Improving the public's willingness to purchase near-expired food to reduce food waste: The case of milk products in China

Cheng, Shujun; Shi, Xuanhao; Ren, Yanjun; Zhao, Minjuan
A icle — Published Ve sion
Imp o ing he public's willingness o pu chase nea -
expi ed ood o educe ood was e: The case o milk
p oduc s in China
Ag icul u al Economics (Zemědělská ekonomika)
P o ided in Coope a ion wi h:
Leibniz Ins i u e o Ag icul u al De elopmen in T ansi ion Economies (IAMO), Halle (Saale)
Sugges ed Ci a ion: Cheng, Shujun; Shi, Xuanhao; Ren, Yanjun; Zhao, Minjuan (2025) : Imp o ing he
public's willingness o pu chase nea -expi ed ood o educe ood was e: The case o milk p oduc s
in China, Ag icul u al Economics (Zemědělská ekonomika), ISSN 1805-9295, Czech Academy o
Ag icul u al Sciences, P ague, Vol. 71, Iss. 2, pp. 86-98,
h ps://doi.o g/10.17221/166/2024-AGRICECON ,
h ps://ag icecon.ag icul u ejou nals.cz/a key/age-202502-0002_imp o ing- he-public-s-
willingness- o-pu chase-nea -expi ed- ood- o- educe- ood-was e- he-case-o -milk-p oduc .php
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/312436
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-nc/4.0/
86
O iginal Pape Ag icul u al Economics – Czech, 71, 2025 (2): 86–98
h ps://doi.o g/10.17221/166/2024-AGRICECON
Suppo ed by he Na ional Na u al Science Founda ion o China (G an Nos.72173097, 72373117), Key Special Funds o  he
Minis y o Ag icul u e and Minis y o Finance (G an No.CARS-07-F-1), and he No hwes A&F Uni e si y (G an
No.JGYJSCXXM202302).
© Theau ho s. This wo k islicensed unde aC ea i e Commons A ibu ion-NonComme cial 4.0 In e na ional (CC BY-NC4.0).
Wi h he apid de elopmen o  he economy and
u baniza ion, global ood was e has been u he ag-
g a a ed (Thybe g and Tonjes 2016), and nea -expi ed
ood (NEF) was e has become an impo an compo-
nen . Acco ding o he Food and Ag icul u e O gani-
za ion o  he Uni ed Na ions (FAO), abou one- hi d
o  he ood (abou 1.3 billion ons) is los o  was ed
du ing p oduc ion and consump ion each yea globally
Imp o ing he public's willingness opu chase nea -
expi ed ood o educe ood was e: The case o milk
p oduc s inChina
Shujun Cheng1, Xuanhao Shi1, Yanjun Ren1,2, Minjuan Zhao1,3*
1College o Economics and Managemen , No hwes A&F Uni e si y, Yangling, P.R. China
2Depa men o Ag icul u al Ma ke s, Leibniz Ins i u e o Ag icul u al De elopmen inT ansi ion
Economies (IAMO), Halle (Saale), Ge many
3College o Economics, Xi'an Uni e si y o Finance and Economics, Xi'an, P.R. China
*Co esponding au ho : minjuan.zhao@nwsua .edu.cn
Shujun Cheng and Xuanhao Shi con ibu ed equally o his wo k
Ci a ion: Cheng S., Shi X., Ren Y., Zhao M. (2025): Imp o ing he public's willingness opu chase nea -expi ed ood o educe
ood was e: The case o milk p oduc s inChina. Ag ic. Econ. – Czech, 71: 86–98.
Abs ac : The nea -expi ed ood (NEF) isasigni ican oppo uni y o educe ood was e, while consume s o en associ-
a e NEF wi h sa e y issues, which esul s inala ge amoun o sa e and heal hy ood being was ed globally. This esea ch
ocuses on ood da e labelling (FDL) and explo es how consume s' label cogni ion impac s hei willingness opu chase
NEF. Using a andom sampling me hod online, weob ain 2113 alid samples om China and conduc anin o ma ion
in e en ion 'quasi-na u al expe imen ' oob ain pa icipan s' FDL cogni ion and willingness opu chase he nea -ex-
pi ed milk (NEM) be o e and a e he in e en ion and e alua e he impac o  he in e en ion h ough hedi e enc-
es-in-di e ences model. The esul s show ha consume s' ini ial pu chase willingness o NEM islow, and hei FDL
cogni ion has aposi i e e ec , especially inEas e n China and highe educa ion consume s. In o ma ion in e en-
ion inc eases consume s' willingness opu chase NEM bychanging hei label cogni ion, and he in e en ion has
amo e p onounced impac among olde , male, and highe educa ion consume s. Conside ing hep essu e on esou ces
and he en i onmen caused by ood was e has become animpedimen osus ainable de elopmen , he indings expand
heapplica ion o  he Knowledge, A i ude, and P ac ice (K-A-P) heo y in he NEF ield and clea ly e eal he impo -
an ole o elimina ing consume p ejudice o FDL in educing ood was e oachie e he Uni ed Na ions Sus ainable
De elopmen Goal 12.3 'Hal e ood was e.'
Keywo ds: di e ence-in-di e ences; expe imen al economics; ood da e labelling; ood indus y; ood was e
87
Ag icul u al Economics – Czech, 71, 2025 (2): 86–98 O iginal Pape
h ps://doi.o g/10.17221/166/2024-AGRICECON
(Gus a ssone al. 2011), h ea ening global ood secu-
i y (Kuipe and Cui 2021). A  he e ail and consume
s ages, ala ge amoun o edible ood isdisca ded (Mel-
bye e al. 2017). Abou 28% o  oods inAus ia go un-
sold because hey pass he bes -be o e o use-by da e
(Lebe so ge and Schneide 2014). In he Uni ed S a es,
mo e han 90% o Ame icans h ow away ood p ema-
u ely because hey ega d da e labels as ood sa e y in-
dica o s ( 2012), which b ings abou 20% o  ood was e
o he consume s age (FDA 2019). A  p esen , only
a ew de eloped economies ha e paid a en ion o his
issue, and he cu en s a us o in ansi ion economies
iss ill unknown.
The consump ion o subop imal ood isa i al oppo -
uni y oimp o e he u iliza ion e iciency o  esou c-
es and educe ood was e (Bai e al. 2022), especially
he NEF (Zhang e al. 2023). NEF e e s o ood abou
o each he es ic ion da e on ood packaging (s ill
wi hin he da e). Howe e , consume s a e eluc an
opu chase he NEF (Aschemann-Wi zel e al. 2018).
Al hough he NEF can educe ood en e p ises' u n-
o e and managemen cos s and enable consume s
obene i om lowe p ices, 'Bad quali y' isconsum-
e s' inhe en imp ession o  he NEF (Xu 2013). Nea ing
expi a ion da e makes consume s eel ha ood sa e y
will no beup o hei expec a ions (Aschemann-Wi -
zel e al. 2018), and exis ing esea ch poin s ou ha
69% o 84% o  consume s belie e ha he quali y
o pe ishable p oduc s dec eases o e ime (Tsi os and
Heilman 2005). Inaddi ion, p ice is an ex e nal cue,
and p oduc ea u es (such as he expi a ion da e ap-
p oaching) a e anin e nal cue, in luencing consume s'
willingness o pu chase he NEF. Al hough educing
he p ice can inc ease he ma ke a ac i eness and
is he mos di ec way oimp o e consume s' pu chase
willingness o  NEF, hey o en in e he quali y le el
based on he p ice and ein o ce he cogni ion o  he
declining quali y (Aschemann-Wi zel e al. 2017; The-
o okis e al. 2012), while his p ice educ ion may jus
aim o educe p essu e on he s ock.
Food da e labelling (FDL) is he c i e ion o de in-
ing he nea -expi ed s a e, while he de ini ion o NEF
has ye obeuni ied globally. InChina, NEF has ye
o be uni ied na ionwide, and only a ew local go -
e nmen s ha e made p elimina y egula ions (see Ta-
ble 1). FDL in China consis s o  manu ac u ing da e
and quali y gua an eed du a ion, collec i ely e e ed
oas'quali y gua an eed da e label,' which isanindi-
ca o o  ood quali y. NEF is he medium o FDL, and
consume s' unde s anding o FDL mainly de e mines
hei willingness oconsume NEF. Howe e , e idence
om ad anced economies sugges s ha many consum-
e s dono know he di e ence be ween he a ious ex-
p essions o FDL on he ma ke (Milne 2012), and hey
o en misunde s and FDL asanindica o o  ood sa e y
(Newsome e  al. 2014; Ve ghese e al. 2015), di ec ly
leading o he w ong cogni ion o  he NEF and esul -
ing a la ge amoun o  sa e, heal hy, and high-quali y
ood being was ed (Hall-Phillips and Shah 2017). The e
isase e e in o ma ion asymme y in he ma ke 's un-
de s anding and use o FDL, and he FDL has anim-
po an ole in educing ood was e (Ka anaugh and
Quinlan 2020). Food manu ac u e s use FDL oindi-
ca e he quali y and eshness o  ood p oduc s ocon-
sume s, while consume s mis akenly belie e ha FDL
isin ended ocon ey ood sa e y in o ma ion.
Based onsuch in o ma ion asymme y, i seems un-
de s andable ha consume s esis NEF o ood sa e y
o a oid he isk o  oodbo ne diseases. The heo y
o Knowledge, A i ude, and P ac ice (K-A-P) ismainly
used oexplain he e ec s o indi idual knowledge and
o iginal belie s on changes o  indi idual beha iou s,
Table 1. The de ini ion o NEF inChina
No. Du a ion o quali y gua an eed C i e ion Example
1Mo e han one yea 45 cans, candy, cookies
2Mo e han six mon hs bu less han one yea 20 ins an noodles, asep ically packaged milk and juice
3Mo e han 90 days and less han six mon hs 15 some acuum-packed and e ige a ed cooked ood
4Mo e han 30 days and less han 90 days 10 some s e ilised packaged mea and esh eggs
5Mo e han 16 days and less han 30 days 5yogu , some desse
6Less han 15 days 1–4 uns e ilised cooked ood, boxed soy p oduc s
The c i e ion column means 'X days be o e he quali y gua an eed da e'
Sou ce: Acco ding o he di e en quali y gua an eed du a ion o di e en oods, he NEF isclassi ied in o six si ua ions,
which is he mos widely ecognised s anda ds o 'The ood da e label c i ical' issued by he Beijing Bu eau o Indus y
and Comme ce
88
O iginal Pape Ag icul u al Economics – Czech, 71, 2025 (2): 86–98
h ps://doi.o g/10.17221/166/2024-AGRICECON
and is he heo e ical basis o o mula ing beha iou
in e en ion s a egies (Pä na e  al. 2005), which has
been widely used in a eas such as was e educ ion
(Liu e al. 2023) and adap a ion oclima e change (Cai
e al. 2024). This heo y di ides indi idual beha iou
ans o ma ion in o h ee s ages: acqui ing knowledge
o in o ma ion, o ming belie s and changing beha iou .
Speci ically, 'Knowledge' e e s o he acquisi ion and
unde s anding o some common-sense knowledge, 'A -
i ude' e e s o he co ec consciousness and posi i e
a i ude owa d some hing, and 'P ac ice' e e s o he
ans o ma ion o indi idual beha iou a e he acqui-
si ion and unde s anding o knowledge and p oducing
co esponding belie s. Consume s will only change
hei exis ing consump ion beha iou i hey awa e ha
i ha ms hei heal h, and changing i can educe he
isk (Con en o and Mu phy 1990). Based on he abo e
heo e ical analysis, andom in e en ion expe imen s
in expe imen al economics can achie e he ealiza-
ion o ace ain s imulus oconsume s oachie e he
change o  hei 'knowledge.' Byin o ming he meaning
o  he FDL, weconduc he FDL in o ma ion in e en-
ion onconsume s ochange hei cogni ion o NEF
ood ocomple e he cogni i e change p ocess o  he
i s s age ('knowledge'). A e he in o ma ion in e -
en ion, his s imulus gi es consume s anew cogni ion
and unde s anding o  he FDL and NEF ('A i ude') and
is expec ed o in luence hei subsequen ood con-
sump ion decisions ('P ac ice'). A u he explana ion
iswhen consume s ealize ha NEF will no pose ood
sa e y o heal h haza ds, he co esponding willingness
oconsume NEF may inc ease.
Many s udies ha e ocused on he ma ke ing me h-
ods o subop imal ood, mainly on he impac o p ice
discoun s on consume s' pu chase willingness and
he ole o  in o ma ion in e en ion in imp o ing
consume s' pe cei ed quali y. On he one hand, me -
chan s a ac consume s' a en ion odi e en ypes
o subop imal ood byse ing p ice discoun s osa is y
hei u ili y a alowe p ice (Aschemann-Wi zel e al.
2018). On he o he hand, p o iding posi i e in o ma-
ion abou ood while lowe ing p ices can ensu e con-
sume s' pe cei ed quali y o subop imal ood and sell
success ully. Aschemann-Wi zel e  al. (2019) use he
cos -sa ing in o ma ion label 'Lowe p ice, sa e mon-
ey' and he was e educ ion label 'Do no was e ood'
oin e e e wi h consume s' pu chase willingness o
subop imal ood and ind ha bo h in e en ion in o -
ma ion imp o es he pe cei ed quali y and pu chase
willingness o subop imal ood. Giesen and Hooge
(2019) use 'Accep ing de ec i e ood: Joining he igh
agains ood was e' and 'I isanapple (ca o ) inaspe-
cial shape: Do no was e hem' in o ma ion in e en-
ions o change consume s' pu chase willingness o
oods wi h appea ance de ec s, and hey ind ha in-
o ma ion wo ks well. Al hough in o ma ion in e en-
ion asane ec i e ool has been widely used in ood
was e mi iga ion, he exis ing in o ma ion in e en ion
ismainly desc ip i e language (Eu opean Commission
2018 Uni ed S a es En i onmen al P o ec ion Agen-
cy 2019; FAO 2020) and indica es p oduc a ibu es
(Aschemann-Wi zel e  al. 2019; Giesen and Hooge
2019; Schneide and Ghosh 2020) oin o m consum-
e s o educe ood was e. Howe e , he e isa ela i e
lack o s a is ical analysis onsubop imal ood h ough
non-p ice o  non-decla a i e s a emen in e en ion
me hods. Fo NEF, he ela ionship be ween consum-
e s' cogni ion o FDL and hei willingness oconsume
NEF has no been empi ically es ed, and s udies o in-
o ma ional in e en ions based on he meaning o FDL
a e a e. Inpa icula , quasi-na u al expe imen s ha
can simula e eal consump ion scena ios can beused
oob ain mo e ealis ic da a, while a oiding po en ial
endogenei y p oblems. No ably, he exis ing s udies a e
mos ly om de eloped economies, and he e isalack
o  ele an s udies om ansi ion economies.
Ou s udy aims oempi ically es he ela ionship
be ween consume s' cogni ion o FDL and hei NEF
pu chase willingness, and hen h ough conduc ing
FDL in o ma ion in e en ion o educe hei mis-
unde s andings abou FDL o explo e he in luence
o in o ma ion on hei pu chase willingness o NEF.
Finally, wea emp odisce n he he e ogenei y among
di e en g oups and con ibu e new ideas o educe
ood was e om ansi ion economy.
MATERIAL AND METHODS
Sampling and da a. Conside ing ha consume s
may be mo e sensi i e o he sa e y o  pe ishables,
ou esea ch ocuses on nea -expi ed pe ishables,
which is ep esen a i e in he NEF esea ch ield.
In his s udy, we ake milk (pe ishable, dai y p od-
uc s), which has a la ge impac on he en i onmen ,
a ep esen a i e o pe ishable ood pu chased bycon-
sume s daily and also acommon ocus in ood was e
esea ch (Blondin e  al. 2017; Conno s and Schuelke
2022), asou esea ch objec .Weob ain he pe cap-
i a milk consump ion da a om he las i e yea s
om he China S a is ical Yea book (The Na ional
Bu eau o S a is ics o China, 2016–2020) and se-
lec s six p o incial capi als acco ding o he ank
89
Ag icul u al Economics – Czech, 71, 2025 (2): 86–98 O iginal Pape
h ps://doi.o g/10.17221/166/2024-AGRICECON
o  he pe capi a milk consump ion, pe capi a g oss
domes ic p oduc (GDP), and egional dis ibu ion
cha ac e is ics.
We conduc anonline ques ionnai e su ey om he
p o essional and well-known Chinese online su ey
pla o m, 'Ques ionnai e S a ' (h ps://www.wjx.cn,
a pla o m simila o Syna a) in China in 2023 wi h
andom sampling. The 'Ques ionnai e S a ' has o e
6.2million egis e ed use s, including o e h ee mil-
lion ac i e use s, co e ing 31p o incial adminis a i e
di isions inChina (excluding Hong Kong, Macao, and
Taiwan). The su ey company consis en ly p o ides
quali ied su ey dis ibu ion and collec ion se ices
h ough sc eening pa icipan s wi h anin eg i y sco e
sys em (Lin and Guan 2021). The p ocess is olun a y
and pa icipan s can s op answe ing and exi any ime.
Be o e he s a o ou su ey, wemake i clea oall
pa icipan s ha he ques ionnai e isen i ely anony-
mous, he collec ed da a will only beused o scien i ic
esea ch, and he e isno igh o w ong inall ques-
ions. Pa icipan s mus bea leas 18yea s old, and
each pa icipan isallowed osubmi only once. Since
he sample g oups o  he 'Ques ionnai e s a ' ha e hei
unique co esponding codes, he numbe o  answe s
o each pa icipan is es ic ed s ic ly. Pa icipan s
a e andomly assigned oei he he in e en ion g oup
o  he con ol g oup. When he numbe o ques ion-
nai es collec ed eaches he expec ed numbe , he
ques ionnai e collec ion au oma ically s ops.
To ensu e he quali y o  he online ques ionnai e,
wese up mul iple a en ion es ques ions and epea -
ed ques ions in di e en places o  he ques ionnai e
o e i y he pa icipan s' se iousness inanswe ing he
ques ionnai e. Inaddi ion, weexclude samples wi h e-
sponse imes ha a e oo long o  oo sho oensu e
he in e en ion's e ec i eness. A e alidi y sc een-
ing, weob ain 2209su ey da a om six ci ies, and he
samples a e e enly dis ibu ed among egions. Toa oid
he in luence caused by he consume s' die a y p e e -
ences, wedele e he samples ha dono consume milk
and ob ain 2113 alid samples. The desc ip i e s a is i-
cal cha ac e is ics o  he samples a e shown inTable2,
which de ails he gene al indi idual and amily cha -
ac e is ics o ou pa icipan s, and di e ences inplace
o  esidence. Wealso illus a e he a ia ions o  he co e
a iables in he s udy he e. Only 17.56% o  Chinese
consume s unde s and he meaning o  he FDL co ec -
ly ini ially, collec i ely e e ed oas'quali y gua an eed
da e' in China, which means consume s' cogni ion
o labels isgene ally low, and he in o ma ion ansmis-
sion unc ion o FDL isse iously damaged. Inaddi ion,
compa ed omean4, consume s' ini ial pu chase will-
ingness o NEM isnega i e. Howe e , a e he in o -
ma ion in e en ion, he co esponding alues change.
In o ma ion in e en ion expe imen al design.
Be o e he in o ma ion in e en ion expe imen , we e-
qui e he pa icipan s oiden i y whe he he quali y
gua an eed da e label isaquali y o asa e y indica o
oob ain hei ini ial label cogni ion, ini ial pu chase
willingness o he NEM, and he co esponding de-
mog aphic cha ac e is ics. Then we conduc an in-
o ma ion in e en ion on he meaning o  he quali y
gua an eed da e label o he in e en ion g oup, which
displays he ex 'Quali y gua an eed da e label –
Aquali y indica o indica ing op imal ood as e and
la ou , which will decline beyond his da e bu isedi-
ble' (in Chinese). Fo he con ol g oup, he sc een dis-
plays he ollowing con en : 'This page isin en ionally
le blank' (in Chinese). Finally, we ob ain hei cog-
ni ion o  he quali y gua an eed da e label and hei
willingness o consume he NEM again o cons uc
he panel da a be o e and a e he in o ma ion in e -
en ion expe imen oiden i y he e ec i eness o  he
in e en ion. The whole p ocess akes place onpa ic-
ipan s' phones o compu e sc eens, and hey canno
goback o  e ise any ques ions hey answe ed.
Model se ing. Fi s ly, weuse o dina y leas squa e
(OLS) eg ession oexplo e he e ec s o  he label cog-
ni ion onconsume s' pu chase willingness o NEM, he
speci ic o mula inEqua ion (1):
Yi = β0 + β1Xi + εi
(1)
whe e: Yi – he consume s' sel - epo ed willingness
(Like se en-le el scale, Y= 1, 2, 3, …, 7) opu chase
NEM be o e ecei ing he in o ma ion in e en ion; he
index = 1, … I – he pa icipan ; Xi – con ains 0–1 a i-
able o  he label cogni ion ha cha ac e ize whe he
he consume answe s he meaning o quali y gua an-
eed da e label co ec ly and ase o con ol a iables
inTable2; εi – he cons an e m; β1 – he eg ession
coe icien o  he a iable.
Second, we use di e ence-in-di e ences (DID) e-
g ession oe alua e he ne e ec o  he in o ma ion
in e en ion on consume s' willingness o pu chase
he NEM. Among he me hods oe alua e he e ec
o policy implemen a ion, he di e ence-in-di e ences
model is he mos p e alen and he oldes quasi-expe -
imen al esea ch me hod widely used in ecen yea s
(Feng e al. 2021; Wang and Ge 2022), which can no
only es he e ec i eness o  he policy o in e en ion,

90
O iginal Pape Ag icul u al Economics – Czech, 71, 2025 (2): 86–98
h ps://doi.o g/10.17221/166/2024-AGRICECON
bu also u he iden i y he ne e ec o  he impac .
The basic idea o  his me hod is o ega d he imple-
men a ion o anew policy o in e en ion asa'na u al
expe imen ' o  'quasi-expe imen ' ha is exogenous
o he economic sys em (Chen and Wu 2015). On he
one hand, he implemen a ion o ou in o ma ion in-
e en ion may make he pa icipan s' label cogni ion
and pu chase willingness di e en be o e and a e he
in e en ion. On he o he hand, he abo e wo indica-
o s may di e be ween he in e en ion and con ol
g oups a  he same ime. The DID model eg ession
es ima ion can e ec i ely con ol he impac o o he
Table 2. Desc ip i e s a is ics
Con ol a iables Desc ip ions De ini ions Mean SD Min. Max. Obs.
Age he age o  he pa icipan s yea 30.24 7.44 18 69 2113
Gende gende o pa icipan s male = 1; emale = 0 0.42 0.49 0 1 2113
Educa ion educa ion le el
o pa icipan s
bachelo deg ee
o abo e = 1; below
bachelo deg ee = 0
0.78 0.41 0 1 2113
Risk a i ude isk a i ude
o pa icipan s
isk a e sion = –1;
isk neu al = 0;
isk p e e ence = 1
–0.45 0.85 –1 12113
Heal h s a e he heal h s a e o pa icipan s'
amily membe s
all heal hy = 1;
someone unheal hy
= 0
0.76 0.43 0 1 2113
Disposable income pe capi a household
disposable income USD/mon h 640.22 505.61 46.89 2816 2113
Family size he numbe o  he pa icipan 's
amily membe s people 2.97 1.33 1 8 2113
Sample loca ions
Eas e n egion Beijing ci y – – – – 365
Guangzhou ci y – – – – 361
Cen al egion He ei ci y – – – – 350
Changchun ci y – – – – 341
Wes e n egion Xining ci y – – – – 344
Chengdu ci y – – – – 352
U ban a ea 95.60%
Co e a iables Desc ip ions De ini ions
In e en ion g oup
(N= 1 127)
Con ol g oup
(N= 986)
mean SD mean SD
Be o e in e en ion
Label cogni ion whe he he quali y gua an eed
da e label isanswe ed co ec ly
co ec = 1;
inco ec = 0 0.20 0.40 0.15 0.36
Pu chase willingness I will buy NEM se en-Like scale 3.95 1.71 3.44 1.72
A e in e en ion
Label cogni ion whe he he quali y gua an eed
da e label isanswe ed co ec ly
co ec = 1;
inco ec = 0 0.62 0.49 0.14 0.34
Pu chase willingness I will buy NEM se en-Like scale 4.10 1.62 3.39 1.63
Se en-Like scale: S ongly disag ee = 1, disag ee = 2, somewha disag ee = 3, gene al = 4, somewha ag ee = 5, ag ee
= 6, s ongly ag ee = 7; wedi ide six ci ies in o h ee egions based on he egional di ision me hod o Lie al. (2020);
SD – s anda d de ia ion; Obs. – numbe o obese a ions; NEM – nea -expi ed milk
Sou ce: Au ho s' wo k
91
Ag icul u al Economics – Czech, 71, 2025 (2): 86–98 O iginal Pape
h ps://doi.o g/10.17221/166/2024-AGRICECON
synch onous in e en ions and he p io di e ence be-
ween he in e en ion and con ol g oups oiden i y
he ne e ec o  he in e en ion (Wang and Ge 2022).
The e o e, ou in o ma ion in e en ion isa'quasi-na -
u al expe imen ,' and he in e en ion's impac can
bee alua ed using he DID me hod. See he ollowing
Equa ion (2) o model cons uc ion:
Yi = β0 + β1 ea i + β2pos + β3 ea i ×
× pos + β4Xi + εi
(2)
whe e: Yi – he pa icipan 's willingness oconsume
NEM be o e and a e he in o ma ion in e en ion;
he index = 0 and 1 – a ime se ies o da a; ea i –
adummy a iable ep esen ing he g oup e ec o  he
in e en ion g oup, and his dummy a iable isex ac ed
om he su ey; pos – a ime dummy a iable, which
cap u es he ime e ec o  he in e en ion pe iod; ea i
× pos – he eal e ec o  he in e en ion g oup du ing
he in e en ion pe iod; Xi – pa icipan s' age, gende ,
educa ion, isk a i ude, pe capi a disposable income,
amily heal h s a us, and amily size; β3 – ou key in e -
es , coe icien which ep esen s he ne e ec o  he
in e en ion.
RESULTS
The co ela ion be ween label cogni ion and NEM
pu chase willingness
OLS benchma k eg ession. Label cogni ion posi-
i ely a ec s NEM's pu chase willingness. Wemainly
ocus on he impac o ini ial label cogni ion oncon-
sume s' ini ial willingness o pu chase NEM, and
he esul s o  OLS benchma k eg ession a e shown
in he i s column o Table3. Fo one uni inc ease
inlabel cogni ion, he co esponding NEM pu chase
willingness inc eases by 0.258, which con i ms ou
conce ns. The highe he label cogni ion, he mo e ex-
plici ly consume s know ha he NEM issimply ood
ha las s in as e o  la ou while s ill edible, b inging
highe pu chase willingness.
Al hough wemainly conce n abou label cogni ion,
wes ill ha e some in e es ing indings ega ding con-
ol a iables. Among he con ol a iables, he age
o  pa icipan s signi ican ly nega i ely a ec s hei
willingness opu chase NEM, which means ha young
consume s may ha e a highe accep ance o  NEM.
The ela ionship be ween consume s' pu chase will-
ingness and pe capi a household disposable income
is e e sed, possibly ela ed o weal hie consume s'
highe demand o eshe oods (Ne e  al. 2015).
Ano he possible eason is ha NEM iso en accom-
panied by sales discoun s, which may be mo e a -
ac i e o consume s wi h lowe disposable income
osa e ondaily expenses and b ing ahighe willing-
ness opu chase.
Robus ness es . Toensu e he obus ness o  he e-
sul s, we es he obus ness by changing dependen
a iables and expanding he sample. i) Changing de-
penden a iable. Weuse he answe ob ained by'I will
buy NEM o educe ood was e' asasubs i u e a iable
Y2 o he dependen a iable o u he eg ession
analysis, and he eg ession esul s a e shown in he
second column o Table3. The e ec o label cogni ion
onNEM pu chase willingness s ill passes he signi i-
Table 3. The eg ession esul o  he co ela ion be ween
label cogni ion and NEM pu chase willingness
Va iables (1) (2) (3)
Y1Y2Y1
Label cogni ion 0.258*** 0.167* 0.252***
(0.097) (0.096) (0.096)
Age –0.021*** –0.011** –0.019***
(0.006) (0.005) (0.005)
Gende –0.038 0.204*** –0.012
(0.077) (0.075) (0.075)
Educa ion 0.111 0.007 0.064
(0.097) (0.094) (0.094)
Risk a i ude 0.002 0.094** –0.016
(0.044) (0.043) (0.043)
Disposable income –0.175*** –0.057 –0.172***
(0.049) (0.047) (0.048)
Family size –0.054* 0.006 –0.051*
(0.030) (0.029) (0.029)
Heal h s a e –0.433*** –0.246*** –0.420***
(0.089) (0.087) (0.088)
Ci y con ols YES YES YES
Cons an 6.099*** 4.682*** 5.987***
(0.353) (0.349) (0.348)
R-squa ed 0.037 0.016 0.033
Obse a ions 2113 2113 2209
*, **, *** P < 0.1, P < 0.05, and P < 0.01, espec i ely; obus
-s a is ics inpa en heses; oa oid he in luence o mul i-
collinea i y on he esul s, weconduc ed he es , and he
esul o mean VIF = 1.31 p o es ha he e isnoob ious
mul icollinea i y inou model; disposable income is ea ed
wi h loga i hm (lnx) in eg essions; Y – he pa icipan 's
willingness opu chase NEM; NEM – nea -expi ed milk
Sou ce: Au ho s' wo k
92
O iginal Pape Ag icul u al Economics – Czech, 71, 2025 (2): 86–98
h ps://doi.o g/10.17221/166/2024-AGRICECON
cance es , which e i ies he obus ness. ii) Expanding
he sample. Since wep e-dele e he samples ha dono
consume milk in he benchma k eg ession oexclude
he in luence o  he ood ype weselec and he die a y
di e ences on he esul s, weexpand o he whole sam-
ple 2209 and conduc he obus ness es again. The
eg ession esul s in he hi d column o Table3 e i y
he obus ness again.
He e ogenei y analysis. Conside ing ha he e
a e al eady some specialised nea -expi ed ood s o es
ineas e n China, wewonde i he e a e signi ican a ea
di e ences in consume cogni ion and hei willing-
ness opu chase. The esul s in he i s h ee columns
inTable4show ha he co ela ion be ween label cog-
ni ion and pu chase willingness passes he signi icance
es in he eas e n a ea, which may be ela ed o he
ac ha consume s in he eas ha e mo e oppo uni-
ies oge in ouch wi h new hings and ela i ely high
pu chase willingness o NEM. In addi ion, i  isalso
cu ious how he educa ional le el o consume s a ec s
his ela ionship. Wecompa e he e ec o  his ac o
byg ouping eg essions o di e en educa ional le -
els (see columns4 and5 inTable4 inde ail), and he
co ela ion be ween label cogni ion and pu chase will-
ingness passes he signi icance es a  he highe edu-
ca ionle el.
The e ec o in o ma ion in e en ion
Desc ip i e analysis: Di e ences be ween be o e
and a e he in o ma ion in e en ion. Since label
cogni ion signi ican ly impac s consume s' willingness
opu chase NEM, we u he demons a e he e ec
o in o ma ion onconsume s' label cogni ion and hei
pu chase willingness h ough andom in o ma ion in-
e en ion expe imen s. The esul s show ha 61.67%
o pa icipan s in he in e en ion g oup co ec ly an-
swe s he meaning o  he quali y gua an eed da e la-
bel a e in o ma ion in e en ion (wi h an inc ease
Table 4. He e ogenei y analysis esul s (OLS model)
Va iables (1) (2) (3) (4) (5)
Eas e n China Cen al China Wes e n China Lowe educa ion Highe educa ion
Label cogni ion 0.355** 0.129 0.265 0.196 0.276**
(0.157) (0.168) (0.185) (0.216) (0.110)
Age –0.022** –0.005 –0.032*** –0.017* –0.022***
(0.010) (0.009) (0.009) (0.009) (0.007)
Gende 0.085 –0.050 –0.161 –0.013 –0.043
(0.130) (0.131) (0.140) (0.175) (0.085)
Educa ion –0.075 0.248 0.178 – –
(0.169) (0.162) (0.173)
Risk a i ude –0.142* 0.018 0.118 0.156 –0.033
(0.074) (0.075) (0.081) (0.100) (0.049)
Disposable income –0.153* –0.191** –0.166* –0.218** –0.162***
(0.081) (0.088) (0.085) (0.111) (0.055)
Family size –0.026 –0.076 –0.057 –0.104* –0.040
(0.047) (0.054) (0.056) (0.061) (0.035)
Heal h s a e –0.582*** –0.347** –0.329** –0.803*** –0.341***
(0.149) (0.158) (0.160) (0.213) (0.097)
Ci y con ols –––YES YES
Cons an 5.862*** 5.347*** 5.922*** 6.935*** 5.981***
(0.547) (0.630) (0.573) (0.726) (0.418)
R-squa ed 0.046 0.019 0.049 0.075 0.030
Obse a ions 726 691 696 466 1647
*, **, *** P < 0.1, P < 0.05, and P < 0.01, espec i ely; obus -s a is ics inpa en heses; disposable income is ea ed wi h
loga i hm (lnx) in eg essions; OLS – o dina y leas squa e
Sou ce: Au ho s' wo k
93
Ag icul u al Economics – Czech, 71, 2025 (2): 86–98 O iginal Pape
h ps://doi.o g/10.17221/166/2024-AGRICECON
o 42.15%), while he cogni i e changes in he con ol
g oup show adec ease, which means in o ma ion in-
e en ion can e ec i ely imp o e consume s' label
cogni ion (Tu ey e al. 2021).
Howe e , i isno enough o us oexplo e changes
incogni ion, and wewould like oknow he u he e -
ec s onpu chase willingness. Fu he , weuse he ke -
nel densi y g aphs oshow he changes inpa icipan s'
NEM pu chase willingness be o e and a e in o ma-
ion in e en ion inbo h he in e en ion and con ol
g oups (see Figu e1). Be o e he in o ma ion in e en-
ion, 38.24% o pa icipan s in he in e en iong oup
has a nega i e a i ude owa d NEM. In con as , a -
e he in e en ion, he p opo ion d ops o33.27%,
and he a e age willingness opu chase NEM inc eas-
es om 3.95 o4.10, while in he con ol g oup, his
ini ial nega i e a i ude isexace ba ed.
The ne e ec o  in o ma ion in e en ion. Al-
hough we can see he changes in label cogni ion
and NEM pu chase willingness be o e and a e he
in o ma ion in e en ion om desc ip i e s a is ics,
we u he use he econome ic analysis DID model
1 2 3 4 5 6 7
The in e en ion g oup
Pu chase willingness o NEM Pu chase willingness o NEM
be o e
a e
0.05
0.10
0.15
0.20
0.25
Densi y
be o e
a e
1 2 3 4 5 6 7
The con ol g oup
0.05
0.10
0.15
0.20
0.25
0.30
Figu e 1. The ke nel densi y g aphs o pa icipan s' pu chase willingness oNEM be o e and a e he in o ma ion
in e en ion
NEM – nea -expi ed milk
Sou ce: Au ho s' wo k
Table 5. The ne e ec o  he in o ma ion in e en ion
and obus eg ession esul s
Va iables (1) (2) (3)
Y1PSM-DID Y2
DID 0.209** 0.220* 0.191*
(0.102) (0.121) (0.103)
Cons an 5.320*** 5.368*** 4.594***
(0.248) (0.283) (0.247)
Con ol a iables YES YES YES
Indi idual ixed e ec YES YES YES
Time ixed e ec YES YES YES
ATT – 8.30 –
R-squa ed 0.062 0.062 0.031
Obse a ions 4226 3378 4226
*, **, *** P < 0.1, P < 0.05, and P < 0.01, espec i ely; obus
-s a is ics inpa en heses; disposable income is ea ed
wi h loga i hm (lnx) in eg essions; Y – he pa icipan 's
willingness opu chase NEM; NEM – nea -expi ed milk;
PSM – p opensi y sco e ma ching; DID – di e ence-in-
di e ences; ATT – a e age ea men e ec on he ea ed
Sou ce: Au ho s' wo k
–30 –20 –10 0 10 20
S anda dised % bias ac oss co a ia es
disposable income
heal h s a e
gende
isk a i ude
educa ion
amily size
unma ched ma ched
age
Figu e 2. Va ia ions ins anda dised bias
Sou ce: Au ho s' wo k