Our study makes three key contributions to this literature. First, we collect novel survey data from a global sample of over one thousand participants.4 Second, we study a signal sent by a court (rather than by legal experts) and vary the signal’s strength. This enables us to test whether individuals change their attitude based on the courts’ evaluation of whether a policy violates the constitution. Arguably, this is more relevant in practice than the ability of some undefined legal experts to sway public opinion. After all, it is the task of the courts to determine and inform the public of whether a political action violates the law or the constitution. Third, by collecting information on the attributes of our respondents—notably, their confidence in the courts and their fairness concerns—we are able to investigate whether there are heterogeneous treatment effects, that is, whether some citizens are more susceptible to a court signal appealing to constitutional loyalty than others. Exploring heterogeneity among subjects seems especially useful if, as Cope (2023) suggests, some individuals are treatment resistant. Most of our participants are unlikely to have strong entrenched opinions on the 4 As the validity of our theoretical arguments might be limited to democratic political systems, our survey is carried out only in democracies as classified by Bjørnskov and Rode (2020). The 1,080 respondents in our survey are distributed unequally across countries, as can be seen in Table A.2 in the Appendix. Many countries in our sample are only represented by few respondents. Thus, we cannot infer the specific treatment effect in these countries. However, due to the global dispersion of our survey respondents, our results are more robust to country-specific idiosyncrasies than the results of previous studies, which relied only on a single country or a small number of countries, but with a large number of respondents per country. 1 3 3 Public Choice (2025) 205:1–18 constitutionality of COVID-19 mitigation policies, as these were new policy instruments designed to deal with a new political challenge. Still, they might not trust the source of the information provided in the treatment. Participants might either not trust legal experts in general or they might question our motives in selecting experts.5 By referring to a court, we can avoid these challenges faced in previous studies. There is only one highest court in the country, leaving no discretion to the experimenter as in choosing relevant legal experts. Moreover, based on participants’ reported trust in the judiciary, we can measure the credibility of the court signal. If some individuals do not trust the courts because judges are perceived as being partisan, corrupt, or incompetent, we can account for this in our analysis. Our analysis yields several novel findings. We show that respondents, on average, reduce their support for COVID-19 mitigation policies by 25–50% of a standard deviation if their country’s highest court indicates that they are—or are deemed likely to be—unconstitutional. Linear probability models (see footnote 17) yield a more intuitive interpretation of our estimated effect size: We find that a probable (clear) violation signal causes 13% (18%) of the respondents to switch from positive support (5 or higher on a 7-point Likert scale) to non-support (4 or lower). In other words, positive support declines from 68% of all participants to 55% and 50%, respectively. Apart from the general importance of demonstrating the concept of constitutional loyalty empirically, this finding is particularly interesting from a political economy perspective. It suggests that courts can sway public opinion against government policies by invoking citizens’ constitutional loyalty. In other words, constitutional courts not only constrain governments directly—through their authority to decide on constitutional matters—but also indirectly, by influencing public opinion and coordinating collective action. Moreover, our analysis shows that the effect size depends on individual citizens’ (i) confidence in the courts and (ii) their moral values. Thus, not all citizens exhibit the same degree of constitutional loyalty, and some are more responsive than others to a court’s signal of unconstitutionality. This implies that the courts’ ability to constrain the government depends on the citizens’ traits and that the courts must maintain high public confidence to be able to harness constitutional loyalty. The remainder of the article is organized as follows: in Sect. 2, we develop our theory and hypotheses. Here, we formalize the concept of constitutional loyalty. Section 3 describes our survey design and distribution. Empirical results are presented and discussed in Sect. 4. Section 5 concludes and outlines possible future research. 2 Theory and hypotheses While it may be apparent why individuals tend to oppose violations of many human rights (Allendoerfer, 2017; Arı & Sonmez, 2025; Tomz & Weeks, 2020), it is less obvious and commonly accepted that the degree of opposition depends on whether those rights are protected by a constitution (e.g., Kantorowicz, 2023). Our basic conjecture in this article is that many people oppose breaches of constitutional law, independent of their immediate con5 Chilton et al. (2025) discuss the ambiguity of referring to legal experts as a potential downside of their treatment in the Chinese context, where they find more support for policies when experts consider them unconstitutional. Such a backlash may arise if experts are motivated, or at least considered to be motivated, by partisanship (see also Braman, 2023). 1 3 4 Public Choice (2025) 205:1–18 sequences. We refer to an individual’s propensity to oppose governmental actions merely because they are unconstitutional as constitutional loyalty. While we are not the first to think about this phenomenon, as is illustrated by our discussion of the literature above, we are unaware of anyone defining and formalizing the concept in question.6 We are also not the first to use the term constitutional loyalty. Fontana and Huq (2018), for example, use the term to indicate that a public official acts in accordance with constitutional law. Such a use of the term, however, devalues the meaning of loyalty, which is commonly understood to involve an element of genuine support. Our proposed definition of constitutional loyalty, in contrast, is based on the most common interpretation of the term loyalty, as it assumes an intrinsic motivation of those who are willing to incur a cost to uphold the constitutional order. 7 Now, how does an individual determine which actions are unconstitutional? Sometimes it is straightforward to identify the violation of a constitutional rule, but in many cases the correct interpretation of the constitution requires expertise. The authority to determine whether an action is unconstitutional then lies with a court. In most countries, the relevant court can rule on legislation or government practices and declare them unconstitutional. However, the court can also offer weaker signals of constitutionality. For instance, the Israeli Supreme Court sometimes issues an “invalidity notice” for problematic legal acts, whereby the act remains intact, but the government is warned that it might be declared unconstitutional if the infringement of rights persists (see, e.g., Bendor, 2020). More generally, written opinions and dissenting opinions occasionally hint at judges’ expectation that certain practices might not hold up against constitutional review. Finally, even oral arguments can offer an informative preview of how judges will decide a case or are likely to decide other cases. Based on these insights, we formulate two connected hypotheses: H1 Receiving a signal from a court that a policy is unconstitutional reduces individuals’ support for the respective policy. H2 A stronger signal that a policy is unconstitutional reduces individuals’ support for the respective policy more than a weaker signal. Both H1 and H2 are based on the notion that individuals who are loyal to the constitution will demonstrate lower support for unconstitutional actions. Thus, upon learning that a court 6 Our definition of constitutional loyalty differs from the concept of “constitutional approval” discussed by Stephanopoulos and Versteeg (2016). The latter is based on people’s stated approval of their constitution. Stephanopoulos and Versteeg explain that constitutional approval implies specific rather than diffuse support (see Easton, 1975), because approval is not abstract. It is based on the concrete content and effects of the constitution. Approval can be based on the representation of the will of the people in the constitution, or it is grounded in reason and justice being reflected in constitutional rules (Harel & Shinar, 2023). Constitutional loyalty, in contrast to constitutional approval, refers to diffuse support or the perceived legitimacy of the constitution as such, i.e., independent of the concrete content of the document. 7 While subjects do not incur a monetary cost when changing their opinion, a psychological/cognitive cost is involved. Individuals who change their mind in light of new information, for example, have to invest energy (Stone et al., 2022). Furthermore, Falk et al. (2018) show that attitudes elicited through surveys are frequently consistent with preferences expressed under conditions where subjects face tangible incentives. In practice, the cost of enforcing the constitution can vary. It ranges from voting for a party whose policies are compatible with the constitution to risking one’s life in protesting against a dictator who violates the constitution. Thus, the willingness to incur small costs to protect the constitution can have important consequences at least in some situations. 1 3 5 Public Choice (2025) 205:1–18 considers a policy potentially unconstitutional, individuals will lower their support (H1) and more so if the signal of unconstitutionality is stronger (H2). So far, we have made an implicit assumption that court signals regarding constitutionality convey credible information. However, according to Caldeira (1986), the courts’ ability to appeal to citizens’ constitutional loyalty should depend on the level of confidence they enjoy in the population. In other words, individuals who do not have confidence in the courts may not respond as strongly to a court signal as those who do trust the courts. This conditional relationship yields our third hypothesis: H3 Individuals with higher confidence in the courts reduce their support for the respective policy more upon receiving a court signal that it is unconstitutional than those with lower confidence in the courts. Another factor which may influence an individual’s responsiveness to a court signal is legal education. There is evidence suggesting that different legally trained groups are overconfident in their abilities (Eigen & Listokin, 2012; Franck et al., 2017; Goodman-Delahunty et al., 2010). Such overconfidence may lead them to believe that—as legal experts—they do not depend on a court for legal interpretation. According to this argument, a court’s signal would not have much effect on the beliefs of those with a legal education.8 H4 Individuals with a law degree reduce their support for the respective policy less upon receiving a court signal that it is unconstitutional than those without a law degree. For our final hypothesis, we consider the role of individuals’ fairness concerns, as constitutional loyalty is likely grounded in an individual’s broader ethical convictions. Moral Foundations Theory (MFT; see Haidt, 2012; Graham et al., 2013) is the most well-established attempt to explain interpersonal variation in human moral reasoning based on a set of historically evolved cultural traits. MFT names fairness as one fundamental dimension of human morality. As the fairness foundation supports cooperation and reciprocity, while encouraging opposition to unequal treatment of individuals, cheating, and rule violations, it is likely to predispose individuals to show constitutional loyalty. This brings us to our fifth and final hypothesis: H5 Individuals with a stronger fairness-based morality reduce their support for the respective policy more upon receiving a court signal that it is unconstitutional than those with weaker fairness-based morality. Overall, our hypotheses predict that constitutional loyalty leads individuals to respond to a court signal and that the size of this effect depends on the strength of the signal, individuals’ level of confidence in the courts, their legal education, and their fairness-based morality. A court signal of unconstitutionality causes individuals that are loyal to the constitution to reduce their support for the policy in question. The reduction in support is larger if the court 8 A plausible counterargument, though lacking empirical evidence, assumes that legally trained individuals have internalized a greater appreciation of constitutional compliance. Consequently, they would reduce their support for the respective policy more upon receiving a court signal that it is unconstitutional than those without a law degree. 1 3 6 Public Choice (2025) 205:1–18 signal is stronger. Individuals who are confident in the courts, who have no law degree, or whose morality is fairness-based show a stronger (negative) reaction to the court signal. 3 Survey design and distribution 3.1 Survey design Our survey design relies on a simple within-subject treatment:9 we ask subjects to rate their level of support for COVID-19 mitigation policies (on a 7-point Likert scale) under three conditions, a baseline and two treatments. In the baseline condition, we simply ask to what degree each subject supports the policy. Then, in the probable violation condition, we ask for their level of support in case the highest relevant court in their country would decide that the policy probably violates a constitutional right. Finally, in the clear violation condition, we elicit the level of support in case the highest court would decide that the policy clearly violates a constitutional right. Hence, our treatment conditions only differ in terms of the incrementally increasing signal (none vs. weak vs. strong) respondents receive from the court, which indicates that the policy might be unconstitutional. This setup is equivalent to the strategy method widely used in economic experiments (see, e.g., Brandts & Charness, 2011; Fischbacher et al., 2012), as we elicit the decisions of respondents in different hypothetical situations.10 To ensure that our results are not policy-specific, we elicit the level of support for three common mitigation policies: outdoor facial mask mandates, prohibitions of going on vacation abroad, and shutdowns of non-essential businesses. The mitigation policies are presented in a randomized order, but the treatment conditions are always in the aforementioned order. We also collect information on basic demographics, confidence in various actors involved in fighting the pandemic, and answers to selected questions from the Moral Foundations Questionnaire (MFQ; see Graham et al., 2011). To test our hypotheses, we rely on three variables of interest, aside from our treatment indicator: confidence in the courts, a dummy variable for whether the subject has a law degree, and how much the subject relies on the fairness foundation in their moral judgments. The confidence variable is measured on a 4-point Likert scale with higher values reflecting more confidence in courts. The fairness variable is derived from an MFQ question asking the subject how important it is for their moral judgment whether “someone was denied his or her rights”, measured on a 6-point Likert scale where higher values correspond to stronger fairness concerns. Table A.1 in the Appendix describes all variables. 9 A within-subject design is ideal to test our hypotheses H3 to H5, as these require an interaction between individual attributes and the treatments (see, e.g., Czibor et al., 2019). A between-subject design would be costly in terms of efficiency and is less suitable for estimating treatment effects conditional on respondent characteristics. 10 Fischbacher et al. (2012) find that those who demonstrate conditional cooperation under the strategy method do so also under the direct-response model, suggesting that both methods yield similar results. Brandts and Charness (2011) also find no difference between the strategy method and the direct-response model for the majority of the experiments they consider. 1 3 7 Public Choice (2025) 205:1–18 3.2 Survey distribution The survey was programmed using the standard software Qualtrics and distributed via Amazon’s MTurk platform, which is widely used in online experiments (Clifford et al., 2015; Horton et al., 2011; Johnson & Ryan, 2020; Macdonald, 2024; Peer et al., 2017; Stritch et al., 2024).11 Three state-of-the-art precautions were taken to ensure high data quality: First, the survey includes three attention checks (e.g., an instructional manipulation check). Respondents who failed an attention check were excluded.12 Second, we used an algorithm to identify and exclude users of VPNs, which our survey instructions explicitly prohibit. This prevents individuals from misreporting their location and is used to filter out fraudulent participants (Kennedy et al., 2020). Third, we recruited only subjects with a positive track record (i.e., at least 100 previously completed tasks and at least 95% of these tasks were approved by the task administrators). Data collection took place between March 2021 and April 2022 to ensure a sufficient sample size. While there is evidence that participants’ exposure to different levels of infection risk can affect their attitude towards government interventions (Alsan et al., 2023), our study design eliminates the need to consider these differences, because we are only interested in how court signals change a given individual’s policy support. Thus, the level of an individual’s exposure is held constant in our comparisons by controlling for respondent fixed effects and we do not need to control for differences in countries’ affectedness by the pandemic or for variation in affectedness over time. Our nonrepresentative global sample includes 1,080 respondents from 58 democratic countries.13 Because we ask each subject about their support for three policies under three conditions (the baseline and two treatments), this yields a total of 3×3×1080 = 9,720 observations. Table A.2 in the Appendix gives an overview of our sample’s country coverage. The number of observations is not equal across countries, but this is not problematic in a within-subject design (see also footnote 4). 4 Results 4.1 Summary statistics Table 1 presents summary statistics. Average support for mitigation policies lies between 3.6 and 5.5 on a scale from 1 to 7 and the mean is reduced in the probable and clear violation conditions relative to the baseline. Although our sample includes respondents of every adult age, the average respondents are only in their mid-30s, which is representative of the pool of MTurk workers.14 Some 30% of the respondents are female. 5% hold a law degree. 11 Some scholars have voiced concerns about the quality of data collected on MTurk (e.g., Ahler et al., 2025; Douglas et al., 2023). However, alternative micro-task markets do not offer a global country coverage comparable to that of MTurk. If random measurement error was a serious problem in our data collection, our results would be biased toward zero. 12 Note that the concerns voiced by Montgomery et al. (2018) regarding post-treatment attention checks concern only between-subject designs. 13 Our sample is neither representative of the global nor of any national population. It is not our goal to measure the effect size in these populations. As in many standard economic experiments, our goal is only to demonstrate the existence of a social phenomenon – constitutional loyalty. 14 One respondent reported an unlikely age of 99, the second-oldest person is 75. 1 3 8 Public Choice (2025) 205:1–18 than expert signals, policies, timeframes, and countries), we cannot determine conclusively what causes this discrepancy. However, our study appears to have some advantages with respect to the treatment resistance identified and studied by Cope (2023). Policymakers can conclude from our study that when a high court, such as the US Supreme Court, enjoys strong public support, it can rely on citizens’ constitutional loyalty to effectively constrain the government. Stoutenborough and Haider-Markel (2008), however, show that confidence in the US Supreme Court varies over time and Vigers and Saad (2024) report that confidence in the US Supreme Court has reached a record low in 2024. When confidence in the Supreme Court declines, it becomes more difficult for the Court to enforce its decisions vis-à-vis the other branches of government. Our findings are also consistent with political events in Germany during the COVID-19 pandemic. Some political parties (e.g., the Free Democratic Party) were reluctant to adopt strict COVID-19 mitigation policies. However, the German constitutional court ruled on November 19, 2021 that the mitigation policies adopted in the spring of 2021 were constitutional. Subsequently, politicians from the new governing coalition—some of whom even brought the previous government to court over its strict mitigation policies—quickly signaled their willingness to adopt harsher mitigation policies to curb the fourth wave of the pandemic. As a final example, consider the initiative of the Israeli government starting in late 2022 to revise the country’s Basic Laws (see Salzberger, 2023 for an overview of these political events). As part of this initiative, the government introduced an amendment that would allow the appointment of a particular person, who was previously convicted of criminal offenses, as a minister. The Israeli Supreme Court rejected the appointment without striking down the Basic Law amendment.19 Yet, some of the judges indicated in the ruling that the amendment might be unconstitutional. The large demonstrations that erupted in response to the government initiative in general lasted for many months. The Supreme Court’s signal of the amendment’s probable unconstitutionality plausibly provided additional backwind to the demonstrations. We have focused here on how individual support for mitigation policies depends on information about their constitutionality, the perceived reliability of that information, and the traits of the respective individual. Some follow-up questions seem interesting: Does individual support for policies and its responsiveness to their constitutionality depend on an individual’s exposure to the pandemic? Are individuals who question the seriousness of the pandemic or the trustworthiness of public (health) authorities as interested as others in the constitutionality of mitigation policies? Do courts use their signaling power strategically? Does the signal’s effect depend on who is the sender (courts, experts, news media, politicians, etc.)? Does it make a difference if the court does not cite the constitution, but, for example, customary or religious laws as being violated by the mitigation policy? Why are women generally more supportive of COVID-19 mitigation policies? And finally, what country-level factors can explain the existence and strength of constitutional loyalty? We leave these questions for future research. Acknowledgements The authors thank Michael Peneder, Hans Pitlik, four anonymous reviewers and the associate editor of Public Choice, and participants of the following conferences and seminars: Austrian Insti19 Israel Supreme Court Case 8948/22, Schienfeld et al. vs. the Israeli Parliament (“Knesset”), Jan. 5, 2023, in Hebrew, h t t p s : / / i m g . m a k o . c o . i l / 2 0 2 3 / 0 1 / 1 8 / b e g a z _ d e r i _ f u l l . p d f. 1 3 15 Public Choice (2025) 205:1–18 tute of Economic Research’s WIFO research seminar, IFN Stockholm conference on “The Economics of Culture and Institutions”, French Law & Economics Association 2022, European Association of Law & Economics 2022, ICON-S 2022, the European Public Choice Society 2023, and the symposium “Cultures of Trust and Institutions of Freedom” at IFN Stockholm (Nov. 30 – Dec. 3, 2023) for helpful comments and suggestions. Funding Open Access funding enabled and organized by Projekt DEAL. Financial support by the German Research Foundation (#381589259) and the Polish National Science Centre (#2016/23/G/HS4/04371) is gratefully acknowledged. Roee Sarel has received financial support for this project in the form of a symposium honorarium by the John Templeton Foundation (#62065). The opinions expressed are those of the authors and do not necessarily reflect the views of the John Templeton Foundation. Declarations Ethics approval, consent, privacy, and compensation In compliance with their ethical obligations, the authors declare that this research has been subject to review regarding its compliance with ethical standards at the University of Hamburg and that survey respondents have given their consent to the collection of their data and its use for research purposes. All data is collected anonymously, and no apparent harm is to be expected for participants of this survey. Survey participants were paid between 2USD and 3USD for a survey that takes about five to ten minutes to fill in. Respondents from low and lower-middle income economies, according to the World Bank, received 2USD. Those in upper-middle income economies received 2.50USD. Respondents from high income economies received 3USD. This amounts to an hourly wage of 12 (/24) to 18 (/36) USD, depending on the respondent’s country of residence and their speed. Pre-registration The survey has been preregistered at the Center for Open Science (OSF): h t t p s : / / o s f . i o / 7 j 8 a m / fi l e s / o s f s t o r a g e / 6 0 5 b b f 7 2 e 1 2 b 6 0 0 0 2 3 a a 7 9 7 e . Replication materials Replication files for our empirical analysis are available at: h t t p s : / / d o i . o r g / 1 0 . 7 9 1 0 / D V N / S S Q W D D . Competing interests The authors have no potential or perceived conflicts of interest arising from this research. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. References Ahler, D. J., Roush, C. E., & Sood, G. (2025). The micro-task market for lemons: Data quality on Amazon’s mechanical Turk. Political Science Research and Methods, 13(1), 1–20. Allendoerfer, M. G. (2017). Who cares about human rights? Public opinion about human rights foreign policy. Journal of Human Rights, 16(4), 428–451. Alsan, M., Braghieri, L., Eichmeyer, S., Kim, M. J., Stantcheva, S., & Yang, D. Y. (2023). Civil liberties in times of crisis. American Economic Journal: Applied Economics, 15(4), 389–421. Arı, B., & Sonmez, B. (2025). Human rights violations and public support for sanctions. Journal of Peace Research, 62(1), 68–84. Baetschmann, G., Staub, K. E., & Winkelmann, R. (2015). Consistent Estimation of the fixed effects ordered logit model. Journal of the Royal Statistical Society: Series A, 178(3), 685–703. Bendor, A. L. (2020). The Israeli judiciary-centered constitutionalism. International Journal of Constitutional Law, 18(3), 730–745. 1 3 16 Public Choice (2025) 205:1–18 Bjørnskov, C., & Gutmann, J. (2024). Coups and constitutional compliance. Paper presented at the Southern Economic Association Annual Meeting in Washington D.C. Bjørnskov, C., & Rode, M. (2020). Regime types and regime change: A new dataset on democracy, coups, and political institutions. The Review of International Organizations, 15(2), 531–551. Braman, E. (2023). Assessing the credibility of constitutional experts. Journal of Law and Courts, 11(1), 86–103. Brandts, J., & Charness, G. (2011). The strategy versus the direct-response method: A first survey of experimental comparisons. Experimental Economics, 14(3), 375–398. Caldeira, G. A. (1986). Neither the purse nor the sword: Dynamics of public confidence in the supreme court. American Political Science Review, 80(4), 1209–1226. Charness, G., Gneezy, U., & Kuhn, M. A. (2012). Experimental methods: Between-subject and within-subject design. Journal of Economic Behavior & Organization, 81(1), 1–8. Chilton, A. S., & Versteeg, M. (2016). International law, constitutional law, and public support for torture. Research & Politics, 3(1), 1–9. Chilton, A. S., & Versteeg, M. (2020). How constitutional rights matter. Oxford University Press. Chilton, A. S., Cope, K. L., Crabtree, C., & Versteeg, M. (2025). Support for constitutional rights during crisis: Evidence from the pandemic. American Journal of Comparative Law, forthcoming. Choutagunta, A., Gutmann, J., & Voigt, S. (2024). Shocking resilience? Effects of extreme events on constitutional compliance. Journal of Institutional Economics, 20, e3. Clifford, S., Jewell, R. M., & Waggoner, P. D. (2015). Are samples drawn from mechanical Turk valid for research on political ideology? Research & Politics, 2(4), 205316801562207. Cope, K. L. (2023). Measuring law’s normative force. Journal of Empirical Legal Studies, 20(4), 1005–1044. Cope, K. L., & Crabtree, C. (2020). A Nationalist backlash to international refugee law: Evidence from a survey experiment in Turkey. Journal of Empirical Legal Studies, 17(4), 752–788. Cope, K. L., & Crabtree, C. (2022). Migrant-family separation and higher-order laws’ diverging normative force. The Journal of Legal Studies, 51(2), 403–426. Czibor, E., Jimenez-Gomez, D., & List, J. A. (2019). The dozen things experimental economists should do (more of). Southern Economic Journal, 86(2), 371–432. Douglas, B. D., Ewell, P. J., & Brauer, M. (2023). Data quality in online human-subjects research: Comparisons between MTurk, prolific, CloudResearch, qualtrics, and SONA. PLOS ONE, 18(3), e0279720. Easton, D. (1975). A re-assessment of the concept of political support. British Journal of Political Science, 5(4), 435–457. Eigen, Z. J., & Listokin, Y. (2012). Do lawyers really believe their own hype, and should they? A natural experiment. The Journal of Legal Studies, 41(2), 239–267. Falk, A., Becker, A., Dohmen, T., Enke, B., Huffman, D., & Sunde, U. (2018). Global evidence on economic preferences. Quarterly Journal of Economics, 133(4), 1645–1692. Fischbacher, U., Gächter, S., & Quercia, S. (2012). The behavioral validity of the strategy method in public good experiments. Journal of Economic Psychology, 33(4), 897–913. Fontana, D., & Huq, A. (2018). Institutional loyalties in constitutional law. University of Chicago Law Review, 85(1), 1–84. Franck, S. D., van Aaken, A., Freda, J., Guthrie, C., & Rachlinski, J. J. (2017). Inside the arbitrator’s Mind. Emory Law Journal, 66(5), 1115–1174. Goodman-Delahunty, J., Granhag, P. A., Hartwig, M., & Loftus, E. F. (2010). Insightful or wishful: Lawyers’ ability to predict case outcomes. Psychology Public Policy and Law, 16, 133–157. Graham, J., Haidt, J., Koleva, S., Motyl, M., Iyer, R., Wojcik, S. P., & Ditto, P. H. (2013). Moral foundations theory: The pragmatic validity of moral pluralism. Advances in Experimental Social Psychology, 47, 55–130. Graham, J., Nosek, B. A., Haidt, J., Iyer, R., Koleva, S., & Ditto, P. H. (2011). Mapping the moral domain. Journal of Personality and Social Psychology, 101(2), 366–385. Gutmann, J., Metelska-Szaniawska, K., & Voigt, S. (2024a). Leader characteristics and constitutional compliance. European Journal of Political Economy, 84, 102423. Gutmann, J., Metelska-Szaniawska, K., & Voigt, S. (2024b). The comparative constitutional compliance database. The Review of International Organizations, 19(1), 95–105. Haidt, J. (2012). The righteous Mind: Why good people are divided by politics and religion. Pantheon. Harel, A., & Shinar, A. (2023). Two concepts of constitutional legitimacy. Global Constitutionalism, 12(1), 80–105. Horton, J. J., Rand, D. G., & Zeckhauser, R. J. (2011). The online laboratory: Conducting experiments in a real labor market. Experimental Economics, 14(3), 399–425. Johnson, D., & Ryan, J. B. (2020). Amazon mechanical Turk workers can provide consistent and economically meaningful data. Southern Economic Journal, 87(1), 369–385. 1 3 17 Public Choice (2025) 205:1–18 Kantorowicz, J. (2023). Testing public reaction to constitutional fiscal rules violations. Constitutional Political Economy, 34(4), 483–509. Kennedy, R., Clifford, S., Burleigh, T., Waggoner, P. D., Jewell, R., & Winter, N. J. G. (2020). The shape of and solutions to the MTurk quality crisis. Political Science Research and Methods, 8(4), 614–629. Macdonald, D. (2024). Political trust and American public support for free trade. Political Behavior, 46(2), 1037–1055. Metelska-Szaniawska, K. (2021). Post-socialist constitutions: The de jure–de facto Gap, its effects and determinants. Economics of Transition and Institutional Change, 29(2), 175–196. Montgomery, J. M., Nyhan, B., & Torres, M. (2018). How conditioning on posttreatment variables can ruin your experiment and what to do about it. American Journal of Political Science, 62(3), 760–775. Peer, E., Brandimarte, L., Samat, S., & Acquisti, A. (2017). Beyond the Turk: Alternative platforms for crowdsourcing behavioral research. Journal of Experimental Social Psychology, 70, 153–163. Salzberger, E. (2023). A Possible Regime Change in Israel. Verfassungsblog: On Matters Constitutional. h t t p s : / / v e r f a s s u n g s b l o g . d e / r e g i m e - c h a n g e - i s r a e l / (December 15, 2024). Stephanopoulos, N. O., & Versteeg, M. (2016). The contours of constitutional approval. Washington University Law Review, 94(1), 113–190. Stone, C., Mattingley, J. B., & Rangelov, D. (2022). On second thoughts: Changes of Mind in decisionmaking. Trends in Cognitive Sciences, 26(5), 419–431. Stoutenborough, J. W., & Haider-Markel, D. P. (2008). Public confidence in the U.S. Supreme court: A new look at the impact of court decisions. The Social Science Journal, 45(1), 28–47. Stritch, J. M., Pedersen, M. J., & Pezo, I. (2024). Crowdsourced data in public administration research: A review and look to the future. Public Administration Review, forthcoming. Tomz, M. R., & Weeks, J. L. P. (2020). Human rights and public support for war. Journal of Politics, 82(1), 182–194. Vigers, B., & Saad, L. (2024). Americans pass judgment on their courts: Sharp decline in confidence in judiciary is among the largest Gallup has ever measured. h t t p s : / / n e w s . g a l l u p . c o m / p o l l / 6 5 3 8 9 7 / a m e r i c a n s - p a s s - j u d g m e n t - c o u r t s . a s p x (December 17, 2024). Voigt, S. (2021). Mind the Gap: Analyzing the divergence between constitutional text and constitutional reality. International Journal of Constitutional Law, 19(5), 1778–1809. Weingast, B. R. (1997). The political foundations of democracy and the rule of law. American Political Science Review, 91(2), 245–263. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. 1 3 18