Gu mann, Je g; Sa el, Roee; Voig , S e an
A icle — Published Ve sion
Measu ing cons i u ional loyal y
Public Choice
P o ided in Coope a ion wi h:
Sp inge Na u e
Sugges ed Ci a ion: Gu mann, Je g; Sa el, Roee; Voig , S e an (2025) : Measu ing cons i u ional
loyal y, Public Choice, ISSN 1573-7101, Sp inge US, New Yo k, NY, Vol. 205, Iss. 1-2, pp. 1-18,
h ps://doi.o g/10.1007/s11127-025-01271-8
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/330403
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h ps://c ea i ecommons.o g/licenses/by/4.0/
Recei ed: 11 Sep embe 2024 / Accep ed: 20 Feb ua y 2025 / Published online: 8 Ma ch 2025
© The Au ho (s) 2025
Je g Gu mann
je g.gu mann@uni-hambu g.de
Roee Sa el
[email p o ec ed]
S e an Voig
[email p o ec ed]
1 Ins i u e o Law and Economics, Uni e si y o Hambu g, Als e e asse 1, D-20354 Hambu g,
Ge many
2 CESi o, Munich, Ge many
Measu ing cons i u ional loyal y
Je gGu mann1,2 · RoeeSa el1· S e anVoig 1,2
Public Choice (2025) 205:1–18
h ps://doi.o g/10.1007/s11127-025-01271-8
Abs ac
We in oduce a new concep , cons i u ional loyal y, which we de ine as he impo ance
ci izens asc ibe o hei go e nmen ’s compliance wi h cons i u ional ules. I s measu e-
men ac oss coun ies is challenging due o di e ences in con ex , his o y, and cul u e.
We o e come his challenge by exploi ing he COVID-19 pandemic as a se ing in which
socie ies a ound he wo ld ace un amilia and simila public heal h challenges, inducing
go e nmen s o adop compa able, ye legally un es ed policies. Based on no el da a om
a global online su ey collec ed be ween 2021 and 2022, we show ha ci izens’ suppo
o COVID-19 mi iga ion policies declines i a cou signals doub s abou hei cons i-
u ionali y. We u he demons a e ha his e ec o cons i u ional loyal y depends on
ci izens’ cha ac e is ics, such as hei con idence in he cou s and hei mo al con ic ions.
Keywo ds Cons i u ional loyal y · COVID-19 · Judicial powe · Mo al ounda ions ·
T us in he judicia y
JEL classi ica ion D02 · H12 · I18 · K40 · P48
1 In oduc ion
Cons i u ions a e legal de ices ha a e mean o cons ain go e nmen s. Ye , he deg ee o
which go e nmen s comply wi h cons i u ional cons ain s a ies signi ican ly ac oss coun-
ies and o e ime (Gu mann e al., 2024b). One ac o ha could de e mine he size o his
1 3
Public Choice (2025) 205:1–18
“de ju e-de ac o gap” is he ex en o which ci izens expec poli icians o comply wi h hei
cons i u ion.1 I ci izens pe cei e he cons i u ion as he legi ima e ounda ion o hei poli i-
cal and legal sys em, hey a e unlikely o ole a e go e nmen ac ions ha iola e cons i u-
ional ules. Whe e his “cons i u ional loyal y” is mo e p onounced, he go e nmen should
be less inclined o o e s ep cons i u ional cons ain s, because i would ha e o expec cos ly
opposi ion. The cou s play an impo an ole in his p ocess by dispe sing in o ma ion on
go e nmen s’ compliance wi h he cons i u ion. By decla ing ha a cons i u ional ule has
(o migh ha e) been b eached, cou s can help ci izens in coo dina ing hei opposi ion
(see, e.g., Weingas , 1997).
An empi ical e alua ion o his mechanism is augh wi h challenges. One impo an
challenge is ha cons i u ional loyal y is needed p ecisely when go e nmen s a e mos
emp ed o ansg ess cons i u ional cons ain s and he iming and na u e o such episodes
o cons i u ional s ess a y ac oss coun ies, making hem di icul o pinpoin and com-
pa e.2 The COVID-19 pandemic, howe e , is an excep ion. I hi mos coun ies a abou he
same ime and ga e ise o e y simila challenges o policymake s all o e he wo ld. In
his a icle, we le e age he pandemic as an exogenous e en ha acili a es he measu e-
men o cons i u ional loyal y, ope a ionalized as he p opensi y o ci izens o ejec a policy
o he simple eason ha i mos likely iola es he coun y’s cons i u ion.
Speci ically, we de ised a global su ey, which elici s suppo o h ee ypical COVID-
19 mi iga ion policies. We hen asked esponden s whe he hei suppo would change i
he highes cou in hei coun y had signaled—wi h di e en le els o ce ain y— ha he
espec i e policies iola e he cons i u ion. Fu he mo e, he su ey elici s h ee key a i-
bu es o each esponden : (1) hei deg ee o con idence in he cou s, (2) whe he hey
ha e a law deg ee, and (3) hei ai ness conce ns. This s udy design enables us o analyze
whe he esponden s’ suppo o a policy is changed by a signal om a high cou indica -
ing ha a policy is p obably (o clea ly) uncons i u ional. Mo eo e , we a e able o e alua e
whe he indi idual esponden cha ac e is ics a ec how esponsi e subjec s a e o he cou
signal. Conside ing such condi ional e ec s is impo an o unde s anding he mechanisms
behind indi iduals’ esponse o cou signals. The e o e, we use a wi hin-subjec design
in he su ey, whe e each esponden is consecu i ely exposed o a baseline and di e en
signals (see Czibo e al., 2019). Ou su ey esponden s we e ec ui ed ia MTu k be ween
Ma ch 2021 and Ap il 2022. The esul ing non ep esen a i e global sample includes 1,080
esponden s, dis ibu ed ac oss 58 democ a ic coun ies.
Ou s udy con ibu es o a ecen and s ill a he small s eam o li e a u e ha asks how
public suppo o policies is a ec ed by hei cons i u ionali y (see Cope & C ab ee, 2022
o an o e iew).3 This li e a u e yields mixed e idence. Fo ins ance, Chil on and Ve s eeg
(2016) ind a sizable, bu s a is ically insigni ican , decline in suppo o he use o o u e
among pa icipan s in he US when hey a e old ha o u e is banned by cons i u ional law.
This inding is no oo su p ising, gi en ha mos US esiden s p obably know ha he US
1 See, e.g., Gu mann e al., 2024b); Me elska-Szaniawska (2021); Voig (2021) o a discussion o he con-
cep o cons i u ional compliance and i s measu emen .
2 Bjø nsko and Gu mann (2024); Chou agun a e al. (2024); Gu mann e al. (2024a) demons a e ha such
episodes migh be igge ed, e.g., by he occu ence o ex eme e en s o by speci ic ypes o poli ical lead-
e s.
3 A somewha ela ed li e a u e deals wi h he same ques ion ega ding in e na ional law (see, e.g., Cope &
C ab ee, 2020, 2022).
1 3
2
Public Choice (2025) 205:1–18
cons i u ion p ohibi s o u e. In ano he s udy, Chil on and Ve s eeg (2020) ell Tu kish pa -
icipan s ha cons i u ional law expe s conside a Wikipedia ban o be in iola ion o Tu -
key’s cons i u ion. They do no ind a decline in suppo o ha policy – suppo e s o he
uling pa y e en inc ease hei suppo . Cope and C ab ee (2022) measu e how public sup-
po o an immig a ion policy in he US shi s when pa icipan s lea n ha expe s conside
i a iola ion o cons i u ional due p ocess igh s. Thei s udy was conduc ed wice: Fi s ,
in 2018, esponden s ea ed wi h his in o ma ion showed inc eased suppo o he policy,
bu his was d i en by suppo e s o P esiden T ump. Second, a eplica ion in No embe
2020 inds no signi ican e ec . In e es ingly, Cope and C ab ee show ha he ea men
leads only 3% o esponden s o upda e hei belie ha he law is uncons i u ional. This
cas s doub on he idea ha public opinion can be swayed by e e ing o he belie s o an
unspeci ied g oup o expe s.
Closely ela ed o ou s udy, Chil on e al. (2025) measu e suppo o COVID-19 mi i-
ga ion policies in he US, Japan, Is ael, Sou h Ko ea, Taiwan, and China du ing he ea ly
days o he pandemic. Using a be ween-subjec expe imen al ea men , hey ind ha some
su ey pa icipan s dec ease hei suppo o some policies i hey a e old ha many legal
expe s conside hese policies uncons i u ional. Howe e , in China, policy suppo ends
o inc ease.
In sum, his li e a u e inds ha indi iduals o en do no educe hei suppo o a policy
a e being old ha expe s conside he policy uncons i u ional. Suppo e s o he go -
e nmen , in pa icula , some imes e en inc ease hei suppo o he policy. Cope (2023)
a gues ha hese weak ea men e ec s and po en ial ‘back i e e ec s’ a e due o subjec s’
en enched p io s abou he cons i u ionali y o he policy in ques ion. In o he wo ds, he
p oblem is ei he ha subjec s ha e al eady o med a s ong opinion ega ding he s udied
policy’s cons i u ionali y o ha he in o ma ion ea men is no compelling enough o
hem o change hei opinion. Ei he way, subjec s hen do no change hei belie abou he
policy’s cons i u ionali y, gi ing hem no eason no o suppo i anymo e.
Ou s udy makes h ee key con ibu ions o his li e a u e. Fi s , we collec no el su ey
da a om a global sample o o e one housand pa icipan s.4 Second, we s udy a signal sen
by a cou ( a he han by legal expe s) and a y he signal’s s eng h. This enables us o es
whe he indi iduals change hei a i ude based on he cou s’ e alua ion o whe he a policy
iola es he cons i u ion. A guably, his is mo e ele an in p ac ice han he abili y o some
unde ined legal expe s o sway public opinion. A e all, i is he ask o he cou s o de e -
mine and in o m he public o whe he a poli ical ac ion iola es he law o he cons i u ion.
Thi d, by collec ing in o ma ion on he a ibu es o ou esponden s—no ably, hei con-
idence in he cou s and hei ai ness conce ns—we a e able o in es iga e whe he he e
a e he e ogeneous ea men e ec s, ha is, whe he some ci izens a e mo e suscep ible o a
cou signal appealing o cons i u ional loyal y han o he s. Explo ing he e ogenei y among
subjec s seems especially use ul i , as Cope (2023) sugges s, some indi iduals a e ea men
esis an . Mos o ou pa icipan s a e unlikely o ha e s ong en enched opinions on he
4 As he alidi y o ou heo e ical a gumen s migh be limi ed o democ a ic poli ical sys ems, ou su ey
is ca ied ou only in democ acies as classi ied by Bjø nsko and Rode (2020). The 1,080 esponden s in
ou su ey a e dis ibu ed unequally ac oss coun ies, as can be seen in Table A.2 in he Appendix. Many
coun ies in ou sample a e only ep esen ed by ew esponden s. Thus, we canno in e he speci ic ea -
men e ec in hese coun ies. Howe e , due o he global dispe sion o ou su ey esponden s, ou esul s
a e mo e obus o coun y-speci ic idiosync asies han he esul s o p e ious s udies, which elied only on
a single coun y o a small numbe o coun ies, bu wi h a la ge numbe o esponden s pe coun y.
1 3
3
Public Choice (2025) 205:1–18
cons i u ionali y o COVID-19 mi iga ion policies, as hese we e new policy ins umen s
designed o deal wi h a new poli ical challenge. S ill, hey migh no us he sou ce o he
in o ma ion p o ided in he ea men . Pa icipan s migh ei he no us legal expe s in
gene al o hey migh ques ion ou mo i es in selec ing expe s.5 By e e ing o a cou ,
we can a oid hese challenges aced in p e ious s udies. The e is only one highes cou in
he coun y, lea ing no disc e ion o he expe imen e as in choosing ele an legal expe s.
Mo eo e , based on pa icipan s’ epo ed us in he judicia y, we can measu e he c ed-
ibili y o he cou signal. I some indi iduals do no us he cou s because judges a e
pe cei ed as being pa isan, co up , o incompe en , we can accoun o his in ou analysis.
Ou analysis yields se e al no el indings. We show ha esponden s, on a e age, educe
hei suppo o COVID-19 mi iga ion policies by 25–50% o a s anda d de ia ion i hei
coun y’s highes cou indica es ha hey a e—o a e deemed likely o be—uncons i u-
ional. Linea p obabili y models (see oo no e 17) yield a mo e in ui i e in e p e a ion o
ou es ima ed e ec size: We ind ha a p obable (clea ) iola ion signal causes 13% (18%)
o he esponden s o swi ch om posi i e suppo (5 o highe on a 7-poin Like scale) o
non-suppo (4 o lowe ). In o he wo ds, posi i e suppo declines om 68% o all pa ici-
pan s o 55% and 50%, espec i ely. Apa om he gene al impo ance o demons a ing
he concep o cons i u ional loyal y empi ically, his inding is pa icula ly in e es ing om
a poli ical economy pe spec i e. I sugges s ha cou s can sway public opinion agains
go e nmen policies by in oking ci izens’ cons i u ional loyal y. In o he wo ds, cons i u-
ional cou s no only cons ain go e nmen s di ec ly— h ough hei au ho i y o decide on
cons i u ional ma e s—bu also indi ec ly, by in luencing public opinion and coo dina ing
collec i e ac ion.
Mo eo e , ou analysis shows ha he e ec size depends on indi idual ci izens’ (i) con-
idence in he cou s and (ii) hei mo al alues. Thus, no all ci izens exhibi he same
deg ee o cons i u ional loyal y, and some a e mo e esponsi e han o he s o a cou ’s sig-
nal o uncons i u ionali y. This implies ha he cou s’ abili y o cons ain he go e nmen
depends on he ci izens’ ai s and ha he cou s mus main ain high public con idence o be
able o ha ness cons i u ional loyal y.
The emainde o he a icle is o ganized as ollows: in Sec . 2, we de elop ou heo y and
hypo heses. He e, we o malize he concep o cons i u ional loyal y. Sec ion 3 desc ibes
ou su ey design and dis ibu ion. Empi ical esul s a e p esen ed and discussed in Sec . 4.
Sec ion 5 concludes and ou lines possible u u e esea ch.
2 Theo y and hypo heses
While i may be appa en why indi iduals end o oppose iola ions o many human igh s
(Allendoe e , 2017; A ı & Sonmez, 2025; Tomz & Weeks, 2020), i is less ob ious and
commonly accep ed ha he deg ee o opposi ion depends on whe he hose igh s a e p o-
ec ed by a cons i u ion (e.g., Kan o owicz, 2023). Ou basic conjec u e in his a icle is ha
many people oppose b eaches o cons i u ional law, independen o hei immedia e con-
5 Chil on e al. (2025) discuss he ambigui y o e e ing o legal expe s as a po en ial downside o hei
ea men in he Chinese con ex , whe e hey ind mo e suppo o policies when expe s conside hem
uncons i u ional. Such a backlash may a ise i expe s a e mo i a ed, o a leas conside ed o be mo i a ed,
by pa isanship (see also B aman, 2023).
1 3
4
Public Choice (2025) 205:1–18
sequences. We e e o an indi idual’s p opensi y o oppose go e nmen al ac ions me ely
because hey a e uncons i u ional as cons i u ional loyal y. While we a e no he i s o
hink abou his phenomenon, as is illus a ed by ou discussion o he li e a u e abo e, we
a e unawa e o anyone de ining and o malizing he concep in ques ion.6 We a e also no
he i s o use he e m cons i u ional loyal y. Fon ana and Huq (2018), o example, use
he e m o indica e ha a public o icial ac s in acco dance wi h cons i u ional law. Such a
use o he e m, howe e , de alues he meaning o loyal y, which is commonly unde s ood
o in ol e an elemen o genuine suppo . Ou p oposed de ini ion o cons i u ional loyal y,
in con as , is based on he mos common in e p e a ion o he e m loyal y, as i assumes
an in insic mo i a ion o hose who a e willing o incu a cos o uphold he cons i u ional
o de . 7
Now, how does an indi idual de e mine which ac ions a e uncons i u ional? Some imes i
is s aigh o wa d o iden i y he iola ion o a cons i u ional ule, bu in many cases he co -
ec in e p e a ion o he cons i u ion equi es expe ise. The au ho i y o de e mine whe he
an ac ion is uncons i u ional hen lies wi h a cou . In mos coun ies, he ele an cou can
ule on legisla ion o go e nmen p ac ices and decla e hem uncons i u ional. Howe e , he
cou can also o e weake signals o cons i u ionali y. Fo ins ance, he Is aeli Sup eme
Cou some imes issues an “in alidi y no ice” o p oblema ic legal ac s, whe eby he ac
emains in ac , bu he go e nmen is wa ned ha i migh be decla ed uncons i u ional i he
in ingemen o igh s pe sis s (see, e.g., Bendo , 2020). Mo e gene ally, w i en opinions
and dissen ing opinions occasionally hin a judges’ expec a ion ha ce ain p ac ices migh
no hold up agains cons i u ional e iew. Finally, e en o al a gumen s can o e an in o ma-
i e p e iew o how judges will decide a case o a e likely o decide o he cases.
Based on hese insigh s, we o mula e wo connec ed hypo heses:
H1 Recei ing a signal om a cou ha a policy is uncons i u ional educes indi iduals’
suppo o he espec i e policy.
H2 A s onge signal ha a policy is uncons i u ional educes indi iduals’ suppo o he
espec i e policy mo e han a weake signal.
Bo h H1 and H2 a e based on he no ion ha indi iduals who a e loyal o he cons i u ion
will demons a e lowe suppo o uncons i u ional ac ions. Thus, upon lea ning ha a cou
6 Ou de ini ion o cons i u ional loyal y di e s om he concep o “cons i u ional app o al” discussed by
S ephanopoulos and Ve s eeg (2016). The la e is based on people’s s a ed app o al o hei cons i u ion.
S ephanopoulos and Ve s eeg explain ha cons i u ional app o al implies speci ic a he han di use sup-
po (see Eas on, 1975), because app o al is no abs ac . I is based on he conc e e con en and e ec s o
he cons i u ion. App o al can be based on he ep esen a ion o he will o he people in he cons i u ion, o
i is g ounded in eason and jus ice being e lec ed in cons i u ional ules (Ha el & Shina , 2023). Cons i u-
ional loyal y, in con as o cons i u ional app o al, e e s o di use suppo o he pe cei ed legi imacy o
he cons i u ion as such, i.e., independen o he conc e e con en o he documen .
7 While subjec s do no incu a mone a y cos when changing hei opinion, a psychological/cogni i e cos
is in ol ed. Indi iduals who change hei mind in ligh o new in o ma ion, o example, ha e o in es
ene gy (S one e al., 2022). Fu he mo e, Falk e al. (2018) show ha a i udes elici ed h ough su eys a e
equen ly consis en wi h p e e ences exp essed unde condi ions whe e subjec s ace angible incen i es.
In p ac ice, he cos o en o cing he cons i u ion can a y. I anges om o ing o a pa y whose policies
a e compa ible wi h he cons i u ion o isking one’s li e in p o es ing agains a dic a o who iola es he
cons i u ion. Thus, he willingness o incu small cos s o p o ec he cons i u ion can ha e impo an con-
sequences a leas in some si ua ions.
1 3
5
Public Choice (2025) 205:1–18
conside s a policy po en ially uncons i u ional, indi iduals will lowe hei suppo (H1) and
mo e so i he signal o uncons i u ionali y is s onge (H2).
So a , we ha e made an implici assump ion ha cou signals ega ding cons i u ional-
i y con ey c edible in o ma ion. Howe e , acco ding o Caldei a (1986), he cou s’ abili y
o appeal o ci izens’ cons i u ional loyal y should depend on he le el o con idence hey
enjoy in he popula ion. In o he wo ds, indi iduals who do no ha e con idence in he
cou s may no espond as s ongly o a cou signal as hose who do us he cou s. This
condi ional ela ionship yields ou hi d hypo hesis:
H3 Indi iduals wi h highe con idence in he cou s educe hei suppo o he espec i e
policy mo e upon ecei ing a cou signal ha i is uncons i u ional han hose wi h lowe
con idence in he cou s.
Ano he ac o which may in luence an indi idual’s esponsi eness o a cou signal is legal
educa ion. The e is e idence sugges ing ha di e en legally ained g oups a e o e con i-
den in hei abili ies (Eigen & Lis okin, 2012; F anck e al., 2017; Goodman-Delahun y e
al., 2010). Such o e con idence may lead hem o belie e ha —as legal expe s— hey do
no depend on a cou o legal in e p e a ion. Acco ding o his a gumen , a cou ’s signal
would no ha e much e ec on he belie s o hose wi h a legal educa ion.8
H4 Indi iduals wi h a law deg ee educe hei suppo o he espec i e policy less upon
ecei ing a cou signal ha i is uncons i u ional han hose wi hou a law deg ee.
Fo ou inal hypo hesis, we conside he ole o indi iduals’ ai ness conce ns, as cons i-
u ional loyal y is likely g ounded in an indi idual’s b oade e hical con ic ions. Mo al
Founda ions Theo y (MFT; see Haid , 2012; G aham e al., 2013) is he mos well-es ab-
lished a emp o explain in e pe sonal a ia ion in human mo al easoning based on a se
o his o ically e ol ed cul u al ai s. MFT names ai ness as one undamen al dimension
o human mo ali y. As he ai ness ounda ion suppo s coope a ion and ecip oci y, while
encou aging opposi ion o unequal ea men o indi iduals, chea ing, and ule iola ions, i
is likely o p edispose indi iduals o show cons i u ional loyal y. This b ings us o ou i h
and inal hypo hesis:
H5 Indi iduals wi h a s onge ai ness-based mo ali y educe hei suppo o he espec-
i e policy mo e upon ecei ing a cou signal ha i is uncons i u ional han hose wi h
weake ai ness-based mo ali y.
O e all, ou hypo heses p edic ha cons i u ional loyal y leads indi iduals o espond o a
cou signal and ha he size o his e ec depends on he s eng h o he signal, indi iduals’
le el o con idence in he cou s, hei legal educa ion, and hei ai ness-based mo ali y.
A cou signal o uncons i u ionali y causes indi iduals ha a e loyal o he cons i u ion o
educe hei suppo o he policy in ques ion. The educ ion in suppo is la ge i he cou
8 A plausible coun e a gumen , hough lacking empi ical e idence, assumes ha legally ained indi iduals
ha e in e nalized a g ea e app ecia ion o cons i u ional compliance. Consequen ly, hey would educe
hei suppo o he espec i e policy mo e upon ecei ing a cou signal ha i is uncons i u ional han
hose wi hou a law deg ee.
1 3
6
Public Choice (2025) 205:1–18
signal is s onge . Indi iduals who a e con iden in he cou s, who ha e no law deg ee, o
whose mo ali y is ai ness-based show a s onge (nega i e) eac ion o he cou signal.
3 Su ey design and dis ibu ion
3.1 Su ey design
Ou su ey design elies on a simple wi hin-subjec ea men :9 we ask subjec s o a e hei
le el o suppo o COVID-19 mi iga ion policies (on a 7-poin Like scale) unde h ee
condi ions, a baseline and wo ea men s. In he baseline condi ion, we simply ask o wha
deg ee each subjec suppo s he policy. Then, in he p obable iola ion condi ion, we ask
o hei le el o suppo in case he highes ele an cou in hei coun y would decide ha
he policy p obably iola es a cons i u ional igh . Finally, in he clea iola ion condi ion,
we elici he le el o suppo in case he highes cou would decide ha he policy clea ly
iola es a cons i u ional igh . Hence, ou ea men condi ions only di e in e ms o he
inc emen ally inc easing signal (none s. weak s. s ong) esponden s ecei e om he
cou , which indica es ha he policy migh be uncons i u ional. This se up is equi alen o
he s a egy me hod widely used in economic expe imen s (see, e.g., B and s & Cha ness,
2011; Fischbache e al., 2012), as we elici he decisions o esponden s in di e en hypo-
he ical si ua ions.10
To ensu e ha ou esul s a e no policy-speci ic, we elici he le el o suppo o h ee
common mi iga ion policies: ou doo acial mask manda es, p ohibi ions o going on aca-
ion ab oad, and shu downs o non-essen ial businesses. The mi iga ion policies a e p e-
sen ed in a andomized o de , bu he ea men condi ions a e always in he a o emen ioned
o de . We also collec in o ma ion on basic demog aphics, con idence in a ious ac o s
in ol ed in igh ing he pandemic, and answe s o selec ed ques ions om he Mo al Foun-
da ions Ques ionnai e (MFQ; see G aham e al., 2011).
To es ou hypo heses, we ely on h ee a iables o in e es , aside om ou ea men
indica o : con idence in he cou s, a dummy a iable o whe he he subjec has a law
deg ee, and how much he subjec elies on he ai ness ounda ion in hei mo al judgmen s.
The con idence a iable is measu ed on a 4-poin Like scale wi h highe alues e lec ing
mo e con idence in cou s. The ai ness a iable is de i ed om an MFQ ques ion asking
he subjec how impo an i is o hei mo al judgmen whe he “someone was denied his
o he igh s”, measu ed on a 6-poin Like scale whe e highe alues co espond o s on-
ge ai ness conce ns. Table A.1 in he Appendix desc ibes all a iables.
9 A wi hin-subjec design is ideal o es ou hypo heses H3 o H5, as hese equi e an in e ac ion be ween
indi idual a ibu es and he ea men s (see, e.g., Czibo e al., 2019). A be ween-subjec design would be
cos ly in e ms o e iciency and is less sui able o es ima ing ea men e ec s condi ional on esponden
cha ac e is ics.
10 Fischbache e al. (2012) ind ha hose who demons a e condi ional coope a ion unde he s a egy
me hod do so also unde he di ec - esponse model, sugges ing ha bo h me hods yield simila esul s.
B and s and Cha ness (2011) also ind no di e ence be ween he s a egy me hod and he di ec - esponse
model o he majo i y o he expe imen s hey conside .
1 3
7
Public Choice (2025) 205:1–18
3.2 Su ey dis ibu ion
The su ey was p og ammed using he s anda d so wa e Qual ics and dis ibu ed ia Ama-
zon’s MTu k pla o m, which is widely used in online expe imen s (Cli o d e al., 2015;
Ho on e al., 2011; Johnson & Ryan, 2020; Macdonald, 2024; Pee e al., 2017; S i ch
e al., 2024).11 Th ee s a e-o - he-a p ecau ions we e aken o ensu e high da a quali y:
Fi s , he su ey includes h ee a en ion checks (e.g., an ins uc ional manipula ion check).
Responden s who ailed an a en ion check we e excluded.12 Second, we used an algo i hm
o iden i y and exclude use s o VPNs, which ou su ey ins uc ions explici ly p ohibi .
This p e en s indi iduals om mis epo ing hei loca ion and is used o il e ou audu-
len pa icipan s (Kennedy e al., 2020). Thi d, we ec ui ed only subjec s wi h a posi i e
ack eco d (i.e., a leas 100 p e iously comple ed asks and a leas 95% o hese asks
we e app o ed by he ask adminis a o s). Da a collec ion ook place be ween Ma ch 2021
and Ap il 2022 o ensu e a su icien sample size. While he e is e idence ha pa icipan s’
exposu e o di e en le els o in ec ion isk can a ec hei a i ude owa ds go e nmen
in e en ions (Alsan e al., 2023), ou s udy design elimina es he need o conside hese
di e ences, because we a e only in e es ed in how cou signals change a gi en indi idual’s
policy suppo . Thus, he le el o an indi idual’s exposu e is held cons an in ou compa i-
sons by con olling o esponden ixed e ec s and we do no need o con ol o di e ences
in coun ies’ a ec edness by he pandemic o o a ia ion in a ec edness o e ime.
Ou non ep esen a i e global sample includes 1,080 esponden s om 58 democ a ic
coun ies.13 Because we ask each subjec abou hei suppo o h ee policies unde h ee
condi ions ( he baseline and wo ea men s), his yields a o al o
3×3×1080 = 9,720
obse a ions. Table A.2 in he Appendix gi es an o e iew o ou sample’s coun y co e -
age. The numbe o obse a ions is no equal ac oss coun ies, bu his is no p oblema ic in
a wi hin-subjec design (see also oo no e 4).
4 Resul s
4.1 Summa y s a is ics
Table 1 p esen s summa y s a is ics. A e age suppo o mi iga ion policies lies be ween
3.6 and 5.5 on a scale om 1 o 7 and he mean is educed in he p obable and clea iola-
ion condi ions ela i e o he baseline. Al hough ou sample includes esponden s o e e y
adul age, he a e age esponden s a e only in hei mid-30s, which is ep esen a i e o he
pool o MTu k wo ke s.14 Some 30% o he esponden s a e emale. 5% hold a law deg ee.
11 Some schola s ha e oiced conce ns abou he quali y o da a collec ed on MTu k (e.g., Ahle e al.,
2025; Douglas e al., 2023). Howe e , al e na i e mic o- ask ma ke s do no o e a global coun y co e age
compa able o ha o MTu k. I andom measu emen e o was a se ious p oblem in ou da a collec ion, ou
esul s would be biased owa d ze o.
12 No e ha he conce ns oiced by Mon gome y e al. (2018) ega ding pos - ea men a en ion checks
conce n only be ween-subjec designs.
13 Ou sample is nei he ep esen a i e o he global no o any na ional popula ion. I is no ou goal o
measu e he e ec size in hese popula ions. As in many s anda d economic expe imen s, ou goal is only o
demons a e he exis ence o a social phenomenon – cons i u ional loyal y.
14 One esponden epo ed an unlikely age o 99, he second-oldes pe son is 75.
1 3
8
Public Choice (2025) 205:1–18
han expe signals, policies, ime ames, and coun ies), we canno de e mine conclusi ely
wha causes his disc epancy. Howe e , ou s udy appea s o ha e some ad an ages wi h
espec o he ea men esis ance iden i ied and s udied by Cope (2023).
Policymake s can conclude om ou s udy ha when a high cou , such as he US
Sup eme Cou , enjoys s ong public suppo , i can ely on ci izens’ cons i u ional loyal y
o e ec i ely cons ain he go e nmen . S ou enbo ough and Haide -Ma kel (2008), how-
e e , show ha con idence in he US Sup eme Cou a ies o e ime and Vige s and Saad
(2024) epo ha con idence in he US Sup eme Cou has eached a eco d low in 2024.
When con idence in he Sup eme Cou declines, i becomes mo e di icul o he Cou o
en o ce i s decisions is-à- is he o he b anches o go e nmen .
Ou indings a e also consis en wi h poli ical e en s in Ge many du ing he COVID-19
pandemic. Some poli ical pa ies (e.g., he F ee Democ a ic Pa y) we e eluc an o adop
s ic COVID-19 mi iga ion policies. Howe e , he Ge man cons i u ional cou uled on
No embe 19, 2021 ha he mi iga ion policies adop ed in he sp ing o 2021 we e con-
s i u ional. Subsequen ly, poli icians om he new go e ning coali ion—some o whom
e en b ough he p e ious go e nmen o cou o e i s s ic mi iga ion policies—quickly
signaled hei willingness o adop ha she mi iga ion policies o cu b he ou h wa e o he
pandemic.
As a inal example, conside he ini ia i e o he Is aeli go e nmen s a ing in la e 2022
o e ise he coun y’s Basic Laws (see Salzbe ge , 2023 o an o e iew o hese poli ical
e en s). As pa o his ini ia i e, he go e nmen in oduced an amendmen ha would
allow he appoin men o a pa icula pe son, who was p e iously con ic ed o c iminal
o enses, as a minis e . The Is aeli Sup eme Cou ejec ed he appoin men wi hou s iking
down he Basic Law amendmen .19 Ye , some o he judges indica ed in he uling ha he
amendmen migh be uncons i u ional. The la ge demons a ions ha e up ed in esponse o
he go e nmen ini ia i e in gene al las ed o many mon hs. The Sup eme Cou ’s signal
o he amendmen ’s p obable uncons i u ionali y plausibly p o ided addi ional backwind o
he demons a ions.
We ha e ocused he e on how indi idual suppo o mi iga ion policies depends on
in o ma ion abou hei cons i u ionali y, he pe cei ed eliabili y o ha in o ma ion, and
he ai s o he espec i e indi idual. Some ollow-up ques ions seem in e es ing: Does
indi idual suppo o policies and i s esponsi eness o hei cons i u ionali y depend on
an indi idual’s exposu e o he pandemic? A e indi iduals who ques ion he se iousness o
he pandemic o he us wo hiness o public (heal h) au ho i ies as in e es ed as o he s in
he cons i u ionali y o mi iga ion policies? Do cou s use hei signaling powe s a egi-
cally? Does he signal’s e ec depend on who is he sende (cou s, expe s, news media,
poli icians, e c.)? Does i make a di e ence i he cou does no ci e he cons i u ion, bu ,
o example, cus oma y o eligious laws as being iola ed by he mi iga ion policy? Why
a e women gene ally mo e suppo i e o COVID-19 mi iga ion policies? And inally, wha
coun y-le el ac o s can explain he exis ence and s eng h o cons i u ional loyal y? We
lea e hese ques ions o u u e esea ch.
Acknowledgemen s The au ho s hank Michael Penede , Hans Pi lik, ou anonymous e iewe s and he
associa e edi o o Public Choice, and pa icipan s o he ollowing con e ences and semina s: Aus ian Ins i-
19 Is ael Sup eme Cou Case 8948/22, Schien eld e al. s. he Is aeli Pa liamen (“Knesse ”), Jan. 5, 2023,
in Heb ew, h 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 i _ u l l . p d .
1 3
15
Public Choice (2025) 205:1–18
u e o Economic Resea ch’s WIFO esea ch semina , IFN S ockholm con e ence on “The Economics o
Cul u e and Ins i u ions”, F ench Law & Economics Associa ion 2022, Eu opean Associa ion o Law &
Economics 2022, ICON-S 2022, he Eu opean Public Choice Socie y 2023, and he symposium “Cul u es
o T us and Ins i u ions o F eedom” a IFN S ockholm (No . 30 – Dec. 3, 2023) o help ul commen s and
sugges ions.
Funding Open Access unding enabled and o ganized by P ojek DEAL.
Financial suppo by he Ge man Resea ch Founda ion (#381589259) and he Polish Na ional Science Cen-
e (#2016/23/G/HS4/04371) is g a e ully acknowledged. Roee Sa el has ecei ed inancial suppo o his
p ojec in he o m o a symposium hono a ium by he John Temple on Founda ion (#62065). The opinions
exp essed a e hose o he au ho s and do no necessa ily e lec he iews o he John Temple on Founda ion.
Decla a ions
E hics app o al, consen , p i acy, and compensa ion In compliance wi h hei e hical obliga ions, he au ho s
decla e ha his esea ch has been subjec o e iew ega ding i s compliance wi h e hical s anda ds a he
Uni e si y o Hambu g and ha su ey esponden s ha e gi en hei consen o he collec ion o hei da a
and i s use o esea ch pu poses. All da a is collec ed anonymously, and no appa en ha m is o be expec ed
o pa icipan s o his su ey. Su ey pa icipan s we e paid be ween 2USD and 3USD o a su ey ha akes
abou i e o en minu es o ill in. Responden s om low and lowe -middle income economies, acco ding o
he Wo ld Bank, ecei ed 2USD. Those in uppe -middle income economies ecei ed 2.50USD. Responden s
om high income economies ecei ed 3USD. This amoun s o an hou ly wage o 12 (/24) o 18 (/36) USD,
depending on he esponden ’s coun y o esidence and hei speed.
P e- egis a ion The su ey has been p e egis e ed a he Cen e o Open Science (OSF): h p s : / / o s . i o / 7 j 8 a
m / i l e s / o s s o a g e / 6 0 5 b b 7 2 e 1 2 b 6 0 0 0 2 3 a a 7 9 7 e .
Replica ion ma e ials Replica ion iles o ou empi ical analysis a e a ailable a : h p s : / / d o i . o g / 1 0 . 7 9 1 0 / D
V N / S S Q W D D .
Compe ing in e es s The au ho s ha e no po en ial o pe cei ed con lic s o in e es a ising om his
esea ch.
Open Access This a icle is licensed unde a C ea i e Commons A ibu ion 4.0 In e na ional License,
which pe mi s use, sha ing, adap a ion, dis ibu ion and ep oduc ion in any medium o o ma , as long as
you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Commons
licence, and indica e i changes we e made. The images o o he hi d pa y ma e ial in his a icle a e
included in he a icle’s C ea i e Commons licence, unless indica ed o he wise in a c edi line o he ma e ial.
I ma e ial is no included in he a icle’s C ea i e Commons licence and you in ended use is no pe mi ed
by s a u o y egula ion o exceeds he pe mi ed use, you will need o ob ain pe mission di ec ly om he
copy igh holde . To iew a copy o his licence, isi h p://c ea i ecommons.o g/licenses/by/4.0/.
Re e ences
Ahle , D. J., Roush, C. E., & Sood, G. (2025). The mic o- ask ma ke o lemons: Da a quali y on Amazon’s
mechanical Tu k. Poli ical Science Resea ch and Me hods, 13(1), 1–20.
Allendoe e , M. G. (2017). Who ca es abou human igh s? Public opinion abou human igh s o eign
policy. Jou nal o Human Righ s, 16(4), 428–451.
Alsan, M., B aghie i, L., Eichmeye , S., Kim, M. J., S an che a, S., & Yang, D. Y. (2023). Ci il libe ies in
imes o c isis. Ame ican Economic Jou nal: Applied Economics, 15(4), 389–421.
A ı, B., & Sonmez, B. (2025). Human igh s iola ions and public suppo o sanc ions. Jou nal o Peace
Resea ch, 62(1), 68–84.
Bae schmann, G., S aub, K. E., & Winkelmann, R. (2015). Consis en Es ima ion o he ixed e ec s o de ed
logi model. Jou nal o he Royal S a is ical Socie y: Se ies A, 178(3), 685–703.
Bendo , A. L. (2020). The Is aeli judicia y-cen e ed cons i u ionalism. In e na ional Jou nal o Cons i u-
ional Law, 18(3), 730–745.
1 3
16
Public Choice (2025) 205:1–18
Bjø nsko , C., & Gu mann, J. (2024). Coups and cons i u ional compliance. Pape p esen ed a he Sou he n
Economic Associa ion Annual Mee ing in Washing on D.C.
Bjø nsko , C., & Rode, M. (2020). Regime ypes and egime change: A new da ase on democ acy, coups,
and poli ical ins i u ions. The Re iew o In e na ional O ganiza ions, 15(2), 531–551.
B aman, E. (2023). Assessing he c edibili y o cons i u ional expe s. Jou nal o Law and Cou s, 11(1),
86–103.
B and s, J., & Cha ness, G. (2011). The s a egy e sus he di ec - esponse me hod: A i s su ey o expe i-
men al compa isons. Expe imen al Economics, 14(3), 375–398.
Caldei a, G. A. (1986). Nei he he pu se no he swo d: Dynamics o public con idence in he sup eme cou .
Ame ican Poli ical Science Re iew, 80(4), 1209–1226.
Cha ness, G., Gneezy, U., & Kuhn, M. A. (2012). Expe imen al me hods: Be ween-subjec and wi hin-sub-
jec design. Jou nal o Economic Beha io & O ganiza ion, 81(1), 1–8.
Chil on, A. S., & Ve s eeg, M. (2016). In e na ional law, cons i u ional law, and public suppo o o u e.
Resea ch & Poli ics, 3(1), 1–9.
Chil on, A. S., & Ve s eeg, M. (2020). How cons i u ional igh s ma e . Ox o d Uni e si y P ess.
Chil on, A. S., Cope, K. L., C ab ee, C., & Ve s eeg, M. (2025). Suppo o cons i u ional igh s du ing
c isis: E idence om he pandemic. Ame ican Jou nal o Compa a i e Law, o hcoming.
Chou agun a, A., Gu mann, J., & Voig , S. (2024). Shocking esilience? E ec s o ex eme e en s on cons i-
u ional compliance. Jou nal o Ins i u ional Economics, 20, e3.
Cli o d, S., Jewell, R. M., & Waggone , P. D. (2015). A e samples d awn om mechanical Tu k alid o
esea ch on poli ical ideology? Resea ch & Poli ics, 2(4), 205316801562207.
Cope, K. L. (2023). Measu ing law’s no ma i e o ce. Jou nal o Empi ical Legal S udies, 20(4), 1005–1044.
Cope, K. L., & C ab ee, C. (2020). A Na ionalis backlash o in e na ional e ugee law: E idence om a
su ey expe imen in Tu key. Jou nal o Empi ical Legal S udies, 17(4), 752–788.
Cope, K. L., & C ab ee, C. (2022). Mig an - amily sepa a ion and highe -o de laws’ di e ging no ma i e
o ce. The Jou nal o Legal S udies, 51(2), 403–426.
Czibo , E., Jimenez-Gomez, D., & Lis , J. A. (2019). The dozen hings expe imen al economis s should do
(mo e o ). Sou he n Economic Jou nal, 86(2), 371–432.
Douglas, B. D., Ewell, P. J., & B aue , M. (2023). Da a quali y in online human-subjec s esea ch: Com-
pa isons be ween MTu k, p oli ic, CloudResea ch, qual ics, and SONA. PLOS ONE, 18(3), e0279720.
Eas on, D. (1975). A e-assessmen o he concep o poli ical suppo . B i ish Jou nal o Poli ical Science,
5(4), 435–457.
Eigen, Z. J., & Lis okin, Y. (2012). Do lawye s eally belie e hei own hype, and should hey? A na u al
expe imen . The Jou nal o Legal S udies, 41(2), 239–267.
Falk, A., Becke , A., Dohmen, T., Enke, B., Hu man, D., & Sunde, U. (2018). Global e idence on economic
p e e ences. Qua e ly Jou nal o Economics, 133(4), 1645–1692.
Fischbache , U., Gäch e , S., & Que cia, S. (2012). The beha io al alidi y o he s a egy me hod in public
good expe imen s. Jou nal o Economic Psychology, 33(4), 897–913.
Fon ana, D., & Huq, A. (2018). Ins i u ional loyal ies in cons i u ional law. Uni e si y o Chicago Law
Re iew, 85(1), 1–84.
F anck, S. D., an Aaken, A., F eda, J., Gu h ie, C., & Rachlinski, J. J. (2017). Inside he a bi a o ’s Mind.
Emo y Law Jou nal, 66(5), 1115–1174.
Goodman-Delahun y, J., G anhag, P. A., Ha wig, M., & Lo us, E. F. (2010). Insigh ul o wish ul: Lawye s’
abili y o p edic case ou comes. Psychology Public Policy and Law, 16, 133–157.
G aham, J., Haid , J., Kole a, S., Mo yl, M., Iye , R., Wojcik, S. P., & Di o, P. H. (2013). Mo al ounda ions
heo y: The p agma ic alidi y o mo al plu alism. Ad ances in Expe imen al Social Psychology, 47,
55–130.
G aham, J., Nosek, B. A., Haid , J., Iye , R., Kole a, S., & Di o, P. H. (2011). Mapping he mo al domain.
Jou nal o Pe sonali y and Social Psychology, 101(2), 366–385.
Gu mann, J., Me elska-Szaniawska, K., & Voig , S. (2024a). Leade cha ac e is ics and cons i u ional com-
pliance. Eu opean Jou nal o Poli ical Economy, 84, 102423.
Gu mann, J., Me elska-Szaniawska, K., & Voig , S. (2024b). The compa a i e cons i u ional compliance
da abase. The Re iew o In e na ional O ganiza ions, 19(1), 95–105.
Haid , J. (2012). The igh eous Mind: Why good people a e di ided by poli ics and eligion. Pan heon.
Ha el, A., & Shina , A. (2023). Two concep s o cons i u ional legi imacy. Global Cons i u ionalism, 12(1),
80–105.
Ho on, J. J., Rand, D. G., & Zeckhause , R. J. (2011). The online labo a o y: Conduc ing expe imen s in a
eal labo ma ke . Expe imen al Economics, 14(3), 399–425.
Johnson, D., & Ryan, J. B. (2020). Amazon mechanical Tu k wo ke s can p o ide consis en and economi-
cally meaning ul da a. Sou he n Economic Jou nal, 87(1), 369–385.
1 3
17
Public Choice (2025) 205:1–18
Kan o owicz, J. (2023). Tes ing public eac ion o cons i u ional iscal ules iola ions. Cons i u ional Poli i-
cal Economy, 34(4), 483–509.
Kennedy, R., Cli o d, S., Bu leigh, T., Waggone , P. D., Jewell, R., & Win e , N. J. G. (2020). The shape
o and solu ions o he MTu k quali y c isis. Poli ical Science Resea ch and Me hods, 8(4), 614–629.
Macdonald, D. (2024). Poli ical us and Ame ican public suppo o ee ade. Poli ical Beha io , 46(2),
1037–1055.
Me elska-Szaniawska, K. (2021). Pos -socialis cons i u ions: The de ju e–de ac o Gap, i s e ec s and de e -
minan s. Economics o T ansi ion and Ins i u ional Change, 29(2), 175–196.
Mon gome y, J. M., Nyhan, B., & To es, M. (2018). How condi ioning on pos ea men a iables can uin
you expe imen and wha o do abou i . Ame ican Jou nal o Poli ical Science, 62(3), 760–775.
Pee , E., B andima e, L., Sama , S., & Acquis i, A. (2017). Beyond he Tu k: Al e na i e pla o ms o
c owdsou cing beha io al esea ch. Jou nal o Expe imen al Social Psychology, 70, 153–163.
Salzbe ge , E. (2023). A Possible Regime Change in Is ael. Ve assungsblog: On Ma e s Cons i u ional. h
p s : / / e a s s u n g s b l o g . d e / e g i m e - c h a n g e - i s a e l / (Decembe 15, 2024).
S ephanopoulos, N. O., & Ve s eeg, M. (2016). The con ou s o cons i u ional app o al. Washing on Uni e -
si y Law Re iew, 94(1), 113–190.
S one, C., Ma ingley, J. B., & Rangelo , D. (2022). On second hough s: Changes o Mind in decision-
making. T ends in Cogni i e Sciences, 26(5), 419–431.
S ou enbo ough, J. W., & Haide -Ma kel, D. P. (2008). Public con idence in he U.S. Sup eme cou : A new
look a he impac o cou decisions. The Social Science Jou nal, 45(1), 28–47.
S i ch, J. M., Pede sen, M. J., & Pezo, I. (2024). C owdsou ced da a in public adminis a ion esea ch: A
e iew and look o he u u e. Public Adminis a ion Re iew, o hcoming.
Tomz, M. R., & Weeks, J. L. P. (2020). Human igh s and public suppo o wa . Jou nal o Poli ics, 82(1),
182–194.
Vige s, B., & Saad, L. (2024). Ame icans pass judgmen on hei cou s: Sha p decline in con idence in judi-
cia y is among he la ges Gallup has e e measu ed. h 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 i c a n s - p
a s s - j u d g m e n - c o u s . a s p x (Decembe 17, 2024).
Voig , S. (2021). Mind he Gap: Analyzing he di e gence be ween cons i u ional ex and cons i u ional eal-
i y. In e na ional Jou nal o Cons i u ional Law, 19(5), 1778–1809.
Weingas , B. R. (1997). The poli ical ounda ions o democ acy and he ule o law. Ame ican Poli ical Sci-
ence Re iew, 91(2), 245–263.
Publishe ’s no e Sp inge Na u e emains neu al wi h ega d o ju isdic ional claims in published maps and
ins i u ional a ilia ions.
1 3
18
