Heida y, Kimia; an de Res , Jean-Pie e; Cus e s, B. H. M.
A icle
Disc imina ion g ounds and pe sonalised p icing:
Consume pe cep ions o ai ness, no m alignmen ,
legali y, and us in ma ke s
In e ne Policy Re iew
P o ided in Coope a ion wi h:
Alexande on Humbold Ins i u e o In e ne and Socie y (HIIG), Be lin
Sugges ed Ci a ion: Heida y, Kimia; an de Res , Jean-Pie e; Cus e s, B. H. M. (2024) : Disc imina ion
g ounds and pe sonalised p icing: Consume pe cep ions o ai ness, no m alignmen , legali y,
and us in ma ke s, In e ne Policy Re iew, ISSN 2197-6775, Alexande on Humbold Ins i u e o
In e ne and Socie y, Be lin, Vol. 13, Iss. 4, pp. 1-37,
h ps://doi.o g/10.14763/2024.4.1809
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/312558
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/3.0/de/deed.en
Volume 13 |
Disc imina ion g ounds and pe sonalised
p icing: Consume pe cep ions o ai ness,
no m alignmen , legali y, and us in
ma ke s
Kimia Heida y Leiden Uni e si y
Jean-Pie e an de Res Leiden Uni e si y
Ba Cus e s Leiden Uni e si y
DOI: h ps://doi.o g/10.14763/2024.4.1809
Published: 18 Oc obe 2024
Recei ed: 10 Oc obe 2023 Accep ed: 26 Ma ch 2024
Compe ing In e es s: The au ho has decla ed ha no compe ing in e es s exis ha
ha e in luenced he ex .
Licence: This is an open-access a icle dis ibu ed unde he e ms o he C ea i e
Commons A ibu ion 3.0 License (Ge many) which pe mi s un es ic ed use,
dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly
ci ed. h ps://c ea i ecommons.o g/licenses/by/3.0/de/deed.en
Copy igh emains wi h he au ho (s).
Ci a ion: Heida y, K. & an de Res , J.-P. & Cus e s, B. (2024). Disc imina ion g ounds
and pe sonalised p icing: Consume pe cep ions o ai ness, no m alignmen , legali y,
and us in ma ke s. In e ne Policy Re iew, 13(4). h ps://doi.o g/10.14763/
2024.4.1809
Keywo ds: Pe sonalised p icing, Consume p o iling, P ice disc imina ion, No ms, An i-
disc imina ion law
Abs ac : This a icle explo es consume pe cep ions o di e en g ounds by which online p ices
can be pe sonalised. We conduc ed a su ey among Du ch consume s (n = 727) p esen ing hem
wi h 25 segmen a ion bases, d awing om legally pe missible and legally p ohibi ed g ounds. We
hen anked hese bases and accompanying consume pe cep ions ac oss i e dimensions: ai ness,
alignmen wi h pe sonal no ms, alignmen wi h social no ms, pe cei ed legali y, and us in
ma ke s. We ind ha while consume pe cep ions gene ally align wi h wha is cu en ly p ohibi ed
in law, he e a e some “new” g ounds, in pa icula in elligence and physical appea ance, ha elici
simila nega i e pe cep ions as legally p ohibi ed g ounds. This aises ques ions ega ding he
u he egula ion o pe sonalised p icing. We discuss p os and cons o upda ing legisla ion o
be e e lec (new) e hical and social no ms.
Issue 4
1. In oduc ion
Online p ice disc imina ion o pe sonalised p icing, he p ac ice o se ing p ices
based on consume cha ac e is ics and beha iou , wi h he help o AI-based ech-
nologies and big da a p ocessing, is slowly becoming a eali y. The apidly inc eas-
ing low o pe sonal da a, in combina ion wi h echnological de elopmen s ha
make i possible o analyse and alue hese da a, makes i inc easingly possible o
companies o pe sonalise p ices based on wha hey ( hink hey) know abou con-
sume s. These da a a e obse ed, de i ed om beha iou , o p o ided by con-
sume s hemsel es and can be in o ma ion abou a pe son's de ice ype o pu -
chasing beha iou , he ype o pages hey iew, social media da a, and e en he
ba e y pe cen age o he de ice ha is used o su he In e ne (OECD, 2018a).
Especially when hese da a a e combined, companies can build mo e sophis ica ed
p o iles and expe imen wi h p icing acco dingly. Pe sonalising p ices and di e en-
ia ing be ween di e en ypes o consume s can be qui e p o i able o companies
(Odlyzko, 2003; Shille , 2014). As he p ac ice con inues o e ol e, consume p o-
iles and he unde lying echnology a e expec ed o become mo e sophis ica ed
(OECD, 2018a).
Di e en ia ing be ween consume s and he p ice hey pay is no a new phenome-
non. Examples o cha ging ships di e en ligh house ees can al eady be ound in
13 h cen u y England (Odlyzko, 2004). F om an economics pe spec i e, such p ac-
ice is iewed in a neu al way. The e m “p ice disc imina ion” ep esen s a alue
ee concep , wi hou he no ma i e (o en nega i e) conno a ions ha a common
unde s anding o he wo d “disc imina ion” sugges s in o he disciplines such as
law (S eppe, 2017). P ice disc imina ion in an economic sense is no always ha m-
ul and can e en be bene icial in e ms o opening he ma ke o consume s who
migh no ha e been able o pu chase a p oduc o se ice unde a uni o m p icing
sys em (Sau e , 2020). A i s glance, economic ex books p esen qui e innocen
and gene ally accep ed ways o dis inguishing be ween consume s. Fo example,
s uden s and he elde ly may be cha ged educed p ices by o e ing discoun s. Loy-
al cus ome s who a e o e ed loyal y discoun s is ano he well-known example. In
addi ion, p ices can di e when p oduc s do no ha e a ixed p ice, such as a a ca
deale ship o s ee endo , whe e consume s wi h high nego ia ing skills may
walk away wi h be e deals han consume s who do no know how o e p iced he
p oduc is o who canno hide hei in e es well enough.
Many o ms o p ice disc imina ion a e much less accep ed. Empi ical esea ch and
anecdo al e idence indica e ha disc imina o y p icing o en mee s wi h esis-
ance om consume s. They gene ally iew he p ac ice as un ai and deem i o
2 In e ne Policy Re iew 13(4) | 2024
be illegi ima e (Poo & Zuide een Bo gesius, 2019; Tu ow e al., 2005; P ies e e
al., 2020). Some examples, al hough deeply oo ed in ou daily li es, s ill aise de-
ba es abou accep abili y and ai ness. A ypical example includes dis inguishing
be ween consume s based on gende , wi h women o en ge ing he sho end o
he s ick and paying a highe p ice o p oduc s and se ices: he so-called “pink
ax” (Depa men o Consume A ai s, 2015). Ano he example is he selec i e
cha ging o highe p ices, o example o consume s who li e in a weal hie
neighbou hood o a “ ine” o loyal cus ome s (Maxwell & Ga ba ino, 2010; CMA,
2020). Social no ms a e an impo an explana ion why some p ac ices a e deemed
mo e (un) ai han o he s. Consume s ha pe cei e a p icing p ac ice o iola e a
social no m, eac nega i ely owa ds he company, such as lowe pu chase in en-
ions (Campbell, 1999) and loss o us (Ga ba ino & Lee, 2003).
The cu en Eu opean legal amewo k does no explici ly p ohibi online p ice dis-
c imina ion, bu i does pose bounda ies as o he igh ul use and p ocessing o
da a o pe sonalising p ices and he anspa ency ha needs o be p o ided by
companies who engage in he p ac ice.1 Se e al au ho s ha e ques ioned whe he
he cu en Eu opean legal amewo k is equipped o add ess he challenges asso-
cia ed wi h online p ice disc imina ion (Sea s, 2021; Ba os Vale, 2020; G ochowski
e al. 2022). One o hese challenges is he disc imina ion ha can occu wi h he
use o AI-based p icing algo i hms. An i-disc imina ion law and da a p o ec ion
law p o ec ce ain “sensi i e” cha ac e is ics and undamen al choices ha canno
be changed, o a leas no wi hou high damage o one’s co e iden i y, such as e-
ligion, gende , and sexual o ien a ion (Cla ke, 2015; Khai an, 2015). The use o
such g ounds in pe sonalised p icing would cons i u e (in)di ec disc imina ion and
likely elici nega i e eac ions om consume s. E en i companies make e o s no
o include such sensi i e (syn he ic) da a in hei analyses, he echnology used
can s ill indi ec ly lead o he sys ema ic disad an aging o ce ain (g oups o )
consume s (Calde s & Žliobai ė, 2013). In o he wo ds: “insensi i e” (non-legally
p ohibi ed) da a can show s ong associa ions wi h sensi i e o legally p ohibi ed
da a (Li, 2022). Mo eo e , some “newe ” g ounds o p o iles ha could be used o
pe sonalised p icing, such as in elligence o b owsing ac i i y, could also cons i-
u e simila immu abili y and ulne abili y o a e a leas no wi hin a pe son’s
con ol, which could wa an u he egula o y p o ec ion (Wach e , 2022). In o de
o assess o wha ex en he cu en legal amewo k is aligned wi h social no ms
and pe cep ions su ounding pe sonalised p icing, i is impe a i e o s udy exis ing
ai ness pe cep ions ega ding bo h legally p ohibi ed and pe missible g ounds.
1. See in his ega d also A icle 6(1)(ea) o he Consume Righ s Di ec i e, which in oduces an in o -
ma ion equi emen o companies engaging in pe sonalised p icing.
3 Heida y, an de Res , Cus e s
The li e a u e ends o pay li le a en ion o he a ious g ounds o disc imina-
ion on he basis o which p ices can be pe sonalised. The e is wo k ha has ex-
plo ed consume pe cep ions o pe sonalised p icing (e.g. Tu ow e al., 2005; Poo
& Zuide een Bo gesius, 2019), as well as some speci ic non-p ohibi ed g ounds
(e.g. loca ion, de ice ype, pu chase his o y) o pe sonalised p icing (P ies e e al.,
2020; Hu nagel e al., 2022). Ye , a mo e comp ehensi e e iew o pe cep ions e-
ga ding disc imina ion g ounds – legally p ohibi ed o no – is missing. While
some g ounds and he p ocessing he eo a e p o ec ed unde da a p o ec ion law
and an i-disc imina ion law, he ques ion emains how well he legally p o ec ed
g ounds align wi h consume (un) ai ness pe cep ions and he ex en o which he
EU legal amewo k can be le e aged o p o ide u u e p o ec ion agains disc imi-
na ion and/o exploi a ion on he basis o hese g ounds – especially gi en ha
hese new segmen a ion c i e ia (i.e. disc imina ion g ounds ha a e no legally
p ohibi ed) could s ill ha e (un o eseen) legally disc imina o y ou comes. The o-
cus o his a icle is mos ly on he ole o an i-disc imina ion law, as he g ounds
ha we examine low mos ly om his amewo k. I builds on he exis ing deba e
ega ding he ole ha an i-disc imina ion law can play o add ess he disc imina-
ion ha can esul om he use o p icing algo i hms (Tanna & Dunning, 2023;
Wach e , 2022) Howe e , we belie e ha he indings ha e implica ions o o he
ields o law, pa icula ly ega ding en o cemen .
The e o e, he aim o his esea ch is wo old. Fi s , we p o ide a comp ehensi e
o e iew o consume pe cep ions ega ding known g ounds ha can be used o
pe sonalise p ices. He e, we do no only analyse g ounds ha a e al eady p ohibi -
ed o decla ed sensi i e by law, bu also pe cep ions ega ding “new” g ounds (i.e.
g ounds ha ha e mo e ecen ly come o play and a e no legally p ohibi ed), as
hey a e hea ily in e wined (Solo e, 2024). Second, using his o e iew, we exam-
ine whe he he e is a gap be ween pe cep ions o disc imina ion g ounds ha a e
p ohibi ed, and hose ha a e no legally p ohibi ed. Ou indings p o ide inpu o
he discussion on he ex en o which cu en legisla ion o e s app op ia e sa e-
gua ds agains pe sonalised p icing and which g ounds, i any, should be be e
p o ec ed in he u u e (e.g. Van de Res e al., 2020). We explo ed consume pe -
spec i es by conduc ing a su ey among 727 Du ch consume s.
The pape is s uc u ed as ollows. In Sec ion 2, we e iew he cu en li e a u e on
pe sonalised p icing and he cu en legal amewo k. In Sec ion 3, we explain he
me hodology used in he empi ical s udy. Sec ion 4 p esen s he su ey esul s and
Sec ion 5 discusses he implica ions o he indings o ou s udy, pa icula ly he
ex en o which he e is a gap be ween he cu en legal landscape and he pe -
4 In e ne Policy Re iew 13(4) | 2024
cep ions o consume s, and whe he he legal amewo k should be econside ed,
including a enues o u u e esea ch. In Sec ion 6, we p o ide conclusions.
2. Li e a u e e iew
2.1 Online p ice disc imina ion
Online p ice disc imina ion is gene ally de ined as cha ging di e en p ices o con-
sume s o he same p oduc , based on in e ences d awn om consume da a wi h
he help o p icing algo i hms and big da a p ocessing (Zuide een Bo gesius &
Poo , 2019). He e, he di e ence in p ice canno be explained by a di e ence in
cos s, bu a he he in o ma ion ha he company has abou i s (p ospec i e)
clien s (Ca oll & Coa es, 1999). P ice di e ences due o, o example, highe ship-
ping o deli e y cos s, o due o insu ed pa ies posing a highe isk o an insu e ,
a e no conside ed p ice disc imina ion (P es on McA ee, 2008). The easoning be-
hind p ice disc imina ion is ha consume s a y in hei willingness o pay o
p oduc s and se ices, as hey alue hese di e en ly (OECD, 2016). Companies ob-
se e such di e ences in consume s’ o en luc ua ing p e e ences and assess he
willingness o pay acco dingly. These assessmen s do no ha e o be accu a e o
he penny in o de o be conside ed p ice disc imina ion (OECD, 2018a). I he e is
no disce nible di e ence be ween consume s and hei willingness o pay, i is o -
en no p o i able o a company o oll ou a disc imina o y p icing s a egy (S ole,
2007).
When c ea ing and op imising segmen s, companies can use a ple ho a o con-
sume da a. These da a include, bu a e no limi ed o, obse ed da a (e.g. in o ma-
ion abou de ice, pu chase his o y), olun ee ed da a (e.g. name, gende ), and in-
e ed da a (e.g. income, li e phase (OECD, 2018b)). Da a o iginally collec ed o
o he pu poses can be eused o his (Cus e s & Bachlechne , 2017). Sec ion 2.2
will map an o e iew o legally p ohibi ed g ounds o disc imina ion, on which
companies could in heo y base hei analyses.
2.2 Legally p ohibi ed g ounds o disc imina ion
The cu en Eu opean legal landscape does no explici ly p ohibi pe sonalised
p icing; he only di ec men ion o pe sonalised p icing can be ound in Di ec i e
2019/2161,2 on he basis o which companies a e equi ed o disclose use o au o-
2. Di ec i e (EU) 2019/2161 o he Eu opean Pa liamen and o he Council o 27 No embe 2019
amending Council Di ec i e 93/13/EEC and Di ec i es 98/6/EC, 2005/29/EC and 2011/83/EU o he
Eu opean Pa liamen and o he Council as ega ds he be e en o cemen and mode nisa ion o
Union consume p o ec ion ules.
5 Heida y, an de Res , Cus e s
ma ed pe sonalised p icing o consume s. In e es ingly, companies a e no e-
qui ed o disclose he pa ame e s used o pe sonalised p icing, only ha he p ice
has been pe sonalised. Ne e heless, he e a e wo conc e e s a ing poin s o
p o ec ed o sensi i e g ounds o p ice disc imina ion. These can be ound in an i-
disc imina ion law and da a p o ec ion law. We will zoom in on he Du ch cons i u-
ion and i s enume a ion o disc imina ion g ounds, as i was ecen ly upda ed o
include new g ounds o disc imina ion and concu en ly, hese we e he g ounds
ha we p esen ed o su ey esponden s.
The Cha e o Fundamen al Righ s o he Eu opean Union (CFEU), which en e ed
in o o ce on 1 Decembe 2009, ensh ines a ious undamen al igh s, eedoms,
and p inciples. The p inciple o equal ea men is p o ec ed in A icle 21 CFEU,
which s a es ha any disc imina ion shall be p ohibi ed and p o ides a non-ex-
haus i e enume a ion o ou een disc imina ion g ounds, mo e han any na ional
cons i u ion o Eu opean membe S a es. The disc imina ion g ounds men ioned
include sex, ace, gene ic ea u es, eligion, disabili y, and age. The a ionale be-
hind hese p o ec ed cha ac e is ics is o p o ec bo h immu able cha ac e is ics
(i.e. cha ac e is ics ha we e no chosen and canno be changed, o a leas no
wi hou high damage o one’s co e iden i y), such as gende and e hnici y, and un-
damen al choices, such as eligion (Cla ke, 2015; Khai an, 2015). Al hough he
scope o he CFEU is o mally es ic ed o EU ins i u ions when implemen ing EU
law (A icle 51 CFEU), he CJEU has ecognized ho izon al e ec s in ecen ju-
isp udence (Mui , 2019).3
The Du ch na ional cons i u ion p o ides a simila non-exhaus i e enume a ion,
al hough wi h less examples lis ed. A icle 1 o he na ional cons i u ion
(G ondwe ) s a es ha all pe sons in he Ne he lands shall be ea ed equally in
equal ci cums ances. Disc imina ion on he g ounds o eligion, belie , poli ical
opinion, ace o sex, o any o he g ounds wha soe e shall no be pe mi ed. As o
2023, he Du ch cons i u ion includes wo new g ounds o disc imina ion: disabili-
y and sexual o ien a ion (Co de , 2023). The Equal T ea men Ac (AWGB), which
da es om 1994 and cons i u es seconda y legisla ion, u he elabo a es on A i-
cle 1 o he cons i u ion. The AWGB p ohibi s making a dis inc ion on he basis o
an exhaus i e enume a ion o sensi i e c i e ia. The mos undamen al objec i e o
his Ac is he p o ec ion o human digni y, mo e speci ically o p omo e equal pa -
icipa ion in socie y, wi hou being subjec ed o disc imina ion o exclusion on he
basis o pe sonal cha ac e is ics.4 The exhaus i e enume a ion in A icle 1 p o-
3. See o example Case C-414/16 Ve a Egenbe ge E angelisches We k ü Diakonie und En wick-
lung eV, §81.
6 In e ne Policy Re iew 13(4) | 2024
hibi s disc imina ion on he basis o eligion, belie , poli ical opinion, ace, gende ,
na ionali y, sexual o ien a ion, and ma i al s a us. In la e subo dina e legisla ion
some addi ional g ounds we e added, including disabili y o ch onic illness, age,
and ype o employmen con ac .5 I is impo an o no e ha h oughou Eu o-
pean cons i u ions, he e exis s a clea lack o ha monisa ion o disc imina ion
g ounds and whe he an exhaus i e o non-exhaus i e enume a ion is used
(Cus e s, 2023). Fo ins ance, weal h and social s a us a e a p o ec ed cha ac e is-
ic in 11 ou o 27 Membe S a es bu a e no men ioned in A icle 21 CFEU.
In addi ion o an i-disc imina ion law, da a p o ec ion law also p o ides legal
bounda ies o he use o ce ain g ounds in pe sonalising p icing. Al hough he
GDPR does no di ec ly deal wi h bias and disc imina ion ha can low om he
use o AI-based p icing algo i hms, i is an impo an ins umen in add essing
hese conce ns as i p o ides bo h ex pos legal emedies and ex an e measu e
aimed a p e en ing un ai p ocessing o pe sonal da a (I ano a, 2020; Li, 2022). A
key objec i e o he GDPR is o p o ec undamen al igh s o da a subjec s, among
which a e he igh o p i acy and o non-disc imina ion.6 In acco dance wi h A i-
cle 5(1)(a) and (b), pe sonal da a shall be p ocessed ai ly, law ully, and anspa -
en ly. Al hough “ ai ness” is an elusi e concep , i a leas means ha da a p ocess-
ing should no c ea e de imen al e ec s and should no disc imina e o exploi
consume ulne abili ies (Cli o d & Ausloos, 2018; Tanna & Dunning, 2023). A i-
cles 9 and 10 GDPR p ohibi he p ocessing o “sensi i e” pe sonal da a, i.e. ade
union membe ship, gene ic and biome ic da a, heal h da a, and c iminal con ic-
ions, gi en ha he exemp ions o A icle 9(2) do no apply. This ies in closely
wi h anspa ency, as wi hou anspa ency i is highly di icul o iden i y b eaches
o (non-)compliance wi h he legal amewo k (Tanna & Dunning, 2023). A icle 22
GDPR, he igh no o be subjec ed o au oma ed decision-making, could be an
impo an ins umen o add ess decisions ha low om p icing algo i hms. How-
e e , he e is cu en ly disag eemen on he applica ion o A icle 22 GDPR o pe -
sonalised p icing, which weakens i s po en ial emedial ole (Zuide een Bo gesius
& Poo , 2019; Wong, 2020). Fu he mo e, a ini y-based pe sonalised p icing,
whe e consume s a e g ouped acco ding o hei in e ed p e e ences a he han
hei pe sonal da a, may ci cum en he legal amewo ks o da a p o ec ion law
4. P eamble AWGB.
5. We gelijke behandeling op g ond an handicap o ch onische ziek e [Equal T ea men (Disabili y
and Ch onic Illness) Ac ] (WGBH/CZ), 2003; We gelijke behandeling op g ond an lee ijd bij a beid
[Equal T ea men on he G ounds o Age a Wo k Ac ] (WGBL), 2004; We onde scheid a beidsduu
[Dis inc ion in Wo king Hou s Ac ] (WOA), 1996; and We onde scheid bepaalde en onbepaalde ijd
[Dis inc ion in De ini e Time Ac ] (WOBOT), 2002.
6. See A icle 7, 8, and 21 CFEU.
7 Heida y, an de Res , Cus e s
and an i-disc imina ion law (Wach e , 2020; Li, 2022).
2.3 Reconside ing p ice disc imina ion g ounds
Apa om g ounds ha a e legally p o ec ed, he e a e many segmen a ion bases
concei able (e.g. loyal y s a us, in elligence/educa ion le el, income) which a e no
p o ec ed in an i-disc imina ion o da a p o ec ion law (See, Bake , 2001; Tannock,
2008; Maxwell & Ga ba ino, 2010). G ounds ha companies ha e s a ed o use,
o example, include physical appea ance (He n, 2020), and ba e y le el (Na el-
ho , 2023). While hese g ounds a e pe missible, hey can lead o consume back-
lash; when consume s deem hem un ai and exploi a i e, he e can be a sudden
and s ong eac ion agains i in social media. Howe e , li le a en ion has been
paid o he (legal) implica ions o he ple ho a o (new) segmen a ion g ounds on
which pe sonalised p icing can be based online, beyond a call o b oaden he con-
sume backlash and co po a e social esponsibili y (CSR) li e a u e o include pe -
sonalised p icing (Van de Res e al., 2022), and a s udy ha explo ed why compa-
nies a e eluc an o use pe sonalised p icing online (Heida y e al., 2022). Gi en a
lack o ha monisa ion o p o ec ed g ounds ac oss na ional cons i u ions, he isk
o unin ended o indi ec disc imina ion, and consume un ai ness pe cep ions, a
econside a ion o he cu en Eu opean legal amewo k none heless seems wa -
an ed.
Fi s , he le el o p o ec ion p o ided agains disc imina ion a ies ac oss membe
s a es in e ms o he numbe o g ounds lis ed, bu also he wo ding o he enu-
me a ion – some Membe S a es p o ide exhaus i e enume a ions o g ounds,
while o he s p o ide a non-exhaus i e enume a ion, o e en no enume a ion a all
(Cus e s, 2023). Al hough he Du ch cons i u ion was ecen ly upda ed and wo
g ounds we e added o e lec de elopmen s in socie y, i akes much ime and e -
o o upda e na ional cons i u ions o e lec such changes. This aises ques ions
ega ding he obus ness o exis ing legal amewo ks in ela ion o de elopmen s
in he echnology su ounding pe sonalised p icing. The e has been qui e some
scep icism ega ding whe he na ional cons i u ions can (and should) keep up wi h
socie al de elopmen s (Ge a ds, 2016). This poin has also been aised o an i-dis-
c imina ion law in gene al (Tanna & Dunning, 2023). This aises he ques ion
whe he pe haps o he modes o egula ion a e mo e sui able o deal wi h he
challenges associa ed wi h pe sonalised p icing, such as he cons an ly changing
g ounds on which a pe sonalised p ice can be based.
Second, he e is a isk o unin ended o indi ec disc imina ion. In he case o di-
ec disc imina ion, a legally p ohibi ed g ound is decisi e o he unequal ea -
8 In e ne Policy Re iew 13(4) | 2024
GROUND UNFAIRNESS PERSONAL NORM SOCIAL NORM LEGALITY TRUST IN MARKET
LOYALTY STATUS 3.82 3.82 4.12 4.14 4.33
MARITAL STATUS 2.96 2.92 3.16 5.00 5.08
NATIONALITY 2.41 2.36 2.76 5.45 5.30
PHOTO/APPEARANCE 2.71 2.50 3.06 5.35 5.27
POLITICAL VIEWS 2.63 2.48 2.90 5.38 5.35
PURCHASE HISTORY 2.99 2.94 3.54 4.69 4.99
RACE/ETHNICITY 2.44 2.28 2.42 5.30 5.07
RELIGION 2.27 2.17 2.33 5.36 5.26
SEXUAL ORIENTATION 2.66 2.52 2.83 5.19 5.12
SOCIAL MEDIA DATA 2.88 2.83 3.48 4.83 4.82
SOCIOECONOMIC STATUS 3.27 3.23 3.74 4.44 4.70
STUDENT STATUS 3.49 3.55 3.80 4.09 4.16
Fi s , ai ness pe cep ions o he disc imina ion g ounds e ealed ha pa icipan s
anked all g ounds as un ai (i.e. all unde 4 on he 7-poin scale). Ou o all
g ounds, he use o loyal y s a us o pe sonalised p icing was deemed he leas
un ai (M = 3.82, SD = 1.70), wi h loca ion (M = 3.57, SD = 1.69) and s uden s a us
(M = 3.49, SD = 1.77) in second and hi d place, espec i ely. Religion was deemed
he mos un ai g ound (M = 2.27, SD = 1.50), wi h na ionali y (M = 2.41, SD = 1.60)
and gende (M = 2.41, SD = 1.60) in sha ed second place and e hnici y (M = 2.44,
SD = 1.70) in hi d place. Fo he comple e anking, see Figu e 2.
15 Heida y, an de Res , Cus e s
FIGURE 2: Pe cei ed ai ness o disc imina ion g ounds (N = 727).
Second, when asked abou he alignmen o he disc imina ion g ounds wi h pe -
sonal no ms, a esponse simila o pe cei ed ai ness was obse ed. Tha is, none
o he g ounds aligned wi h pe sonal no ms (i.e. all unde 4 on he 7-poin scale).
The g ounds ha aligned he leas wi h pa icipan s’ pe sonal no ms we e eligion
(M = 2.17, SD = 1.55), gende (M = 2.25, SD = 1.59), and e hnici y (M = 2.28, SD =
1.57). Loyal y s a us (M = 3.82, SD = 1.74), s uden s a us, and loca ion epo edly
misaligned he leas wi h pe sonal no ms. Fo he comple e anking, see Figu e 3.
16 In e ne Policy Re iew 13(4) | 2024
FIGURE 3: Pe cei ed alignmen wi h pe sonal no ms o disc imina ion g ounds (N = 727).
Thi d, he g ounds we e also anked in e ms o he (pe cei ed) alignmen wi h ex-
is ing social no ms. He e we howe e obse ed a small posi i e di e ence wi h
pe sonal no ms, (579) = -7.47, p < .001, D = .31. The sco es o social no ms we e
consis en ly highe o each g ound han o pe sonal no ms. The g ounds we e
pe cei ed o align less wi h pe sonal no ms han wi h social no ms. In o he
wo ds, o all g ounds pa icipan s assumed highe accep ance in socie y han hei
own accep ance.
The g ounds ha we e pe cei ed o align he leas wi h socie al no ms, we e eli-
gion (M = 2.33, SD = 1.45), e hnici y (M = 2.42, SD = 1.53), and gende (M = 2.73, SD
= 1.78). Loyal y s a us (M = 4.12, SD = 1.74), loca ion (M = 4.03, SD = 1.60), and s u-
den s a us (M = 3.8, SD = 1.82) we e pe cei ed o align mo e wi h socie al no ms
bu s ill sco ed unde 4 on he 7-poin scale; see Figu e 4 o he comple e ank-
ing.
17 Heida y, an de Res , Cus e s
FIGURE 4: Pe cei ed alignmen wi h social no ms o disc imina ion g ounds (N = 727).
Fou h, we asked pa icipan s o wha ex en each g ound should be legally p o-
hibi ed. The h ee g ounds ha we e deemed he mos illegi ima e, we e na ional-
i y (M = 5.45, SD = 1.60), heal h da a (M = 5.4, SD = 1.59), and in elligence (M =
5.39, SD = 1.72). S uden s a us (M = 4.09, SD = 1.89), loyal y s a us (M = 4.14, SD =
1.83), and loca ion da a (M = 4.31, SD = 1.70) we e deemed he leas illegi ima e,
albei hey s ill we e anked o e 4 on he 7-poin scale. Figu e 5 shows he com-
ple e anking.
18 In e ne Policy Re iew 13(4) | 2024
FIGURE 5: Pe cei ed illegi imacy o disc imina ion g ounds (N = 727).
Fi h and las , pa icipan s had o indica e o wha ex en he use o a ce ain
g ound would diminish hei us in he ma ke . The use o poli ical iews (M =
5.35, SD = 1.54), na ionali y (M = 5.3, SD = 1.69), and in elligence (M = 5.29, SD =
1.75) we e he op h ee g ounds ha would lead o a loss o us . S uden s a us
(M = 4.16, SD = 1.81), loyal y s a us (M = 4.33, SD = 1.73), and loca ion (M = 4.35,
SD = 1.58) came ou as he h ee g ounds ha would lead o less loss in us , al-
bei s ill all g ounds sco ed o e 4 on he 7-poin scale, hence all leading o less
us in ma ke s. Fo he comple e anking, see Figu e 6.
19 Heida y, an de Res , Cus e s
FIGURE 6: Repo ed loss o us in he ma ke in case o use o disc imina ion g ounds (N = 727).
We compa ed he legally p ohibi ed g ounds o he g ounds ha a e no (ye )
legally p ohibi ed. We c ea ed wo ca ego ies, bundling all pe cep ions o each o
he i e dimensions and conduc ed a Pai ed Samples es o compa e he a e age
sco es o he wo ca ego ies. Fo ai ness, pe sonal no ms, and social no ms, he
mean alue o all legal g ounds was signi ican ly highe han ha o he “non-le-
gal” g ounds. Legal g ounds we e pe cei ed as mo e un ai han non-legal g ounds
(M = -.43, SD = 1.09), less in alignmen wi h pe sonal no ms (M = -.45, SD = 1.10)
and social no ms (M = -.60, SD = 1.16). Addi ionally, o e all, legal g ounds we e
deemed mo e illegi ima e (i.e. less pe missible) han “non-legal g ounds” (M = .46,
SD = 1.21) and we e epo ed o lead o a highe loss in us (M = .32, SD = 1.16).
In addi ion o ca ego ising g ounds based on hei cu en legali y, we also dis in-
guished be ween immu able and ( echnically) mu able g ounds. We selec ed h ee
unambiguous g ounds o each ca ego y. We conside ed e hnici y, sexual o ien a-
ion, and in elligence o be immu able, and b owse ype, ba e y pe cen age, and
de ice ype o be mu able. As o he immu able g ounds, e hnici y and sexual o i-
en a ion a e legally p o ec ed disc imina ion g ounds, whe eas in elligence is no .
We ound signi ican di e ences (p < .001) be ween he g oups: immu able
g ounds we e conside ed mo e un ai han mu able g ounds (M = 2.53 s. M =
2.82), less aligned wi h pe sonal no ms (M = 2.40 s. M = 2.72), and less aligned
wi h social no ms (M = 2.68 s. M = 3.10). Mo eo e , immu able g ounds we e
deemed o be mo e illegi ima e han mu able g ounds (M = 5.30 s. M = 4.93) and
20 In e ne Policy Re iew 13(4) | 2024
would diminish pa icipan s’ us in he ma ke mo e (M = 5.16 s. M = 4.93).
5. Discussion
The aim o his esea ch was o p o ide a comp ehensi e o e iew o consume
pe cep ions ega ding g ounds used o pe sonalised p icing, and o examine
whe he he e is a gap be ween pe cep ions o g ounds ha a e p ohibi ed and no
p ohibi ed. Whe e p e ious esea ch ocused on gene al (un) ai ness pe cep ions
su ounding pe sonalised p icing (Tu ow e al., 2009; Poo & Zuide een Bo ge-
sius, 2019) o zoomed in on speci ic non-legally p ohibi ed g ounds (P ies e e al.,
2020; Hu nagel e al., 2022), ou esea ch se ou o map an o e iew o pe cep-
ions o a ious g ounds ac oss se e al dimensions. By combining bo h ypes o
g ounds (legally and no legally p ohibi ed), i is possible o assess he ex en o
which cu en legisla ion aligns wi h socie al pe cep ions and whe he g ounds
ha a e (no ) ye legally p ohibi ed, e oke simila esponses as legally p ohibi ed
g ounds. Ou indings p o ide aluable inpu o he discussion on he ex en o
which cu en legisla ion o e s – o should o e – app op ia e sa egua ds agains
he challenges associa ed wi h pe sonalised p icing.
We ind qui e some o e lap be ween consume pe cep ions o un ai ness (“illegi i-
macy”) and illegali y. The g ounds ha pa icipan s in ou su ey conside mos
un ai a e also he g ounds ha mos need p o ec ion by law, a inding la gely con-
sis en wi h wha is al eady es ablished in an i-disc imina ion and da a p o ec ion
law. Mo eo e , om he g ounds ha a e no p ohibi ed, in elligence and physical
appea ance sco ed high on un ai ness and we e epo ed o align he leas wi h
pe sonal and social no ms. Pe sonalising p ices on b owse ype and ba e y pe -
cen age also sco ed ela i ely high in e ms o un ai ness and misalignmen wi h
pe sonal and social no ms.
Pe cep ions o (social) no m misalignmen could ha e a - eaching nega i e conse-
quences o companies and he digi al ma ke . Fo he la e , we included he di-
mension “ us in ma ke ” o assess he ex en o which he use o a g ound would
lead o (sel - epo ed) loss o us among pa icipan s. We ound ha he use o
g ounds ha we e deemed he leas aligned wi h social no ms and he mos ille-
gi ima e we e accompanied wi h a highe sel - epo ed loss o us in he digi al
ma ke . “Newe ” g ounds ha sco ed high on no m misalignmen , such as in elli-
gence and appea ance, showed a simila co ela ion.
21 Heida y, an de Res , Cus e s
5.1 Policy implica ions
F om a policy pe spec i e, i is in e es ing o zoom in on he g ounds ha a e no
ye legally p ohibi ed, bu ha we e pe cei ed as un ai and in iola ion o pe son-
al and social no ms: in elligence, appea ance, ba e y pe cen age, and b owse
ype. One could a gue ha hese g ounds could po en ially be included in legisla-
ion as p ohibi ed g ounds – especially in elligence and appea ance.
The e a e a leas wo a gumen s o his. One is he iewpoin ha he law is a
codi ica ion o social no ms (Basu, 2002). The law is no a s a ic sys em, bu
changes o e ime, e lec ing changes in no ms and pe cep ions in socie y. I social
no ms and pe cep ions change, in his case because new echnologies enable new
g ounds o (p ice) disc imina ion, his can be su icien eason o change he legis-
la ion acco dingly. Legally p ohibi ing disc imina ion g ounds ha a e conside ed
un ai by people would mean a u he alignmen be ween he legal sys em and
social no ms and pe cep ions. Basically, his is a ai ness a gumen , which applies
he s onges o inhe en ly immu able pe sonal cha ac e is ics.
A second a gumen o conside changing he law has a mo e economic pe spec-
i e: a lack o p o ec ion o g ounds o p ice disc imina ion ha a e conside ed
un ai would also epo edly lead o a ela i ely high loss o us in he ma ke
(C oss, 2005). Hence, apa om he un ai ness a an indi idual le el, he e is a
la ge economic e ec ha may p o ide an a gumen o he legisla o o s ep in
and o e p o ec ion h ough egula ion.
When econside ing he g ounds ha need p o ec ion, he ques ion is which
g ounds o include. The cu en lack o ha monisa ion o g ounds in EU Membe
S a es show ha he e is no a sha ed unde s anding o ai ness, o a leas no
ag eemen on which g ounds equi e p o ec ion (Cus e s, 2023). Al hough he e-
po ed loss o us and un ai ness pe cep ions associa ed wi h newe g ounds
could o m a jus i ica ion o p o ec hese g ounds – o a leas eassess he cu -
en legal amewo k – he apid pace in which he echnology is de eloping
makes i di icul o p edic o wha ex en legal egula ion could keep up wi h
hese de elopmen s. New g ounds could eme ge ha a e also pe cei ed as un ai ,
bu companies could also ind ways o ci cum en p ohibi ed g ounds by using
p oxies o engaging in indi ec disc imina ion, bo h o which a e p o en o be di i-
cul o de ec and en o ce (Zuide een Bo gesius, 2018). The e o e, adding se e al
new p ohibi ed disc imina ion g ounds o he legal amewo k does no seem o
be he way o wa d acco ding o some au ho s (Solo e, 2024).
22 In e ne Policy Re iew 13(4) | 2024
Ou o all he non-legally p ohibi ed g ounds, p icing based on in elligence and
appea ance ecei ed he mos nega i e eac ions. In elligence and appea ance a e
bo h ( echnically) immu able, meaning ha hey a e unchangeable, o a leas no
changeable wi hou signi ican e o . Disc imina ion based on in elligence o on
appea ance is no legally p ohibi ed10 and happens on a daily basis (Tannock,
2008; Liu, 2017). Pho os a e no only conside ed biome ic da a,11 bu eadily e-
eal all kinds o physical a ibu es, such as ace, eligion, heal h, and e hnici y.
While in elligence is o a la ge pa inna e, en i onmen al ac o s such as a high
socioeconomic s a us can con ibu e o de eloping in ellec ual skills, and he o he
way a ound: mo e in elligen indi iduals end o achie e highe le el o educa ion,
occupa ional s a us, income, and e en be e heal h ou comes (Decke s e al.,
2017; Bosma e al., 2007). Howe e , since da a a e so hea ily in e wined, i would
be di icul o subs an ia e why exac ly hese wo g ounds would need o be added
o he exis ing legal amewo k (Solo e, 2024).
Al e na i ely, ins ead o p o iding jus i ica ion o adding new p ohibi ed g ounds,
ou indings p o ide insigh in o he deg ee o ha m ha migh be in lic ed when
speci ic g ounds a e used o p ice pe sonalisa ion. In ou indings, socioeconomic
s a us comes o wa d as a ela i ely accep able – o less unaccep able – g ound o
base p ices on. Socioeconomic s a us and i s supposed p o ec ion a e a poin o
discussion among EU Membe S a es (Gan y & Beni o Sanchez, 2021). Socioeco-
nomic s a us is no (ye ) legally p o ec ed in Du ch cons i u ional law, pa ly be-
cause he e is s ill no consensus ega ding whe he i is an inhe en o immu able
cha ac e is ic (Tweede Kame , 2020).12 In he con ex o pe sonalised p icing, he
e ec o his p icing s a egy on consume wel a e is ambiguous (Elegido, 2011). In
some cases, i could be bene icial o dis inguish be ween consume s and hei so-
cioeconomic posi ion, as i migh allow ce ain consume s access o a p oduc o
se ice ha hey – because o hei socioeconomic posi ion – would no ha e
been able o a o d o he wise. Howe e , he ulne abili ies ha come wi h being
pa o a ce ain socioeconomic class (e.g. low le el o digi al li e acy), could e y
well be exploi ed by companies (S ycha z & Dui en oo de, 2021). Fu he mo e,
socioeconomic s a us and social class a e hea ily in e wined wi h sensi i e da a
such as poli ical p e e ences and e hnici y (Gandy, 2009). In line wi h Solo e
(2024), we p opose ha a case-by-case analysis is needed, ocusing on how ce ain
da a a e used in pe sonalised p icing, a he han he ype o da a. I such analyses
10. Apa om Belgium, F ance and Se bia, whe e disc imina ion on he basis o physical appea ance is
p ohibi ed.
11. Reci al 51 GDPR.
12. The g ound is cu en ly p o ec ed in 11 Membe S a es.
23 Heida y, an de Res , Cus e s
would e eal he need o include new disc imina ion g ounds in legisla ion, his
does no necessa ily need o be done in o mal legisla ion, such as in e na ional
ea ies, na ional cons i u ions, o equal ea men ac s. The open-endedness o
he lis ings o disc imina ion g ounds (i.e. non-exhaus i e lis ings ha use ph ases
like “o o he cha ac e is ics”) in many legal ins umen s allows judges and cou s
o quali y new g ounds as illegal in pa icula con ex s.
Fu he mo e, company beha iou ha exploi s consume cha ac e is ics can possi-
bly be deemed as un ai unde compe i ion and consume law (Sau e , 2020; Li e
al., 2023; Dui en oo de, 2023). Ne e heless, online p ice disc imina ion emains
di icul o de ec due o he di icul ies in isola ing he g ounds ha he p ice di -
e ence was based on. Conside ing how he cu en Eu opean legal amewo k
could be le e aged o add ess changes in he g ounds ha could be used o on-
line p ice disc imina ion is he e o e only one pa o sol ing he puzzle. Inc eased
anspa ency abou he unde lying mechanisms o a pe sonalised p ice will u he
ou knowledge abou he cu en s a e o he a and he (pe sonal) in o ma ion
ha can be used o his end. The indi idual access and in o ma ion igh s in he
GDPR could aid in o e coming consume s’ e iden ia y de iciencies (A icle 13-15
GDPR; Hacke , 2018). Clea ing up he cu en con usion su ounding he applicabil-
i y o A icle 22 GDPR would be ano he s ep owa ds p o iding a mo e obus p o-
ec ion agains online p ice disc imina ion (Wong, 2020). Fu he mo e, da a p o ec-
ion law can p o ide o ex an e measu es o implemen mo e anspa ency abou
he g ounds used in da a p ocessing (Li, 2022).
5.2 Theo e ical implica ions
Ou indings a e consis en wi h p e ious indings and obse a ions ha pe son-
alised p icing is gene ally iewed by he public as un ai and illegi ima e (Poo &
Zuide een Bo gesius, 2019; Tu ow e al., 2005, P ies e e al., 2020). The di e -
ences ha we ind be ween g ounds a e also in line wi h p e ious s udies ha o-
cused on speci ic non-legally p ohibi ed g ounds. Fo ins ance, in line wi h Hu -
nagel e al. (2022), we ind ha pe sonalised p icing based on loca ion is deemed
as less un ai han de ice ype. Fu he mo e, in line wi h P ies e e al., (2021), we
also ind ha loca ion is deemed as less un ai han pu chase his o y. The g ounds
ha we in es iga ed all ank unde he second hal o he se en-poin scale o ai -
ness, pe sonal no ms, and social no ms inasmuch ha no g ound is conside ed ai
o pe sonalised p icing – only less un ai . A simila end is obse ed o he pe -
cei ed illegi imacy and he e ec ha he use o he disc imina ion g ounds has
o consume s’ us in ma ke s. In sum, his con i ms ha he cu en pe cep ion o
online p ice disc imina ion among consume s is s ill nega i ely loaded. Con-
24 In e ne Policy Re iew 13(4) | 2024
Hu nagel, G., Schwaige , M., & We i z, L. (2022). Seeking he pe ec p ice: Consume esponses o
pe sonalized p ice disc imina ion in e-comme ce. Jou nal o Business Resea ch, 143, 346–365. h p
s://doi.o g/10.1016/j.jbus es.2021.10.002
I ano a, Y. (2020). The da a p o ec ion impac assessmen as a ool o en o ce non-disc imina o y
AI. In L. An unes, M. Naldi, G. F. I aliano, K. Rannenbe g, & P. D ogka is (Eds.), P i acy Technologies
and Policy (Vol. 12121, pp. 3–24). Sp inge . h ps://doi.o g/10.1007/978-3-030-55196-4_1
Khai an, T. (2015). A heo y o disc imina ion law. Ox o d Uni e si y P ess. h ps://doi.o g/10.1093/ac
p o :oso/9780199656967.001.0001
La son, J., Ma u, S., & Angwin, J. (2015). Unin ended consequences o geog aphic a ge ing [P ojec
epo ]. P oPublica. h ps://s a ic.p opublica.o g/p ojec s/p ince on-
Li, Z. (2022). A ini y-based algo i hmic p icing: A dilemma o EU da a p o ec ion law. Compu e
Law & Secu i y Re iew, 46. h ps://doi.o g/10.1016/j.cls .2022.105705
Liu, X. (2017). Disc imina ion and lookism. In K. Lippe -Rasmussen (Ed.), The Rou ledge handbook o
he e hics o disc imina ion (pp. 276–288). Rou ledge. h ps://doi.o g/10.4324/9781315681634
Malgie i, G. (2023). In/accep able ma ke ing and consume s’ p i acy expec a ions: Fou es s om
EU da a p o ec ion law. Jou nal o Consume Ma ke ing, 40(2), 209–223. h ps://doi.o g/10.1108/JC
M-03-2021-4571
Maxwell, S. (2002). Rule-based p ice ai ness and i s e ec on willingness o pu chase. Jou nal o
Economic Psychology, 23(2), 191–212. h ps://doi.o g/10.1016/S0167-4870(02)00063-6
Maxwell, S., & Ga ba ino, E. (2010). The iden i ica ion o social no ms o p ice disc imina ion on he
in e ne . Jou nal o P oduc & B and Managemen , 19(3), 218–224. h ps://doi.o g/10.1108/1061042
1011046193
Mikians, J., Gya ma i, L., E amilli, V., & Laou a is, N. (2012). De ec ing p ice and sea ch
disc imina ion on he in e ne . P oceedings o he 11 h ACM Wo kshop on Ho Topics in Ne wo ks,
79–84. h ps://doi.o g/10.1145/2390231.2390245
Mille , A. A. (2014). Wha do we wo y abou when we wo y abou p ice disc imina ion? The law
and e hics o using pe sonal in o ma ion o p icing. Jou nal o Technology Law & Policy, 19(1),
41–104. h ps://schola ship.law.u l.edu/j lp/ ol19/iss1/2
Mui , E. (2019). The ho izon al e ec s o cha e igh s gi en exp ession o in EU legisla ion, om
Mangold o Baue . Re iew o Eu opean Adminis a i e Law, 12(2), 185–215. h ps://doi.o g/10.7590/1
87479819X15840066091312
Na elho , Y. (2023, Ap il 10). Di é en s a i s pou une même cou se: Ube ne me pas ou le
monde su un même pied d’égali é [Di e en a es o he same jou ney: Ube does no pu
e e yone on he same oo ing]. De niè e Heu e. h ps://www.dhne .be/conso/au o-mo o/2023/04/1
0/di e en s- a i s-pou -une-meme-cou se-ube -ne-me -pas- ou -le-monde-su -un-meme-pied-deg
ali e-CE3CGQL3EVH6HAJN33NRKM36Q4/
Odlyzko, A. (2003). P i acy, economics, and p ice disc imina ion on he In e ne . P oceedings o he
5 h In e na ional Con e ence on Elec onic Comme ce, 355–366. h ps://doi.o g/10.1145/948005.9480
51
Odlyzko, A. (2004). The e olu ion o p ice disc imina ion in anspo a ion and i s implica ions o
he in e ne . Re iew o Ne wo k Economics, 3(3). h ps://doi.o g/10.2202/1446-9022.1055
31 Heida y, an de Res , Cus e s
O ganisa ion o Economic Co-ope a ion and De elopmen . (2015). Da a-d i en inno a ion: Big da a
o g ow h and well-being. OECD Publishing. h ps://doi.o g/10.1787/9789264229358-en
O ganisa ion o Economic Co-ope a ion and De elopmen . (2016). P ice Disc imina ion –
Backg ound no e by he Sec e a ia (Repo No. DAF/COMP(2016)15). h ps://one.oecd.o g/documen /
DAF/COMP(2016)15/en/pd
O ganisa ion o Economic Co-ope a ion and De elopmen . (2018b). Pe sonalised p icing in he
digi al e a – No e by he Ne he lands (Repo No. DAF/COMP/WD(2018)124). h ps://one.oecd.o g/do
cumen /DAF/COMP/WD(2018)124/en/pd
O ganisa ion o Economic Co-ope a ion and De elopmen . (2018a). Pe sonalised p icing in he
digi al e a – Backg ound no e by he Sec e a ia (Repo No. DAF/COMP(2018)13). h ps://one.oecd.o
g/documen /DAF/COMP(2018)13/en/pd
Poo , J., & Zuide een Bo gesius, F. J. (2019). Does e e yone ha e a p ice? Unde s anding people’s
a i ude owa ds online and o line p ice disc imina ion. In e ne Policy Re iew, 8(1). h ps://doi.o g/
10.14763/2019.1.1383
P es on McA ee, R. (2008). P ice disc imina ion. In W. D. Collins & J. Angland (Eds.), Issues in
compe i ion law and policy 465 (pp. 465–484). ABA Sec ion o An i us Law.
P ies e , A., Robbe , T., & Ro h, S. (2020). A special p ice jus o you: E ec s o pe sonalized
dynamic p icing on consume ai ness pe cep ions. Jou nal o Re enue and P icing Managemen ,
19(2), 99–112. h ps://doi.o g/10.1057/s41272-019-00224-3
Rosenc ance, L. (2000, Sep embe 13). Cus ome ou age p omp s Amazon o change p ice- es ing
policy. Compu e wo ld. h ps://www.compu e wo ld.com/a icle/2597088/cus ome -ou age-p omp
s-amazon- o-change-p ice- es ing-policy.h ml
Sau e , W. (2020). A du y o ca e o p e en online exploi a ion o consume s? Digi al dominance
and special esponsibili y in EU compe i ion law. Jou nal o An i us En o cemen , 8(2), 406–427. h
ps://doi.o g/10.1093/jaen o/jnz023
Sea s, A. M. (2021). The limi s o online p ice disc imina ion in Eu ope. Science and Technology Law
Re iew, 21(1), 1-42 Pages. h ps://doi.o g/10.7916/STLR.V21I1.5761
Shille , B. (2014). Fi s deg ee p ice disc imina ion using big da a (Wo king Pape No. 58). h ps://Econ
Pape s. epec.o g/RePEc:b d:wpape :58
Solo e, D. J. (2024). Da a is wha da a does: Regula ing use, ha m, and isk ins ead o sensi i e da a.
No hwes e n Uni e si y Law Re iew, 118(4), 1081–1138. h ps://doi.o g/10.2139/ss n.4322198
S eppe, R. (2017). Online p ice disc imina ion and pe sonal da a: A Gene al Da a P o ec ion
Regula ion pe spec i e. Compu e Law & Secu i y Re iew, 33(6), 768–785. h ps://doi.o g/10.1016/j.c
ls .2017.05.008
S ole, L. A. (2007). P ice disc imina ion and compe i ion. In M. A ms ong & R. Po e (Eds.),
Handbook o indus ial o ganiza ion (Vol. 3, pp. 2221–2299). Else ie . h ps://doi.o g/10.1016/S157
3-448X(06)03034-2
S ycha z, J., & Dui en oo de, B. (2021). The exploi a ion o ulne abili y h ough pe sonalised
ma ke ing communica ion: A e consume s p o ec ed? In e ne Policy Re iew, 10(4), 1–27. h ps://do
i.o g/10.14763/2021.4.1585
Tanna, M., & Dunning, W. (2023). Bias and disc imina ion in he use o AI in he inancial sec o . In
32 In e ne Policy Re iew 13(4) | 2024
N. Remolina & A. Gu ea-Ma inez (Eds.), A i icial in elligence in inance (pp. 320–349). Edwa d
Elga Publishing. h ps://doi.o g/10.4337/9781803926179.00025
Tannock, S. (2008). The p oblem o educa ion-based disc imina ion. B i ish Jou nal o Sociology o
Educa ion, 29(5), 439–449. h ps://www.js o .o g/s able/40375368
Tu ow, J., Feldman, L., & Mel ze , K. (2005). Open o exploi a ion: Ame ica’s shoppe s online and
o line. (Depa men al Pape s (ASC)) [Wo king pape ]. h ps:// eposi o y.upenn.edu/handle/20.500.1
4332/2000
Tweede Kame de S a en-Gene aal [ House o Rep esen a i es o he Ne he lands]. (2020). He
Voo s el an we an de leden Be gkamp, Özü ok en Van den Hul houdende e kla ing da e g ond
bes aa een oo s el in o e weging e nemen o e ande ing in de G ondwe , s ekkende o oe oeging
an handicap en seksuele ge ich heid als non-disc imina ieg ond (32411) [P oposal o a law by he
membe s Be gkamp, Özü ok and Van den Hul decla ing ha he e a e g ounds o conside a p oposal o
amend he Cons i u ion o add disabili y and sexual o ien a ion as g ounds o non-disc imina ion
(32411)] (No. P oposal). Go e nmen o he Ne he lands. h ps://www.ee s ekame .nl/behandeling/2
0200623/ oo ze ing_behandeling/docume
Van Bekkum, M., & Zuide een Bo gesius, F. (2023). Using sensi i e da a o p e en disc imina ion
by a i icial in elligence. Does he GDPR need a new excep ion? Compu e Law & Secu i y Re iew, 48,
1–12. h ps://doi.o g/10.1016/j.cls .2022.105770
Van de Res , J.-P. I., Sea s, A. M., Miao, L., & Wang, L. (2020). A no e on he u u e o pe sonalized
p icing: Cause o conce n. Jou nal o Re enue and P icing Managemen , 19(2), 113–118. h ps://doi.o
g/10.1057/s41272-020-00234-6
Van de Res , J.-P., Sea s, A. M., Kuokkanen, H., & Heida y, K. (2022). Algo i hmic p icing in
hospi ali y and ou ism: Call o esea ch on e hics, consume backlash and CSR. Jou nal o
Hospi ali y and Tou ism Insigh s, 5(4), 771–781. h ps://doi.o g/10.1108/JHTI-08-2021-0216
Wach e , S. (2020). A ini y p o iling and disc imina ion by associa ion in online beha iou al
ad e ising. Be keley Technology Law Jou nal, 35(2), 367–430. h ps://doi.o g/10.15779/Z38JS9H82M
Wach e , S. (2022). The heo y o a i icial immu abili y: P o ec ing algo i hmic g oups unde an i-
disc imina ion law (Ve sion 1). a Xi . h ps://doi.o g/10.48550/ARXIV.2205.01166
Wong, B. (2020). Online pe sonalised p icing as p ohibi ed au oma ed decision-making unde
A icle 22 GDPR: A scep ical iew. In o ma ion & Communica ions Technology Law, 30(2), 193–207. h
ps://doi.o g/10.1080/13600834.2020.1860460
Zuide een Bo gesius, F. (2018). Disc imina ion, a i iical in elligence and algo i hmic decision-making
[S udy]. Council o Eu ope. h ps:// m.coe.in /disc imina ion-a i icial-in elligence-and-algo i hmic-
decision-making/1680925d73
Zuide een Bo gesius, F., & Poo , J. (2017). Online p ice disc imina ion and EU da a p i acy law.
Jou nal o Consume Policy, 40(3), 347–366. h ps://doi.o g/10.1007/s10603-017-9354-z
33 Heida y, an de Res , Cus e s
Appendices
Appendix 1: Explana ion o disc imina ion g ounds
NOTE: The unde lined g ounds a e legally p ohibi ed ei he in he Eu opean legal amewo k o in
he Du ch cons i u ion. The g ounds in i alics a e legally p ohibi ed in some o he na ional
cons i u ions in he EU.
GROUND ORIGIN REFERENCE/FIELD OF LAW ARTICLE EXAMPLES USED IN
SURVEY
AGE Law An i-disc imina ion law 21 CFEU; A icle 1
G ondwe ;
“Young”/”old", age ca ego y,
e c.
BATTERY
PERCENTAGE Li e a u e OECD 2018a; Na elho ,
2023
The ba e y pe cen age o
he de ice on which he
p ice is shown.
BROWSER TYPE Li e a u e Mikians e al., 2012; OECD
2018a
Google Ch ome o Sa a i,
cookies, incogni o mode,
e c.
BROWSING
ACTIVITY Li e a u e OECD 2018a
B owsing his o y, o igin
page, ime spen on
webpages, e c.
CRIMINAL
RECORD/
OFFENCES
Law Da a p o ec ion law A icle 9 GDPR C iminal eco d, c iminal
o ences, allega ions, e c.
DEVICE TYPE Li e a u e OECD 2018; Hu nagel e al.,
2022
Mobile, PC o able , Apple
o Windows, e c.
GENDER/SEX Law An i-disc imina ion law 21 CFEU; A icle 1
G ondwe ; A icle 1 AWGB
Female, male, ansgende ,
non-bina y, e c.
HEALTH DATA Law Da a p o ec ion law A icle 9 GDPR
Medical eco d, in o ma ion
abou smoking, blood
p essu e, e c.
IDEOLOGY/
PHILOSOPHY Law An i-disc imina ion law 21 CFEU; A icle 1
G ondwe ; A icle 1 AWGB
Pe sonal con ic ions abou
socie y, humani y, e c.
IMPAIRMENT/
DISABILITY Law An i-disc imina ion law 21 CFEU; A icle 1
G ondwe
Men al o physical disabili y,
ch onic illness, e c.
INCOME/WEALTH Law Solo e, 2024 ‘Rich’/‘poo ’, high o low
income, e c.
INTELLIGENCE Li e a u e Tannock, 2008 IQ-sco e, highes le el o
educa ion ob ained, e c.
LOCATION Li e a u e
La son, Ma u & Angwin,
2015; OECD, 2018a; P ies e
e al., 2020
The coun y o place in
which one esides o is
loca ed.
LOYALTY
STATUS Li e a u e Maxwell & Ga ba ino, 2010;
OECD, 2018a; CMA, 2020
Loyal cus ome , new
cus ome , bulk cus ome s,
e c.
MARITAL
STATUS Law An i-disc imina ion law A icle 1 AWGB Ma ied, ci il pa ne ship,
unma ied, e c.
34 In e ne Policy Re iew 13(4) | 2024
GROUND ORIGIN REFERENCE/FIELD OF LAW ARTICLE EXAMPLES USED IN
SURVEY
NATIONALITY Law An i-disc imina ion law 21 CFEU; A icle 1 AWGB Place o bi h, ci izenship,
e c.
PHOTO/
APPEARANCE Li e a u e He n, 2020; Solo e, 2024
Con o mi y wi h beau y
ideals, physical appea ance,
e c.
POLITICAL
VIEWS Law An i-disc imina ion law;
Da a p o ec ion law
21 CFEU; A icle 1
G ondwe ; A icle 1
AWGB; A icle 9 GDPR
“Le ”/“ igh ”, libe al/
conse a i e, e c.
PURCHASE
HISTORY Li e a u e OECD 2018a;
Ea lie pu chases, ype o
pu chases, amoun spen ,
e c.
RACE/
ETHNICITY Law An i-disc imina ion law;
Da a p o ec ion law
21 CFEU; A icle 1
G ondwe ; A icle 9 GDPR Skin colou , ances y, e c.
RELIGION Law An i-disc imina ion law;
Da a p o ec ion law
21 CFEU; A icle 1
G ondwe ; A icle 1
AWGB; A icle 9 GDPR
Ch is ian, Muslim, Jew,
A heis , e c.
SEXUAL
ORIENTATION Law An i-disc imina ion law;
Da a p o ec ion law
21 CFEU; A icle 1
G ondwe ; A icle 1
AWGB; A icle 9 GDPR
He e osexual, homosexual,
bisexual, e c.
SOCIAL MEDIA
DATA Li e a u e OECD 2018a; Likes, commen s,
in e ac ions, e c.
SOCIOECONOMIC
STATUS Law Solo e, 2024; Gan y &
Beni o Sanchez, 2021
Posi ion on social scale, i.e.
h ough educa ion o job
posi ion.
STUDENT
STATUS Li e a u e Ca oll & Coa es, 1999 En ollmen in (highe )
educa ional ins i u ion.
Appendix 2: Su ey ins umen
In oduc ion
Thank you o you ime and pa icipa ion in his su ey. You answe s con ibu e o esea ch on pe sonalising online
p ices. You pa icipa ion is highly app ecia ed. Comple ing he ques ionnai e akes app oxima ely 5-6 minu es. Please
no e ha i is easies o ill in his su ey on a compu e o lap op.
We kindly eques ha you answe all ques ions u h ully. You da a and answe s will be ea ed con iden ially. Only he
esea che s ha e access o you da a. Pa icipa ion is comple ely olun a y, and you can s op a any ime du ing he
su ey. I is also possible o eques he dele ion o he da a p o ided a e wa ds by con ac ing he esea che s.
Fo ques ions o commen s, please con ac Kimia Heida y a k.heida [email protected] .nl.
By clicking he 'I ag ee' bu on, you indica e ha you ha e ead he abo e in o ma ion, ha you a e awa e ha
pa icipa ion is olun a y and ha you ag ee o you da a being used o esea ch pu poses. I you do no wan o
pa icipa e, you can s op now by closing his page. Thank you again o you coope a ion.
▢ I ag ee o he e ms and condi ions and would like o pa icipa e in he ques ionnai e.
Sec ion 1 – Demog aphic in o ma ion
1. Wha is you age?
2. Wha gende do you iden i y wi h mos ?
▢ Male
▢ Female
▢ Non-bina y/O he
▢ P e e no o say
3. In which coun y do you eside?
[d opdown menu wi h 206 coun ies]
4. Wha is he highes le el o o mal educa ion you ha e a ained?
▢ P ima y school
▢ High school
35 Heida y, an de Res , Cus e s
▢ Highe p o essional educa ion (HBO)
▢ Bachelo ’s Deg ee
▢ Mas e ’s Deg ee
▢ Doc o a e
▢ O he , namely: ….
5. Wha is you g oss annual income?
▢ Less han €20,000
▢ €20,000 – €49,999
▢ €50,000 – €74,999
▢ €75,000 – €99,999
▢ Mo e han €100,000
▢ P e e no o say
Sec ion 2 – Scena io
Please ead he ollowing ex ca e ully be o e clicking 'nex ':
'Companies a e inc easingly expe imen ing wi h hei p ices online. This also applies o o e ing pe sonalised p ices:
cha ging di e en p ices o di e en consume s based on hei da a, such as pe sonal cha ac e is ics o online beha io .
This may esul in a highe o lowe p ice o ce ain g oups o consume s.'
On he nex page we p esen you wi h a numbe o g ounds on he basis o which companies could pe sonalise he p ice.
We a e cu ious abou you opinion on his.
Click 'nex ' o con inue o he ques ions. Please no e ha once you click on his, you will no be able o e u n o he
p e ious page
[Pa icipan s we e andomly assigned o one ou o i e g oups and being shown 5 (ou o 25) g ounds]
Sec ion 3 – Ques ions ela ed o g ounds
[All answe s on a scale o 1 (‘s ongly disag ee’) o 7 (‘s ongly ag ee’): 1 = s ongly disag ee, 2 = disag ee, 3 = somewha
disag ee, 4 = nei he ag ee no disag ee, 5 = somewha ag ee, 6 = ag ee, 7 = s ongly ag ee]
6. S a emen : ‘The use o his g ound o se ing p ices is ai .’
[Indica e answe o each g ound]
7. S a emen : ‘Pe sonally, I ind i accep able o base p ices on his g ound.’
[Indica e answe o each g ound]
8. S a emen : ‘In socie y, i is conside ed accep able o base p ices on his g ound.’
[Indica e answe o each g ound]
9. S a emen : ‘The use o his g ound o se ing p ices should be legally p ohibi ed.’
[Indica e answe o each g ound]
10. S a emen : ‘The use o his g ound o se ing p ices would diminish my us in he ma ke .’
[Indica e answe o each g ound]
Sec ion 4 – Gene al ques ions
11. Please indica e o wha ex en you ag ee o disag ee wi h he ollowing s a emen s:
[Answe s on a scale o 1 (‘s ongly disag ee’) o 7 (‘s ongly ag ee’)]
- I is accep able i an online s o e cha ges di e en p ices o consume s based on pe sonal da a.
- I is ai i an online s o e cha ges di e en p ices o consume s based on pe sonal da a.
- I is easonable i an online s o e cha ges di e en p ices o consume s based on pe sonal da a.
A en ion check
12. I is e y impo an o he quali y o he esea ch ha you pay close a en ion when comple ing his ques ionnai e.
Please indica e neu al o his s a emen .
▢ Disag ee
▢ Somewha disag ee
▢ Neu al
▢ Somewha ag ee
▢ Ag ee
13. Please indica e o wha ex en you ag ee o disag ee wi h he ollowing s a emen :
‘I ind i accep able o cha ge people wi h a highe willingness o pay a highe p ice.’
[Answe s on a scale o 1 (‘s ongly disag ee’) o 7 (‘s ongly ag ee’)]
14. How o en ha e you made an online pu chase on a e age in he las 6 mon hs? Choose he answe ha comes
closes .
▢ Daily
▢ Weekly
▢ Mon hly
▢ Less han mon hly
▢ Ne e
15. Wha is you a e age daily (non-wo k- ela ed) in e ne use in hou s?
▢ Less han one hou
▢ 1 o 2 hou s
▢ 2 o 4 hou s
▢ 4 o 6 hou s
36 In e ne Policy Re iew 13(4) | 2024
▢ Mo e han 6 hou s
16. How much money ha e you spen in o al on online pu chases in he pas 6 mon hs?
▢ €0
▢ €1 – €50
▢ €51 – €100
▢ €101 – €300
▢ €301 – €500
▢ €501–€1000
▢ Mo e han €1000
30. Do you ha e any commen s o would you like o sha e some hing ha was no co e ed in he su ey?
[open ques ion]
in coope a ion wi hPublished by
37 Heida y, an de Res , Cus e s
