Strategic behaviour and decision making in competitive hospital markets: an experimental investigation

Han, Johann; Kai ies-Schwa z, Nadja; Vomho , Ma kus
A icle — Published Ve sion
S a egic beha iou and decision making in compe i i e
hospi al ma ke s: an expe imen al in es iga ion
In e na ional Jou nal o Heal h Economics and Managemen
P o ided in Coope a ion wi h:
Sp inge Na u e
Sugges ed Ci a ion: Han, Johann; Kai ies-Schwa z, Nadja; Vomho , Ma kus (2024) : S a egic
beha iou and decision making in compe i i e hospi al ma ke s: an expe imen al in es iga ion,
In e na ional Jou nal o Heal h Economics and Managemen , ISSN 2199-9031, Sp inge US, New
Yo k, NY, Vol. 24, Iss. 3, pp. 333-355,
h ps://doi.o g/10.1007/s10754-024-09366-3
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/315274
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 p://c ea i ecommons.o g/licenses/by/4.0/
Vol.:(0123456789)
In e na ional Jou nal o Heal h Economics and Managemen (2024) 24:333–355
h ps://doi.o g/10.1007/s10754-024-09366-3
1 3
RESEARCH ARTICLE
S a egic beha iou anddecision making incompe i i e
hospi al ma ke s: anexpe imen al in es iga ion
JohannHan1,2· NadjaKai ies‑Schwa z3,4· Ma kusVomho 3,4
Recei ed: 24 Ma ch 2022 / Accep ed: 10 Feb ua y 2024 / Published online: 15 Ma ch 2024
© The Au ho (s) 2024
Abs ac
We in es iga e quali y p o ision and he occu ence o s a egic beha iou in compe i i e
hospi al ma ke s whe e p o ide s a e assumed o be semi-al uis ic owa ds pa ien s. Fo
his, we employ a labo a o y expe imen wi h a hospi al ma ke aming. Subjec s decide
on he quali y le els o one o h ee compe ing hospi als espec i ely. We a y he o gani-
za ional aspec o whe he quali y decisions wi hin hospi als a e made by indi iduals o
eams. Realized mone a y pa ien bene i s go o eal pa ien s ou side he lab. In bo h se -
ings, we ind ha deg ees o coope a ion quickly con e ge owa ds nega i e alues, imply-
ing absence o collusion and pa ien cen ed o compe i i e quali y choices. Mo eo e , hos-
pi als ea quali y as a s a egic complemen and adjus hei quali y choice in he same
di ec ion as hei compe i o s. The esponse magni ude o eam ma ke s is weake ; his is
d i en by non-coope a i e o al uis ic eams, which end o se le els o quali y ha a e
s a egically independen .
Keywo ds Quali y compe i ion· Hospi al ma ke s· Team decisions· Al uism·
Labo a o y expe imen
JEL Classi ica ion C92· D03· D43· D64· I11· L13
* Nadja Kai ies-Schwa z
nadja.kai ies-schwa z@uni-duesseldo .de
Johann Han
[email p o ec ed]
Ma kus Vomho
[email p o ec ed]
1 Facul y o Economics andBusiness Adminis a ion, Uni e si y o Duisbu g-Essen, Essen,
Ge many
2 CINCH (Compe en inCompe i ion andHeal h), Be line Pla z 6-8, 45127Essen, Ge many
3 Ins i u e o Heal h Se ices Resea ch andHeal h Economics, Cen e o Heal h andSocie y,
Medical Facul y andUni e si y Hospi al Düsseldo , Hein ich-Heine-Uni e si y Düsseldo ,
Düsseldo , Ge many
4 Ins i u e o Heal h Se ices Resea ch andHeal h Economics, Ge man Diabe es Cen e , Leibniz
Cen e o Diabe es Resea ch a Hein ich-Heine-Uni e si y Düsseldo , Düsseldo , Ge many
334
J.Han e al.
1 3
In oduc ion
In hospi al ma ke s whe e p ices a e egula ed and pa ien s can choose hospi als eely,
quali y has become a c ucial s a egic a iable. The e is a ich se o li e a u e s udying he
e ec s o compe i ion on quali y (see, o example, Gayno & Town, 2012; Gayno e al.,
2015 o su eys). Howe e , li le is known abou he s a egic conside a ions behind hos-
pi als’ quali y p o ision beha iou and whe he he way i is ma e ialized wi hin hospi als
has an e ec on he quali y p o ided.1 This is ele an in heal hca e ma ke s, since p o id-
e s who decide on he quali y o ca e on ei he an indi idual o eam le el a e ypically
assumed o be semi-al uis ic (Chalkley & Malcomson, 1998; Chone & Ma, 2011; Ellis &
McGui e, 1986, 1990; Jack, 2005; Liu & Ma, 2013). Se e al labo a o y expe imen s ha e
iden i ied such al uism in medical p o ision beha iou o indi idual decisions (B osig-
Koch e al., 2016, 2017a; Godage & Wiesen, 2013). Hence, when he e ogeneously semi-
al uis ic p o ide s a e compe ing, he ou come migh be di e en om classic ma ke
expe imen s wi h p o i maximizing agen s.2
Using a heo e ical model, B ekke e al. (2017) ind ha he s a egic na u e o qual-
i y compe i ion c ucially depends on he hospi als’ deg ee o al uism and he in e ac ion
be ween quali y and cos -con ainmen incen i es. Wi hou oom o cos con ainmen , hey
show ha quali y is a s a egic complemen i al uism is ela i ely low compa ed o he
cos -subs i u abili y be ween quali y and ou pu ; i al uism is ela i ely high, quali y can
also be a subs i u e. In con as o he heo e ical p edic ions, he empi ical li e a u e on
hospi als’ s a egic in e ac ions conside ing he s a egic na u e o quali y in hospi al ma -
ke s is a he small and no s aigh o wa d— ha is, i a all hey ind posi i e signi ican
associa ions be ween i als’ quali y choices. G a elle e al. (2014) in es iga e whe he a
hospi al’s quali y depends on he quali y o i al hospi als using da a om he English
Na ional Heal h Se ice and ind mixed esul s: only some quali y indica o s a e posi i ely
associa ed and a e hus complemen s. Ye , hey do no ind e idence o any nega i e asso-
cia ion. In a ollow-up s udy, Longo e al. (2017) in es iga e a longe ime ho izon and only
ind signi ican posi i e associa ions o one ou o eigh quali y indica o s. Guccio and Lisi
(2016) use da a om I alian hospi als and in es iga e he s a egic na u e o caesa ean sec-
ion a es, showing ha a hospi al’s p o ision beha iou is s ongly a ec ed by he beha -
iou o he o he hospi als wi hin one egion.
Howe e , da a om he ield migh be p one o ce ain challenges, making a con-
olled analysis di icul . Fo ins ance, a necessa y assump ion o p o ide coo dina ion
and demand e ec s is ha quali y is obse able, which is no necessa ily he case. While
e o ms including policies a ge ing inc eased anspa ency by publicizing physician qual-
i y in o ma ion ha e gained popula i y (D ano e & Jin, 2010), o he aspec s, such as he
1 No e ha Bloom (2015) inds ha compe i ion has a posi i e impac on ou come as well as managemen
quali y in he English public hospi al sec o , ye managemen quali y is measu ed by a su ey ool co e ing
hings apa om he decision-making p ocesses, including incen i es, moni o ing, o a ge se ing.
2 Economis s ha e used labo a o y expe imen s o s udy compe i ion in ma ke s o qui e some ime (see
Hol , 1995 and Po e s & Sue ens, 2013 o a mo e ecen example). The majo i y o hese expe imen s is
ei he se in a quan i y o p ice compe i ion en i onmen . An impo an dis inc ion be ween games o quan-
i y and games o p ice se ing is ha he o me a e conside ed as s a egic subs i u es and he la e as s a-
egic complemen s in compe i ion. Gene ally, o s a egic subs i u es, he op imal esponse o an agg essi e
s a egy by he compe i o is o become less agg essi e – i.e., mo e in o he opposi e di ec ion – while o
s a egic complemen s he op imal esponse is o become mo e agg essi e and ollow he di ec ion o he
mo e (Ti ole, 1988).
335
S a egic beha iou anddecision making incompe i i e hospi al…
1 3
mul idimensionali y o quali y—o en in combina ion wi h subjec i e pe cep ions o o e -
all quali y, as well as neglec ing quali y epo s en i ely—lead o a lack o anspa ency in
he ield. This makes coo dina ion much mo e di icul .
E en i quali y esponses we e pe ec ly obse able om ield da a, i would be di icul
o in e whe he p o ide s di ec ly eac ed o hei compe i o s’ quali y choices o whe he
hei quali y choice was an ou come o he decision p ocess wi hin he hospi al and hus
de i ed om adhe ing o a ce ain social no m o ea men s yle. Hence, in o de o unde -
s and hospi als’ quali y esponses, one should also accoun o he o ganiza ional s uc u e
o he decision p ocesses ha leads o he quali y p o ided. Typically, he decision on s a-
egic o ien a ion is no jus made by an indi idual bu in o ganized eams like managemen
boa ds, including, o example, he head o he hospi al, he chie inancial o ice , and he
senio physician. Each o hem migh weigh hospi al pe o mance and hus quali y di e -
en ly. Then, he uni a y playe assump ion ha is common in he indus ial o ganiza ion li -
e a u e migh be less easonable i he in ol ed s akeholde s pu di e en weigh s on p o i
and pa ien cen edness.3
The objec i e o his pape is o in es iga e how compe ing and semi-al uis ic hospi als
p o ide quali y when decisions a e made on ei he he indi idual o eam le el. Fo his, we
use da a om he i s pa o a con olled labo a o y expe imen on indi idual and eam
hospi al ma ke s conduc ed by Han e al. (2017). We conduc ed addi ional expe imen al
sessions o inc ease he numbe o obse a ions o eam ma ke s. While Han e al. (2017)
compa e pa s 1 and 2 o he expe imen and ocus on he analysis o a me ge ’s o al
e ec s on quali y p o ision, we ocus on hospi als’ s a egic in e ac ions when a ying he
way decisions a e made (i.e., whe he a an indi idual o eam le el). In he expe imen ,
pa icipan s ake on he ole o a decision make in a hospi al and ha e o choose a quali y
le el o hei espec i e hospi als in 15 consecu i e pe iods. The ma ke consis s o h ee
hospi als ha compe e o pa ien s by hei quali y choice in a epea ed Salop oligopoly
game. We also implemen an expe imen al condi ion wi h a eam decision p ocess. He e,
hospi al boa ds o h ee membe s ha e o ag ee on a quali y le el h ough a simple majo -
i y ule. Decisions a ec eal pa ien s ou side he labo a o y, since he gene a ed pa ien
u ili y is mone ized and ans e ed o a cha i y ha p o ides medical ca e o pa ien s wi h-
ou access o o mal heal hca e.
Ou pape con ibu es o he g owing li e a u e on heal h economic expe imen s on p o-
ide beha iou (Hennig-Schmid e al., 2011; Hennig-Schmid & Wiesen, 2014; G een,
2014; B osig-Koch e al., 2016, 2017a, 2017b, 2020; Laga de & Blaauw, 2017; Kes e -
nich e al., 2015; Di Guida e al., 2019; Ma insson & Pe sson, 2019; Rei e al., 2020;
Kai ies-Schwa z & Soucek, 2020). While he e is some esea ch in es iga ing compe i i e
ma ke s wi h semi-al uis ic heal hca e p o ide s, showing ha semi-al uism yields be e
pa ien ou comes han expec ed o p o i maximizing agen s (B osig-Koch e al., 2017b;
Han e al., 2017), he e is no e idence on s a egic beha iou in hospi al ma ke s wi h qual-
i y compe i ion ha includes a compa ison o indi idual and eam decisions.4 Ou pape
3 Con ibu ions by Pauly and Redisch (1973) and Ha is (1977) add ess he issue o s akeholde s wi h di -
e en objec i es in hospi als, while Phelps (2002) in es iga es di e en p e e ences in he uling body o
hospi als when i comes o quali y decisions. Fo a de ailed discussion o eam decisions in hospi als, see
Ba os and Oli ella (2011).
4 B osig-Koch e al. (2017a) conside collusi e beha iou , bu only o wo physicians. Han e al. (2017)
in es iga e he e ec s on quali y o a me ge due in hospi al ma ke s wi h quali y compe i ion and do no
in es iga e coo dina ion beha iou .
336
J.Han e al.
1 3
also con ibu es o wo s ands o li e a u e on classic ma ke expe imen s. The i s s and
in es iga es aci collusion in epea ed Be and and Cou no ma ke s (Po e s & Sue ens,
), showing ha i is mo e di icul o es ablish collusion when s a egic a iables a e sub-
s i u es. The second one analyses eam decisions in ma ke games, sugges ing ha eams
a e gene ally close o he heo e ical p edic ions o he game han indi iduals a e (see, o
example, Bo ns ein e al., 2008; Kugle e al., 2012 o su eys).5 Howe e , ou se ing is
unique because, gi en he ex e nali ies o pa ien s, semi-al uism is a plausible assump-
ion in hese ma ke s, and he heo e ical benchma ks o he game wi h p o i maximizing
agen s migh be less ele an .
Ou main esul s can be summa ized as ollows. We show ha indi idual and eam
ma ke s s a a sligh ly posi i e deg ees o coope a ion. Howe e , hey quickly con e ge
owa ds nega i e alues, indica ing pa ien -cen ed beha iou . We also show ha ini ial
quali y is signi ican ly mo e homogeneous in eam ma ke s han in indi idual ones. This
migh be explained by he p eceding eam decision p ocess, wi h he majo i y o ing, il e -
ing ou ex eme quali y choices om he beginning. Towa ds he inal pe iods, bo h hos-
pi al ma ke s coo dina e a le els be ween he compe i i e and op imal pa ien ou come.
We also conside s a egic quali y p o ision beha iou and ind ha hospi als led by an
indi idual ea quali y as a s a egic complemen and adjus hei quali y choice in he
same di ec ion as hei compe i o s. This also holds o hospi als led by eams. Howe e ,
hei esponse magni ude is weake . This is d i en by non-coope a i e, o al uis ic, eam
ma ke s, which do no seem o be esponsi e o changes in hei i als’ quali ies bu end o
se quali y independen ly.
The es o he pape p oceeds as ollows. Sec ion "Expe imen al design and heo e ical
amewo k" ou lines he expe imen al design and heo e ical amewo k. Sec ion“Resul s”
p esen s he esul s, and Sec ion “Conclusion” concludes he s udy.
Expe imen al design and heo e ical amewo k
Expe imen al design anddecision si ua ion
Indi idual decisions
Ou se ing is simila o a classical ma ke expe imen bu wi h a heal hca e aming.
Ins ead o i ms, he e a e hospi als compe ing o pa ien s. Pa icipan s a e andomly
ma ched in o a ma ke and a e in cha ge o quali y decisions o one o h ee compe ing
hospi als. A he beginning o each pe iod, subjec s simul aneously choose he quali y le el
q
ha hey wan o p o ide, wi h
q∈{1, 2, 3, …, 13}
.6 The quali y le el a ec s pa ien
u ili y— he highe he be e —which de e mines he pa ien s’ hospi al choice and he e-
o e ma ke demand and hospi al p o i s. O e all, pa icipan s play 15 pe iods whe e hey
decide on he quali y. In line wi h he heo e ical model o quali y compe i ion, hospi als
decide on he quali y o ca e. We assume ha a minimum s anda d o quali y is always
5 The e idence has a endency, bu i is ambiguous o some deg ee. Fo example, Bo ns ein and Gneezy
(2002) ind ha eams con e ge quicke on he compe i i e solu ion, while Raab and Schippe (2009) ind
no di e ences be ween indi idual and eam decisions o h ee i m Cou no oligopolies.
6 To educe complexi y o pa icipan s, hey we e p o ided wi h a disc e e se o quali y le els o choose
om, e en i he heo e ical amewo k was based on con inuous unc ions.

337
S a egic beha iou anddecision making incompe i i e hospi al…
1 3
ul illed o ensu e ha a pa ien ’s p oblem is sol ed. Any addi ional quali y will con inu-
ously inc ease pa ien bene i s. Such addi ional quali y p o ision migh also be e lec ed in
(publicly a ailable) quali y indica o s ha make quali y obse able o he pa ien . Assume,
o ins ance, he a ea o obs e ics: in he e en o a p ema u e bi h, he p esence o a pae-
dia ician leads o an inc eased quali y le el. In Ge many, o example, his aspec is also
e lec ed in quali y indica o s.
Subjec s a e p o ided wi h ull in o ma ion abou he possible consequences o hei
own and hei compe i o hospi als’ quali y decisions wi h espec o p o i s and pa ien
bene i . A p o i and pa ien bene i able is handed ou wi h he ins uc ions (see Appendix
A.1.1 and A.2.1 o he ins uc ions ansla ed in o English, as well as he o iginal ones in
Ge man), and a calcula o is implemen ed in he compu e p og am whe e subjec s can
simula e di e en quali y combina ions in he ma ke wi h he espec i e ou comes (see
Appendix A.2.2 o ansla ed sc eensho s o bo h indi idual and eam ma ke condi ions).
Pa ien u ili y is measu ed as he o al u ili y de i ed by he uni mass o pa ien s in he
ma ke om he quali y p o ided by he h ee hospi als (mo e de ail in Sec ion "Beha -
iou al benchma ks and hypo heses"). Subjec s can use he implemen ed calcula o o cal-
cula e hei own con ibu ion and he con ibu ion by he o he hospi als. The e a e no eal
pa ien s in he lab. To c ea e a mo e ealis ic decision si ua ion ha allows o al uism
owa ds pa ien s, we implemen ed a ans e o he mone a y equi alen o quali y choices,
simila o Eckel and G ossman (1996), Hennig-Schmid e  al. (2011), and B osig-Koch
e al., (2016, 2017a, 2017b). Pa icipan s in he expe imen knew ha he highe he le el
o quali y p o ided, he mo e money would go o a cha i y g an ing access o heal hca e o
uninsu ed pa ien s in Ge many who would no be ea ed o he wise.
Team decisions
Ou main ce e is pa ibus a ia ion is he way decisions a e made.7 In addi ion o indi-
idual decisions, we ha e a eam condi ion whe e subjec s a e ma ched in eams o h ee
( h oughou he expe imen , he same eam membe s cons i u e a eam) and o m a hospi al
boa d ha makes decisions abou hospi al quali y. The ma ch emains un il he end o he
game. Wi h h ee subjec s on each hospi al boa d, nine playe s o m a ma ke . Hospi al
quali y le el
q
is de e mined by a majo i y o e o he hospi al boa d membe s. The deci-
sion mechanism is a simple majo i y ule, simila o Gille e al. (2011). I a leas wo
membe s p opose he same quali y le el, i is implemen ed as he hospi al quali y le el o
he cu en pe iod. I no majo i y is eached wi hin a o ing ound, he same o ing p ocess
is con inued un il a majo i y decision is eached. Ou expe imen al condi ions hus a y
ega ding whe he decisions a e made a he indi idual o eam le el (see Table1).
Beha iou al benchma ks andhypo heses
In he ollowing sec ion, we de i e beha iou al benchma ks and hypo heses o ou expe i-
men . Fo he indi idual hospi al ma ke s, we e e o he heo e ical amewo k in oduced
in Han e al. (2017), who in es iga e compe i i e ma ke ou comes wi h and wi hou hos-
pi al me ge s. Fo he eam hospi al ma ke s, we p esen he ela ed li e a u e and in e ou
hypo heses om i .
7 Fo a compa ison o a monopoly and a compe i i e ma ke , see Han e al. (2017).
338
J.Han e al.
1 3
Indi idual decisions
We begin by in oducing he model amewo k o indi idual quali y decisions. Fo his,
we conside a Salop model wi h an exogenously ixed numbe o h ee hospi als ha com-
pe e in e ms o ea men quali y (Salop, 1979). A uni mass o pa ien s is uni o mly dis-
ibu ed on a ci cle. Pa ien s ecei e medical ea men a hospi als ha a e equidis an ly
loca ed along he ci cle. A pa ien ’s u ili y depends on he quali y le el
qi
ecei ed in hos-
pi al i wi h
i∈{1, 2, 3}
, as well as on he a el dis ance be ween he hospi al’s loca ion
xi
and he pa ien ’s loca ion
z
. The disu ili y om a eling is measu ed by
>0
. Pa ien s a e
ully insu ed, i.e., p ices o ea men do no a ec hei u ili y. Fu he mo e, i is assumed
ha he ‘basic’ alua ion o ea men is su icien ly la ge o ensu e ha ecei ing ea -
men is always p e e ed o emaining un ea ed.8 Gi en he hospi al’s loca ion
xi
and he
pa ien ’s loca ion
z
, he pa ien ’s u ili y
uz
,
xi
is gi en by:
The o al u ili y o pa ien s being ea ed by he hospi al i is Bi=

zi+1
i
∫
o
(
+qi− s
)
ds +

z
i
−1
∫
o(
+qi− s
)
ds
,
which de e mines he con ibu ion o he pa ien in he
ollowing. I can be shown ha hospi al i’s demand
Di
depends on he quali y choices o all
h ee hospi als ac i e in he ma ke and is gi en by:
Hospi als compe e o pa ien s in e ms o quali y.9 Since p ices p o ea men a e exog-
enously gi en by a egula o and ma ginal cos s
c
> 0 pe quali y a e cons an , hospi al i’s
p o i unc ion can be w i en as:
The h ee compe ing hospi als simul aneously choose hei quali y le el in o de o maxi-
mize hei p o i unc ion as s a ed in Eq.(3).10 I we ake he i s -o de condi ion (FOC),
we can sol e o he bes -quali y- esponse unc ion o each hospi al:
The co esponding symme ic Nash equilib ium quali y
q∗
i
and p o i le el
π∗
i
o hospi al i
is gi en by:
(1)
uz,xi
= +q
i
−
|
|
z−x
i|
|.
(2)
D
i=
1
3
+
2qi−
∑
j≠iqj
2
.
(3)
πi
=
(
p−cq
i)
D
i.
(4)
q
i�qj�=
1
2
p
c
+
∑
j≠iqj
4
−
6
.
(5)
q
∗
i=
p
c
−
3
andπ∗
i=
c
9.
8 The assump ion o linea demand is a s ong one. Changing he demand unc ion possibly changes he
p edic ion o he model. This is a clea limi a ion o he model and he co esponding expe imen .
9 In ou model app oach, hospi als canno endogenously choose loca ions since hei loca ions a e ixed.
10 We assume in he ollowing ha disu ili y om a eling is su icien ly small, i.e.,
<
3p∕c
.
339
S a egic beha iou anddecision making incompe i i e hospi al…
1 3
To de i e heo e ical benchma ks and ou beha iou al expec a ions o he s a egic
na u e o quali y p o ision wi hin ou expe imen al design, we now in oduce ou expe i-
men al pa ame iza ion o he o mal model. We chose ou pa ame e s in a way ha sa is-
ies he pa icipa ion cons ain o pa ien s— ha is, demanding ea men e en unde he
hospi al’s minimal quali y p o ision o 1 (
=5
) and ha pa ien s do no a el beyond
one o hei neighbou ing hospi als o ecei e a ea men (
=36
). Regula ed p ices
a e se a
p=44
and ea men cos s a
c=2
. Subjec s could choose a quali y le el
qi∈{1, 2, 3, …, 13}
.11
Based on his pa ame iza ion, we in oduce some ele an benchma ks (see Table2).
The symme ic Nash equilib ium is eached a a join quali y le el o 10, whe e each hos-
pi al makes a p o i o 8 Tale 12 while he o al u ili y o pa ien s (con ibu ion o pa ien s)
is 4 Tale . The pa ien op imum is a a quali y le el o 13, wi h a lowe pe hospi al p o i
o 6 Tale and a highe con ibu ion o he pa ien o 5 Tale . When all hospi als choose he
minimum quali y le el o 1, he join p o i maximum (JPM) is eached. E e y hospi al ge s
a p o i o 14 Tale in his pe iod, and 1 Tale goes o he pa ien popula ion. The bench-
ma k de ec e e s o de ec ing om his collusi e JPM. The op imal de ec ion choice o
pu ely p o i maximizing p o ide s is a quali y le el o 5 o 6,13 bo h leading o p o i s o
15.11 Tale (a ull p o i and pa ien u ili y able can be ound in Appendix A.1).
Nex , we in oduce a measu e o he deg ee o coope a ion wi hin a ma ke . Follow-
ing Sue ens and Po e s (2007) and Po e s and Sue ens (2009), we ansla e hese quali y
benchma ks in o deg ees o coope a ion. This gi es us a s aigh o wa d indica ion abou
how collusi e a heal hca e ma ke ac ually is. The deg ee o coope a ion o heal hca e
ma ke
k
in pe iod
is calcula ed by:
Table 1 Expe imen al condi ions
o e iew
a All obse a ions o he indi idual ma ke s and o 13 eam ma -
ke s a e aken om Han e al. (2017). We an u he sessions o ge
addi ional 25 independen obse a ions o eam ma ke s. To be in
line wi h Han e al. (2017), all sessions consis ed o wo pa s, whe e
he i s pa was an indi idual o eam compe i ion scena io and he
second pa wen along wi h a change in ma ke concen a ion. In
his s udy, we a e only in e es ed in he i s pa , so we can pool he
obse a ions wi hin all indi idual and eam condi ions om Han e al.
(2017). See Appendix A.1.2 o a de ailed explana ion on how we
de i ed he espec i e sample sizes
Condi ions Numbe o
subjec s
Numbe o hos-
pi als
Num-
be o
ma ke sa
Indi idual 213 213 71
Team 342 114 38
To al 555 327 109
11 A quali y le el o ze o is excluded in he design as his would mean non- ea men . T ea ing a pa ien
should a leas lead o he lowes quali y le el possible.
12 Tale is ou expe imen al cu ency uni . The exchange a es di e o he wo condi ions: 1 Tale = 0.07€
in he indi idual ma ke condi ions and 1 Tale = 0.21€ in he eam ma ke condi ions. We do so in o de o
make indi idual ea nings compa able, al hough we inc ease he numbe o pa icipan s pe ma ke om one
o h ee in he eam condi ions. Fo mo e de ail, see Sec ion"Expe imen al p ocedu e".
13 Due o he disc e e pa ame iza ion o ou expe imen , he bes esponse is ambiguous.
340
J.Han e al.
1 3
Gi en ou pa ame iza ion, i holds ha
𝜌k =0
o a e age quali y choices a he non-
coope a i e Nash equilib ium o 10. The JPM would esul in
𝜌k =1
, while a uni o m
quali y choice a he pa ien op imum would esul in
𝜌k =−0.33
.
So a , ou heo e ical amewo k assumes pu e p o i -maximizing hospi als. Howe e ,
when assuming semi-al uis ic hospi als, simila o B ekke e al., (2011, 2017), a hospi al
i
’s objec ion unc ion can be w i en as:
whe e
B
i=
z
i+1
i
∫
o
( +qi− s)ds +
zi−1
∫
o
( +qi− s)
ds
is he o al u ili y o he pa ien s being
ea ed in hospi al
i
and
𝛼>0
is a measu e o he deg ee o al uism.14 The op imal Nash
equilib ium quali y, bes esponse unc ion, and s a egic ela ionship hus depend on he
deg ee o al uism (see Appendix A.3 o he o mal model equa ions). The bes esponse
unc ion allows hospi als’ quali y le els o be s a egic complemen s as well as a s a egic
subs i u es. Howe e , o semi-al uis ic hospi als ha maximize hei objec i e unc ions,
quali y would be a s a egic complemen . Mo eo e , we show ha ou pa ame iza ion
quali y le els o semi-al uis ic hospi als a e highe (i.e.
q∗
i𝛼
>
q∗
i)
and p o i s a e lowe
compa ed o pu e p o i -maximizing hospi als (i.e.
π∗
i𝛼<π∗
i)
.
The ocus o his pape is o be e unde s and s a egic beha iou wi hin hospi al ma -
ke s. The e o e, we nex ansla e ou expec a ions o he deg ee o coope a ion. Gi en ou
pa ame iza ion, he deg ee o coope a ion is
𝜌k =−0.33
o uni o m quali y choices a he
pa ien u ili y op imum o 13 (see Table2). Hence, in con as o p o i -maximizing p o id-
e s, o which he Nash equilib ium p edic s a symme ic quali y choice o 10 ha ans-
la es in o a deg ee o coope a ion o 0, ma ke s wi h semi-al uis ic p o ide s should sup-
ply highe quali y le els and ha e a nega i e deg ee o coope a ion. In Po e s and Sue ens
(2009), nega i e alues o he deg ee o coope a ion
𝜌k
a e in e p e ed as high compe i i e-
ness beyond he Nash equilib ium. In ou se ing, howe e , nega i e alues also co espond
o a e age quali y choices owa ds he pa ien u ili y op imum.15
Hypo hesis 1 (Deg ee o Coope a ion) In a ma ke wi h semi-al uis ic hospi als, we
expec a nega i e deg ee o coope a ion.
Hypo hesis 2 (S a egic Na u e) In ma ke s wi h semi-al uis ic hospi als, we expec ha
hospi als’ quali y le els a e s a egic complemen s.
(6)
𝜌
k =
A g.Ma ke Quali yk −Quali y
Nash
Quali y
JPM
−Quali y
Nash
.
(7)
Ωi
=
(
p−cq
i)
⋅D
i
+𝛼⋅B
i
14 Fo simplici y, we assume homogenei y in he deg ee o al uism.
15 While i would be in e es ing o be able o di e en ia e be ween he d i e s o s a egic beha iou , he
p ima y aim o his pape is o iden i y he la e and o compa e he ou comes o ma ke s wi h an indi-
idual decision p ocess o hose wi h a eam decision p ocess (see below).
347
S a egic beha iou anddecision making incompe i i e hospi al…
1 3
Resul 1 (Co esponding Hypo heses 1 and 3—Deg ee o Coope a ion) Indi idual and
eam ma ke s s a a sligh ly posi i e deg ees o coope a ion, bu hey quickly con e ge o
a e age le els signi ican ly below 0, indica ing pa ien cen ed o compe i i e beha iou .
While he a e age deg ees o coope a ion (quali y le els) a e highe (lowe ) o indi idual
han o eam ma ke s, he di e ence is no signi ican .
Coo dina ion beha iou
So a , we ha e conside ed he agg ega e deg ee o coope a ion. Howe e , he la e does
no shed ligh on ma ke he e ogenei y and whe he ma ke s coo dina e on ce ain quali y
le els. To ge a be e unde s anding o coo dina ion beha iou wi hin ma ke s, we now
conside he di e ence be ween hospi als’ quali y choices wi hin each ma ke . Fo his,
we conside he absolu e di e ence be ween he highes and lowes hospi al quali y o a
espec i e hospi al wi hin each ma ke in e e y pe iod. A small quali y sp ead sugges s ha
hospi als wi hin a ma ke join ly g a i a e owa ds a e e ence quali y le el and indica es
coo dina ion beha iou . Mo eo e , he quali y sp ead ends o be la ge wi h hospi als ha
conside quali y as a s a egic subs i u e a he han as a s a egic complemen . Fo e e -
ence, a p o i maximizing de ia ion om JPM would imply a sp ead o 4 o 5, and mo ing
om he Nash equilib ium o he pa ien op imum would imply a sp ead o 3 (see Table2).
The a e age quali y sp eads in ou expe imen al condi ions can be in e ed om Table6
and Fig.2.27 The quali y sp ead is signi ican ly highe o indi idual decision make s in
he i s i e pe iods, wi h 3.02 quali y uni s on a e age compa ed o 2.30 in eams (MWU,
p = 0.003 o pe iods 1–5). O e ime, he sp eads con e ge o simila le els a ound 2 and
a e no signi ican ly di e en anymo e (MWU, p = 0.889 o pe iods 5–10; MWU, p = 0.957
o pe iods 10–15). I seems ha he eam decision p ocess ‘ il e s ou ’ ex eme decisions
in he beginning and c ea es mo e s able ma ke quali ies o e all. The hi d line in Fig.2
co esponds o he sp ead o quali y p oposals in he majo i y decision p ocess o he eam
decisions. As expec ed, i s end is pa allel o he ac ual eam decision. We he eby con ib-
u e e idence compa ing he coo dina ion beha iou o indi iduals and eams in compe i-
i e ma ke s wi h semi-al uis ic beha iou (see Hypo hesis 4).
While ma ke s end o g a i a e owa d he same quali y le els, hese quali y le els
could be di e en o each ma ke . Figu e3 explo es he he e ogenei y o esul ing a e age
ma ke quali ies in pe iods 11 o 14. Due o end-game e ec s, pe iod 15 is neglec ed. The
le skewness o he dis ibu ion sugges s ha ma ke s coo dina e on le els be ween he
compe i i e and pa ien op imal ou come in he inal pe iods.
Using he a e age s anda d de ia ion o ma ke quali ies, we ind simila esul s as o
quali y sp eads (see Table7). The a e age s anda d de ia ion o ma ke quali ies is signi i-
can ly di e en o eam and indi idual ma ke s (MWU, p = 0.001). This esul is mos ly
d i en by pe iods 1–5 (MWU, p = 0.000). The a e age s anda d de ia ion o ma ke quali-
ies o pe iods 6–10 (MWU, p = 0.122) and o pe iods 11–15 (MWU, p = 0.208) is no sig-
ni ican ly di e en o eam and indi idual ma ke s.
27 The ela i ely high quali y and o e sp eads in he espec i e indi idual and eam ma ke s in he i s
pe iod indica e ha we do no ha e an ancho ing e ec . In case o ancho ing on, e.g. he Nash equilib ium
o pa ien op imal quali y le els, we should see much lowe sp eads.

348
J.Han e al.
1 3
Resul 2 (Co esponding Hypo hesis 4—Coo dina ion) Ini ially, quali y le els a e mo e
homogeneous in eam ma ke s—i.e., he quali y sp ead is signi ican ly highe o indi id-
ual ma ke s han o eam ma ke s. This migh be explained by he p eceding eam deci-
sion p ocess, wi h majo i y o ing il e ing ou ex eme quali y choices om he beginning.
Towa ds he inal pe iods, bo h ma ke s coo dina e a le els be ween he compe i i e and
pa ien op imal ou come.
S a egic na u e o quali y
Ha ing iden i ied some deg ee o pa ien cen edness (compe i i eness), especially o
eam ma ke s, we now in es iga e he di ec ion in which hospi als eac o hei compe i-
o s’ a e age quali y choice. Fo his, we un he ixed e ec s model sepa a ely o indi-
idual and eam ma ke s:
The model in es iga es how s ong and in which di ec ion a hospi al
i
in pe iod
eac s
o he a e age quali y change om i als in he p e ious pe iod. On he le -hand side,
we ha e he di e ence in quali y choice o hospi al
i
be ween he p e ious pe iod
−1
and he cu en pe iod
. The independen a iable is he di e ence in he a e age quali y
choice om pe iod
−2
o
−1
by he compe ing hospi als. We include indi idual ixed
e ec s. The esul s a e p esen ed in Table8. Column (1) shows he esul s o indi idual
ma ke s and column (2) o eam ma ke s. F om column (1), we can in e ha in indi idual
ma ke s he coe icien is posi i e and s a is ically signi ican a he 1 pe cen le el. Hence,
as p edic ed by ou model, quali y choices a e ea ed as s a egic complemen s. This is in
line wi h Hypo hesis 2 and implies ha hospi als maximizing hei objec i e unc ion play
ecip ocally by ollowing he p e ious ma ke quali y le els. F om column (2), we can see
ha he coe icien is s ill posi i e bu smalle and signi ican a he 5 pe cen le el. Hence,
he e ec is weake in he eam condi ions, bu i is s ill in line wi h complemen a i y.
Nex , we sepa a e ma ke s by hei a e age deg ee o coope a ion— ha is, coope a i e
ma ke s wi h an a e age deg ee o coope a ion
𝜌>0
and hose wi h an a e age deg ee o
coope a ion
𝜌≤0
. This migh gi e us clea e insigh in o he ela ionship be ween coop-
e a i eness and s a egic complemen a i y. Columns (3) and (4) compa e indi idual and
eam ma ke s in ma ke s wi h a posi i e deg ee o coope a ion. Fo bo h, he coe icien is
posi i e and s a is ically signi ican a he 1 pe cen le el. Hence, one can in e ha bo h
indi idual and eam coope a i e ma ke s display s a egic complemen a i y. Columns (5)
and (6) conside non-coope a i e indi idual and eam ma ke s wi h an a e age deg ee o
coope a ion
𝜌≤0
. He e, we can in e ha while he coe icien o non-coope a i e indi-
idual ma ke s is s ill highly signi ican , he one o non-coope a i e eam ma ke s is no
signi ican ly di e en om 0. Thus, eams in mo e pa ien cen ed (compe i i e) ma ke s
do no seem o be esponsi e o changes in hei i als’ quali ies bu end o se quali ies
independen ly. We he eby con ibu e e idence compa ing he s a egic na u e o quali y
p o ision o indi iduals and eams in compe i i e ma ke s wi h semi-al uis ic beha iou
(see Hypo hesis 4).28
(12)
Δchoicei =𝛽o+𝛽1Δa g. i alchoice −1+𝜈i+ui .
28 A andom e ec s model on changes in choices, which clus e s on ma ke le el, yields e y simila
esul s.
349
S a egic beha iou anddecision making incompe i i e hospi al…
1 3
Resul 3 (Co esponding Hypo hesis 2—S a egic Na u e) Indi idual hospi al ma ke s
ea quali y le els as s a egic complemen s and adjus hei quali y choice in he same
di ec ion as hei compe i o s. This also holds o eam ma ke s. Howe e , hei esponse
magni ude is weake . This migh be d i en by non-coope a i e o semi-al uis ic eam ma -
ke s ha a e no esponsi e o changes in hei i als’ quali ies bu end o se quali ies
independen ly.
Conclusion
In his pape , we in es iga ed quali y compe i ion in heal hca e ma ke s whe e p o ide s
a e assumed o be al uis ic owa ds pa ien s wi h a ocus on wo aspec s ha a e espe-
cially ele an om a beha iou al pe spec i e in his con ex : i s , how hospi als eac o
a change in hei compe i o s’ quali y choice, and second, how he decision p ocess o
Table 6 A e age quali y
sp eads—indi idual e sus eam
by pe iod in e als on ma ke
le el
N equals he numbe o hospi al ma ke s. An indi idual ma ke con-
sis s o h ee indi idual hospi als and a eam ma ke consis s o h ee
hospi als wi h h ee boa d membe s each
Pe iods
1 1–5 6–10 11–15 15 To al
Indi idual 4.718 3.017 2.403 2.096 1.986 2.505
SD 2.630 1.288 1.461 1.325 1.938 1.407
N 71 71 71 71 71
Team 3.395 2.295 2.342 1.979 1.553 2.205
SD 2.087 0.865 1.186 0.926 0.978 1.006
N 38 38 38 38 38
Fig. 2 Quali y sp ead in indi idual and eam ma ke s
350
J.Han e al.
1 3
quali y choices—i.e., indi idual o eam decision made wi hin hospi als—a ec s quali y
p o ision beha iou . We implemen ed a labo a o y expe imen based on a Salop model,
whe e hospi als compe ed o pa ien s ia quali y p o ision in a ma ke wi h egula ed
p ices. The ma ke consis ed o h ee hospi als. We also implemen ed an expe imen al con-
di ion wi h a eam decision p ocess. He e, hospi al boa ds we e made up o h ee membe s
who had o ag ee on a quali y le el ia a simple majo i y ule. Decisions a ec eal pa ien s
ou side he labo a o y since he gene a ed pa ien u ili y is mone ized and ans e ed o a
cha i y ha p o ides medical ca e o pa ien s wi hou access o o mal heal hca e.
Wi h espec o he gene al compe i ion analysis, we do no ind much e idence o col-
lusion wi hin compe i i e heal hca e ma ke s. In pa icula , we show ha indi idual and
Fig. 3 A e age ma ke quali y be ween pe iods 11 and 14
Table 7 A e age S anda d
De ia ion—Indi idual s. Team
Condi ion by Pe iod In e als on
Ma ke Le el
N equals he numbe o hospi al ma ke s. An indi idual ma ke con-
sis s o h ee indi idual hospi als and a eam ma ke o h ee hospi als
wi h h ee boa d membe s each
Pe iods
1 1–5 6–10 11–15 15 To al
Indi idual 2.483 1.603 1.297 1.139 1.064 1.389
SD 1.376 0.657 0.777 0.696 1.018 0.537
N 71 71 71 71 71
Team 1.57 1.068 1.097 0.926 0.755 1.05
SD 0.957 0.402 0.561 0.426 0.467 0.329
N 38 38 38 38 38
351
S a egic beha iou anddecision making incompe i i e hospi al…
1 3
eam ma ke s s a a sligh ly posi i e deg ees o coope a ion bu hen quickly con e ge
owa ds nega i e alues, indica ing pa ien cen ed (compe i i e) beha iou . This is in line
wi h e idence om p e ious heal h economic expe imen s o duopolies o gene al p ac i-
ione s by B osig-Koch e al. (2017b). They show ha collusion, while being obse ed, is
less equen han in ela ed p ice compe i ion expe imen s wi hou a heal h aming.
Compa ing indi idual ma ke s wi h eam ma ke s, we show ha while he a e age
deg ees o coope a ion (quali y le els) a e desc ip i ely highe (lowe ) o indi idual han
o eam ma ke s, he di e ence is no s a is ically signi ican . To some ex en , his esul
is di e en om Gille e al. (2011) who ind highe p ices o Be and quali y compe i-
ions, which should ansla e in o lowe quali y le els and a highe deg ee o coope a ion
in ou se ing, wi h he majo i y o ing ule compa ed o indi idual decision making. The
e ec in ou se ing migh be o se by semi-al uis ic beha iou . We also show ha ini-
ially, quali y le els a e signi ican ly mo e homogeneous in eam ma ke s han in indi idual
ones. This migh be explained by he p eceding eam decision p ocess wi h majo i y o ing
il e ing ou ex eme quali y choices om he beginning. This is in line wi h e idence om
Bo ns ein and Gneezy (2002) o eam ma ke s and p ice compe i ion wi hou al uism,
showing ha eam ma ke s s abilize much quicke han indi idual ma ke s. Also, Bo n-
s ein e al. (2008) ind ha eam ma ke s end o beha e mo e s able o e all. Towa ds he
inal pe iods, howe e , we show ha bo h ypes o ma ke s coo dina e a le els be ween he
compe i i e and pa ien op imal ou come. Po e s and Sue ens (2009) ind simila endgame
e ec s wi h nega i e deg ees o coope a ion o s a egic subs i u es and deg ees close o
he compe i i e benchma k o complemen s in he inal pe iod in ma ke s wi hou semi-
al uis ic clien s.
We also conside s a egic beha iou and ind ha hospi als led by an indi idual ea
quali y as a s a egic complemen and adjus hei quali y choice in he same di ec ion as
hei compe i o s. This also holds o hospi als led by eams. Howe e , hei esponse mag-
ni ude is weake , d i en by non-coope a i e o al uis ic eam ma ke s, which do no seem
o be esponsi e o changes in hei i als’ quali y le els bu end o se quali y le els inde-
penden ly. Ou esul is in line wi h p e ious e idence by Po e s and Sue ens (2009) o
non-collusi e ma ke s in games wi h s a egic subs i u es. I ein o ces he idea ha coop-
e a ion is easie when goods a e seen as s a egic complemen s, as Hol (1995) and Sue ens
and Po e s (2007) sugges . This inding can also be linked o he heo e ical conside a ions
Table 8 Reg ession esul s on changes in choices o all, coope a i e, and non-coope a i e ma ke s
Fixed e ec s model sepa a ely o indi idual and eam ma ke s. Resul s a e clus e ed on ma ke le el.
S anda d e o s in pa en heses
*p < 0.05; **p < 0.01; ***p < 0.001
(1)
Indi idual
Ma ke s
(2)
Team
Ma ke s
(3)
Indi idual
Coope a-
i e
(4)
Team
Coope a i e
(5)
Indi idual
Non-Coope -
a i e
(6)
Team
Non-Coope a i e
A g. i alchange 0.268*** 0.095* 0.295*** 0.492*** 0.248*** 0.056
(0.032) (0.045) (0.056) (0.011) (0.039) (0.041)
Cons an 0.040*** 0.086*** 0.022*** -0.006** 0.052*** 0.102***
(0.004) (0.008) (0.003) (0.001) (0.006) (0.008)
N 2769 1482 936 156 1833 1326
Adjus ed R20.035 0.004 0.040 0.091 0.031 0.001
352
J.Han e al.
1 3
o B ekke e al. (2017) ega ding compe i i e heal hca e ma ke s wi h al uis ic p o ide s.
They a gue ha quali y is a s a egic complemen as long as al uism is ela i ely low com-
pa ed o he cos subs i u abili y be ween quali y and ou pu , while s ong al uism could
yield quali y as a s a egic subs i u e. Howe e , we do no ind such ex eme esul s excep
o ou mos al uis ic ma ke s—i.e., non-coope a i e eam ma ke s. He e, quali y is no
longe ea ed as a complemen bu is s a egically independen . We he eby also con ibu e
o ecen ield s udies in es iga ing he s a egic na u e o quali y wi hin hospi al ma ke s.
In line wi h ou esul s, hey ind ei he no esponse o i als’ quali y in hospi al se ings
o some e idence o complemen a i y o a ew selec ed quali y indica o s (G a elle e al.,
2014; Longo e al., 2017). Gi en ou expe imen al esul s, his migh be d i en by he hos-
pi als’ deg ee o coope a i eness o al uism wi hin he ma ke .
O e all, ou esul s con ibu e no el expe imen al e idence ega ding compe i ion in
heal hca e ma ke s. While he p e ious li e a u e sugges s ha he highe he deg ee o
obse abili y o hospi al quali y indica o s o pa ien s, he s onge he hospi als’ esponse
o a i al hospi al’s quali y (see, o example, G a elle e al., 2014), ou indings sugges
ha hospi als’ s a egic quali y p o ision beha iou migh also depend on he deg ee o
al uism and he o ganiza ional s uc u e— ha is, whe he he quali y decisions o he hos-
pi al a e made by indi iduals o by eams. While egula o s o en os e quali y imp o e-
men in heal hca e wi h incen i es o compe i ion—e.g., by add essing ma ke concen a-
ion o public epo ing o quali y da a— he o ganiza ional s uc u e o hospi als migh be
an addi ional channel o conside .
Fu u e esea ch could analyse in mo e de ail o wha ex en indi idual cha ac e is ics
(e.g. socio-demog aphics, isk a i udes, and pe sonali y ai s) migh change he o e all
indings on he s a egic na u e o quali y. Since ele an cha ac e is ics can di e o he
indi idual ma ke si ua ion and he majo i y o ing p ocess, his could be help ul in inding
ou wha d i es quali y decisions in he espec i e si ua ions.
Supplemen a y In o ma ion The online e sion con ains supplemen a y ma e ial a ailable a h ps:// doi.
o g/ 10. 1007/ s10754- 024- 09366-3.
Funding Open Access unding enabled and o ganized by P ojek DEAL. Financial suppo p o ided by
he Bundesminis e ium ü Bildung und Fo schung (BMBF, Ge man Fede al Minis y o Educa ion and
Resea ch) (g an numbe : 01 EH1602A) is g a e ully acknowledged.
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 Com-
mons 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
Ash on, M. C., & Lee, K. (2009). The HEXACO-60: A sho measu e o he majo dimensions o pe sonal-
i y. Jou nal o Pe sonali y Assessmen , 91, 340–345.
Ba os, P. P., & Oli ella, P. (2011). Hospi als: Teaming up. In The Ox o d handbook o heal h economics
(pp. 432–462). h ps:// doi. o g/ 10. 1093/ ox o dhb/ 97801 99238 828. 013. 0019

353
S a egic beha iou anddecision making incompe i i e hospi al…
1 3
Bloom, N., P oppe , C., Seile , S., & Van Reenen, J. (2015). The impac o compe i ion on managemen
quali y: E idence om public hospi als. The Re iew o Economic S udies, 82(2), 457–489.
Bo ns ein, G., & Gneezy, U. (2002). P ice compe i ion be ween eams. Expe imen al Economics, 5(1),
29–38.
Bo ns ein, G., Kugle , T., Budescu, D. V., & Sel en, R. (2008). Repea ed p ice compe i ion be ween
indi iduals and be ween eams. Jou nal o Economic Beha io and O ganiza ion, 66(3–4), 808–
821. h ps:// doi. o g/ 10. 1016/j. jebo. 2006. 06. 004
B ekke, K. R., Siciliani, L., & S aume, O. R. (2011). Hospi al compe i ion and quali y wi h egula ed
p ices. Scandina ian Jou nal o Economics, 113(2), 444–469. h ps:// doi. o g/ 10. 1111/j. 1467- 9442.
2011. 01647.x
B ekke, K. R., Siciliani, L., & S aume, O. R. (2017). Hospi al me ge s wi h egula ed p ices. Scandina-
ian Jou nal o Economics, 119(3), 597–627. h ps:// doi. o g/ 10. 1111/ sjoe. 12191
B osig-Koch, J., Hennig-Schmid , H., Kai ies-Schwa z, N., Koko , J., & Wiesen, D. (2024). A new look
a physicians’ esponses o inancial incen i es: Quali y o ca e, p ac ice cha ac e is ics, and mo i-
a ions. Jou nal o Heal h Economics, 102862.
B osig-Koch, J., Hehenkamp, B., & Koko , J. (2017b). The e ec s o compe i ion on medical se ice
p o ision. Heal h Economics, 26, 6–20. h ps:// doi. o g/ 10. 1002/ hec. 3583
B osig-Koch, J., Hennig-Schmid , H., Kai ies-Schwa z, N., & Wiesen, D. (2016). Using a e- ac ual
ield and lab expe imen s o in es iga e how ee- o -se ice and capi a ion a ec medical se ice
p o ision. Jou nal o Economic Beha io and O ganiza ion, 131, 17–23. h ps:// doi. o g/ 10. 1016/j.
jebo. 2015. 04. 011
B osig-Koch, J., Hennig-Schmid , H., Kai ies-Schwa z, N., & Wiesen, D. (2017a). The e ec s o in o-
ducing mixed paymen sys ems o physicians: Expe imen al e idence. Heal h Economics, 26, 243–
262. h ps:// doi. o g/ 10. 1002/ hec. 3292
Chalkley, M., & Malcomson, J. M. (1998). Con ac ing o heal h se ices when pa ien demand does
no e lec quali y. Jou nal o Heal h Economics, 17(1), 1–19. h ps:// doi. o g/ 10. 1016/ S0167-
6296(97) 00019-2
Choné, P., & Ma, C. A. (2011). Op imal heal h ca e con ac unde physician agency. Annals o Eco-
nomics and S a is ics/annales D’économie E De S a is ique, 101(102), 229–256. h ps:// doi. o g/ 10.
2307/ 41615 481
Coope , D. J., & Kagel, J. H. (2005). A e wo heads be e han one? Team e sus indi idual play in
signaling games. Ame ican Economic Re iew, 95(3), 477–509. h ps:// doi. o g/ 10. 1257/ 00028 28054
201431
Di Guida, S., Gy d-Hansen, D., & Oxholm, A. S. (2019). Tes ing he my h o ee- o -se ice and o e -
p o ision in heal h ca e. Heal h Economics, 28, 717–722.
Dohmen, T., Falk, A., Hu man, D., Sunde, U., Schupp, J., & Wagne , G. G. (2011). Indi idual isk
a i udes: Measu emen , de e minan s, and beha io al consequences. Jou nal o he Eu opean Eco-
nomic Associa ion, 9(3), 522–550.
D ano e, D., & Zhe Jin, G. (2010). Quali y disclosu e and ce i ica ion: Theo y and p ac ice. Jou nal o
Economic Li e a u e, 48(4), 935–963. h ps:// doi. o g/ 10. 1257/ jel. 48.4. 935
Eckel, C. C., & G ossman, P. J. (1996). Al uism in anonymous dic a o games. Games and Economic
Beha io , 16(2), 181–191. h ps:// doi. o g/ 10. 1006/ game. 1996. 0081
Ellis, R. P., & McGui e, T. G. (1986). P o ide beha io unde p ospec i e eimbu semen : Cos sha ing
and supply. Jou nal o Heal h Economics, 5(2), 129–151. h ps:// doi. o g/ 10. 1016/ 0167- 6296(86)
90002-0
Ellis, R. P., & McGui e, T. G. (1990). Op imal paymen sys ems o heal h se ices. Jou nal o Heal h
Economics, 9(4), 375–396. h ps:// doi. o g/ 10. 1016/ 0167- 6296(90) 90001-J
Engel, C. (2010). The beha iou o co po a e ac o s: How much can we lea n om he expe imen al
li e a u e? Jou nal o Ins i u ional Economics, 6(4), 445–475. h ps:// doi. o g/ 10. 1017/ S1744 13741
00001 35
Fischbache , U. (2007). z- ee: Zu ich oolbox o eady-made economic expe imen s. Expe imen al
Economics, 10(2), 171–178. h ps:// doi. o g/ 10. 1007/ s10683- 006- 9159-4
Gayno , M., & Town, R. J. (2012). Compe i ion in heal h ca e ma ke s. In Handbook o heal h econom-
ics (Vol. 2, pp. 499–637). Else ie .
Gayno , M., Ho, K., & Town, R. J. (2015). The indus ial o ganiza ion o heal h-ca e ma ke s. Jou nal o
Economic Li e a u e, 53(2), 235–284.
Gille , J., Sch am, A., & Sonnemans, J. (2011). Ca el o ma ion and p icing: The e ec o manage ial
decision-making ules. In e na ional Jou nal o Indus ial O ganiza ion, 29(1), 126–133. h ps://
doi. o g/ 10. 1016/j. ijind o g. 2010. 03. 015
354
J.Han e al.
1 3
Godage , G., & Wiesen, D. (2013). P o i o pa ien s’ heal h bene i ? Explo ing he he e ogenei y in
physician al uism. Jou nal o Heal h Economics, 32(6), 1105–1116. h ps:// doi. o g/ 10. 1016/j. jheal
eco. 2013. 08. 008
G a elle, H., San os, R., & Siciliani, L. (2014). Does a hospi al’s quali y depend on he quali y o o he
hospi als? A spa ial econome ics app oach. Regional Science and U ban Economics, 49, 203–216.
h ps:// doi. o g/ 10. 1016/j. egsc iu be co. 2014. 09. 005
G een, E. P. (2014). Paymen sys ems in he heal hca e indus y: An expe imen al s udy o physician incen-
i es. Jou nal o economic beha io & o ganiza ion, 106, 367–378.
G eine , B. (2015). Subjec pool ec ui men p ocedu es: O ganizing expe imen s wi h ORSEE. Jou nal
o he Economic Science Associa ion, 1(1), 114–125. h ps:// doi. o g/ 10. 1007/ s40881- 015- 0004-4
Guccio, C., & Lisi, D. (2016). Thus do all. Social in e ac ions in inapp op ia e beha io o childbi h
se ices in a highly decen alized heal hca e sys em. Regional Science and U ban Economics, 61,
1–17. h ps:// doi. o g/ 10. 1016/j. egsc iu be co. 2016. 08. 004
Han, J., Kai ies-Schwa z, N., & Vomho , M. (2017). Quali y compe i ion and hospi al me ge s: An
expe imen . Heal h Economics, 26, 36–51. h ps:// doi. o g/ 10. 1002/ hec. 3574
Ha is, J. E. (1977). The in e nal o ganiza ion o hospi als: some economic implica ions. The Bell Jou -
nal o Economics. h ps:// doi. o g/ 10. 2307/ 30032 97
Hennig-Schmid , H., Sel en, R., & Wiesen, D. (2011). How paymen sys ems a ec physicians’ p o-
ision beha iou : An expe imen al in es iga ion. Jou nal o Heal h Economics, 30(4), 637–646.
h ps:// doi. o g/ 10. 1016/j. jheal eco. 2011. 05. 001
Hennig-Schmid , H., & Wiesen, D. (2014). O he - ega ding beha iou and mo i a ion in heal h ca e p o-
ision: An expe imen wi h medical and non-medical s uden s. Social Science and Medicine, 108,
156–165.
Hol , C. A. (1995). Indus ial o ganiza ion: A su ey o labo a o y esea ch. The Handbook o Expe i-
men al Economics, 349, 402–403.
Huck, S., Kon ad, K. A., Mülle , W., & No mann, H. T. (2007). The me ge pa adox and why aspi a ion
le els le i ail in he labo a o y. The economic jou nal, 117(522), 1073–1095.
Jack, W. (2005). Pu chasing heal h ca e se ices om p o ide s wi h unknown al uism. Jou nal o
Heal h Economics, 24(1), 73–93. h ps:// doi. o g/ 10. 1016/j. jheal eco. 2004. 06. 001
Kai ies-Schwa z, N., & Soucek, C. (2020). Pe o mance pay in hospi als: An expe imen on bonus-
malus incen i es. In e na ional Jou nal o En i onmen al Resea ch and Public Heal h, 17(22),
8320.
Kes e nich, I., Schumache , H., & Win e , J. (2015). P o essional no ms and physician beha io : Homo
oeconomicus o homo hippoc a icus? Jou nal o Public Economics, 131, 1–11. h ps:// doi. o g/ 10.
1016/j. jpube co. 2015. 08. 009
Koche , M. G., & Su e , M. (2007). Indi idual e sus g oup beha io and he ole o he decision mak-
ing p ocedu e in gi -exchange expe imen s. Empi ica, 34(1), 63–88. h ps:// doi. o g/ 10. 1007/
s10663- 006- 9026-8
Kugle , T., Kausel, E. E., & Koche , M. G. (2012). A e g oups mo e a ional han indi iduals? A e iew
o in e ac i e decision making in g oups. Wiley In e disciplina y Re iews: Cogni i e Science, 3(4),
471–482. h ps:// doi. o g/ 10. 1002/ wcs. 1184
Laga de, M., & Blaauw, D. (2017). Physicians’ esponses o inancial and social incen i es: A medically
amed eal e o expe imen . Social Science and Medicine, 179, 147–159.
Liu, T., & Ma, C. A. (2013). Heal h insu ance, ea men plan, and delega ion o al uis ic physician.
Jou nal o Economic Beha io and O ganiza ion, 85, 79–96. h ps:// doi. o g/ 10. 1016/j. jebo. 2012.
11. 002
Longo, F., Siciliani, L., G a elle, H., & San os, R. (2017). Do hospi als espond o i als’ quali y and
e iciency? A spa ial panel econome ic analysis. Heal h Economics, 26, 38–62. h ps:// doi. o g/ 10.
1002/ hec. 3569
Ma insson, P., & Pe sson, E. (2019). Physician beha io and condi ional al uism: The e ec s o pay-
men sys em and unce ain heal h bene i . Theo y and Decision, 87, 365–387.
Mo a, M. (2004). Compe i ion policy: Theo y and p ac ice. Camb idge Uni e si y P ess.
Mülle , W., & Tan, F. (2013). Who ac s mo e like a game heo is ? G oup and indi idual play in a
sequen ial ma ke game and he e ec o he ime ho izon. Games and Economic Beha io , 82,
658–674. h ps:// doi. o g/ 10. 1016/j. geb. 2013. 09. 007
Pauly, M. V., & Redisch, M. (1973). The no - o -p o i hospi al as a physicians’ coope a i e. Ame ican
Economic Re iew, 63(1), 87–99.
Po e s, J., & Sue ens, S. (2009). Coope a ion in expe imen al games o s a egic complemen s and sub-
s i u es. The Re iew o Economic S udies, 76(3), 1125–1147.
355
S a egic beha iou anddecision making incompe i i e hospi al…
1 3
Po e s, J., & Sue ens, S. (2013). Oligopoly expe imen s in he cu en millennium. Jou nal o Economic
Su eys, 27(3), 439–460. h ps:// doi. o g/ 10. 1111/ joes. 12025
Raab, P., & Schippe , B. C. (2009). Cou no compe i ion be ween eams: An expe imen al s udy. Jou nal
o Economic Beha io and O ganiza ion, 72(2), 691–702. h ps:// doi. o g/ 10. 1016/j. jebo. 2009. 07.
004
Rei , S., Ha ne , L., & Seebaue , M. (2020). Physician beha iou unde p ospec i e paymen schemes-
e idence om a e ac ual ield and lab expe imen s. In e na ional Jou nal o En i onmen al
Resea ch and Public Heal h, 17(15), 5540.
Salop, S. (1979). Monopolis ic compe i ion wi h ou side goods. The Bell Jou nal o Economics, 10(1),
141–156. h ps:// doi. o g/ 10. 2307/ 30033 23
Sue ens, S., & Po e s, J. (2007). Be and colludes mo e han Cou no . Expe imen al Economics, 10(1),
71–77.
Su e , M. (2009). Indi idual beha io and g oup membe ship: Commen . Ame ican Economic Re iew,
99(5), 2247–2257. h ps:// doi. o g/ 10. 1007/ s10683- 006- 9132-2
Ti ole, J. (1988). The heo y o indus ial o ganiza ion. MIT P ess.
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.