Analy ics and Insigh s Team, The T easu y
Resea ch Repo
S a is ical analysis o e hnic wage gaps in New Zealand
Analy ical Pape , No. 18/03
P o ided in Coope a ion wi h:
The T easu y, New Zealand Go e nmen
Sugges ed Ci a ion: Analy ics and Insigh s Team, The T easu y (2018) : S a is ical analysis o e hnic
wage gaps in New Zealand, Analy ical Pape , No. 18/03, ISBN 978-1-988556-62-8, New Zealand
Go e nmen , The T easu y, Welling on
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/205389
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h ps://c ea i ecommons.o g/licenses/by/4.0/
S a is ical Analysis o E hnic Wage Gaps
in New Zealand
Analy ical Pape 18/03
Sep embe 2018
DISCLAIMER
The iews, opinions, indings, and conclusions o ecommenda ions exp essed in his epo a e
s ic ly hose o he au ho s. They do no necessa ily e lec he iews o he New Zealand T easu y,
S a is ics New Zealand o he New Zealand Go e nmen . The New Zealand T easu y and he
New Zealand Go e nmen ake no esponsibili y o any e o s o omissions in, o o he co ec ness
o , he in o ma ion con ained in his Analy ical Pape .
The esul s in his epo a e no o icial s a is ics, hey ha e been c ea ed o esea ch pu poses
om he In eg a ed Da a In as uc u e (IDI) managed by S a is ics New Zealand.
Access o he anonymised da a used in his s udy was p o ided by S a is ics NZ in acco dance wi h
secu i y and con iden iali y p o isions o he S a is ics Ac 1975. Only people au ho ised by he
S a is ics Ac 1975 a e allowed o see da a abou a pa icula pe son, household, business o
o ganisa ion and he esul s in his pape ha e been con iden ialised o p o ec hese g oups om
iden i ica ion.
Ca e ul conside a ion has been gi en o he p i acy, secu i y and con iden iali y issues associa ed
wi h using adminis a i e and su ey da a in he IDI. Fu he de ail can be ound in he P i acy
impac assessmen o he In eg a ed Da a In as uc u e a ailable om www.s a s.go .nz.
The esul s a e based in pa on ax da a supplied by Inland Re enue o S a is ics NZ unde he Tax
Adminis a ion Ac 1994. This ax da a mus be used only o s a is ical pu poses, and no indi idual
in o ma ion may be published o disclosed in any o he o m, o p o ided o Inland Re enue o
adminis a i e o egula o y pu poses.
Any pe son who has had access o he uni - eco d da a has ce i ied ha hey ha e been shown,
ha e ead, and ha e unde s ood sec ion 81 o he Tax Adminis a ion Ac 1994, which ela es o
sec ecy. Any discussion o da a limi a ions o weaknesses is in he con ex o using he IDI o
s a is ical pu poses, and is no ela ed o he da a’s abili y o suppo Inland Re enue’s co e
ope a ional equi emen s.
ANALYTICAL PAPER 18/03
S a is ical Analysis o E hnic Wage Gaps in
New Zealand
MONTH/YEAR
Sep embe 2018
AUTHORS
Analy ics and Insigh s eam, The T easu y
ISBN (ONLINE) 978-1-98-855662-8
URL T easu y websi e a Sep embe 2018:
h ps:// easu y.go .nz/publica ions/ap/ap-18-03
Pe sis en URL: h p://pu l.oclc/nz /p-2042
ACKNOWLEDGEMENTS
A numbe o T easu y s a ga e use ul commen s
on ea lie d a s o his pape . Any emaining
e o s a e he sole esponsibili y o he au ho s.
NZ TREASURY
New Zealand T easu y
PO Box 3724
Welling on 6008
NEW ZEALAND
Email
Telephone
Websi e
in o ma ion@ easu y.go .nz
64-4-472 2733
www. easu y.go .nz
PURPOSE OF THE ANALYTICAL PAPERS
SERIES The T easu y’s aim in publishing he Analy ical
Pape s se ies is o make his analysis a ailable o
a wide audience and o in o m and encou age
public deba e, wi h he ul ima e aim o in o ming
ou policy ad ice.
Analy ical Pape s a e commissioned as pa o he
T easu y’s co e unc ion in de eloping and
p o iding ad ice o Minis e s. They include wo k
unde aken by s a a he T easu y o o he
go e nmen depa men s, as well as wo k
unde aken o he T easu y by ex e nal
esea che s o consul an s.
Analy ical Pape s do no hemsel es ep esen
policy ad ice.
© C own Copy igh
This wo k is licensed unde he C ea i e Commons A ibu ion 4.0 In e na ional licence. In essence,
you a e ee o copy, dis ibu e and adap he wo k, as long as you a ibu e he wo k o he C own and
abide by he o he licence e ms.
To iew a copy o his licence, isi h ps://c ea i ecommons.o g/licenses/by/4.0/. Please no e ha no depa men al o
go e nmen al emblem, logo o Coa o A ms may be used in any way ha in inges any p o ision o he
Flags, Emblems, and Names P o ec ion Ac 1981. A ibu ion o he C own should be in w i en o m and
no by ep oduc ion o any such emblem, logo o Coa o A ms.

AP 18/03 | S a is ical Analysis o E hnic Wage Gaps in New Zealand
i
Execu i e summa y
This pape es ima es he con ibu ion o di e ences in measu ed pe sonal and job
cha ac e is ics o New Zealand’s e hnic wage gaps. The e a e subs an ial and pe sis en
gaps be ween he a e age hou ly wages o Pākehā, Māo i and Paci ic employees. Su ey
es ima es published by S a is ics NZ show ha he a e age hou ly wage ea ned by Māo i
employees was 82% o he a e age hou ly Pākehā wage in 2017, while he a e age wage
ea ned by Paci ic employees was 77% o he a e age Pākehā wage.
The pu pose o he pape is o in es iga e he ex en o which e hnic g oup di e ences in
demog aphic, p oduc i i y- ela ed o job cha ac e is ics may be con ibu ing o he e hnic
pay gaps. The T easu y has p oduced his pape as one o i s con ibu ions o He kai kei
aku inga, he p e ious Go e nmen ’s Māo i economic de elopmen s a egy.
The pape uses 2016-17 da a om he Household Labou Fo ce Su ey and s anda d
Oaxaca-Blinde decomposi ion me hods. The s a is ical models used he e indica e ha
e hnic di e ences in pe sonal and job cha ac e is ics accoun o 68–73% o he Māo i-
Pākehā hou ly wage gap o males, 75% o he Māo i-Pākehā wage gap o emales, 39–
55% o he Paci ic-Pākehā wage gap o males, and 41–55% o he Paci ic-Pākehā wage
gap o emales.
Educa ional le el and occupa ion a e he wo ac o s ha ha e he la ges impac on
Māo i-Pākehā and Paci ic-Pākehā wage dispa i ies, amongs all hose conside ed. In ou
models o he Māo i-Pākehā wage gap:
•di e ences in highes quali ica ion accoun o 18–22% o he wage gap o males and
22–25% o he wage gap o emales
•di e ences in occupa ion accoun o 26–28% o he wage gap o males and 21–24%
o he wage gap o emales.
In ou models o he Paci ic-Pākehā wage gap:
•di e ences in highes quali ica ion accoun o 19–22% o he wage gap o males and
20–22% o he wage gap o emales
•di e ences in occupa ion accoun o 31–33% o he wage gap o males and 29–33%
o he wage gap o emales.
This implies ha educing e hnic di e ences in educa ional a ainmen and/o occupa ional
s a us could help o educe he size o he agg ega e wage dispa i ies.
The ‘unexplained’ pa s o he wage gaps could be caused by a a ie y o ac o s,
including e hnic g oup di e ences in skills ha ha en’ been included in he models (such
as he ield o he quali ica ion ha is held, English language p o iciency, de ailed
occupa ional skills, o i m-speci ic skills and expe ience); di e ences in p e e ences o
di e en jobs because o hei non-wage cha ac e is ics, and disc imina ion.
Fu u e esea ch could use o he me hods, such as ield expe imen s and in e iews wi h
employe s, o gain insigh s in o he causes and e ec s o e hnic disc imina ion in he
labou ma ke .
AP 18/03 | S a is ical Analysis o E hnic Wage Gaps in New Zealand
ii
Con en s
Execu i e summa y .......................................................................................................... i
1 In oduc ion ............................................................................................................. 1
2 Me hods ................................................................................................................... 2
2.1 Da a sou ces ................................................................................................................... 2
2.2 S udy popula ion ............................................................................................................. 2
2.3 De ini ions and ela i e size o each e hnic g oup .......................................................... 2
2.4 Measu e o ea nings ....................................................................................................... 3
2.5 O he a iables ................................................................................................................ 4
2.6 Oaxaca-Blinde decomposi ion ....................................................................................... 4
3 E hnic g oup di e ences in wages and cha ac e is ics ....................................... 6
4 Con ibu ion o di e ences in cha ac e is ics o he e hnic wage gaps ........... 10
5 Summa y and discussion .................................................................................... 14
5.1 Summa y .......................................................................................................................14
5.2 Wha ha e we lea n abou e hnic wage gaps? ............................................................14
5.3 Wha ha e we lea n abou disc imina ion? ..................................................................15
5.4 O he ypes o e idence on disc imina ion ....................................................................16
5.5 Fu u e esea ch .............................................................................................................17
Re e ences ..................................................................................................................... 18
Lis o igu es
Figu e 1 – Ra io o Māo i and Paci ic people’s hou ly wages o Pākehā hou ly wages ...................... 1
Figu e 2 – Pe cen age o wage and sala y ea ne s in each e hnic g oup and hei
combina ions, HLFS June qua e s 2016-17 ....................................................................................... 3
Figu e 3 – Li e acy and nume acy skills by e hnici y and highes quali ica ion – people
aged 15 o 65.....................................................................................................................................15
Lis o ables
Table 1: E hnic g oup de ini ions ......................................................................................................... 2
Table 2: Mean e hnic g oup hou ly wages and hei a ios: HLFS June qua e s 2016-17 ................ 6
Table 3: A e age pe sonal and job cha ac e is ics o employees in each e hnic g oup ..................... 6
Table 4: Decomposi ion esul s using Pākehā wage eg ession coe icien s as weigh s .................11
Table 5: Decomposi ion esul s using ull sample wage eg ession coe icien s as weigh s ............12
AP 18/03 | S a is ical Analysis o E hnic Wage Gaps in New Zealand
1
1 In oduc ion
This pape es ima es he con ibu ion o di e ences in measu ed pe sonal and job
cha ac e is ics o New Zealand’s e hnic wage gaps. The e a e subs an ial and pe sis en
gaps be ween he a e age hou ly wages o Pākehā, Māo i and Paci ic employees. Su ey
es ima es published by S a is ics NZ show ha he a e age hou ly wage ea ned by Māo i
employees was 82% o he a e age hou ly Pākehā wage in 2017, while he a e age wage
ea ned by Paci ic employees was 77% o he a e age Pākehā wage. These a ios ha e
no changed ma e ially in ecen yea s, as shown in Figu e 1 ( he e has been mo emen
om yea o yea bu no consis en upwa d o downwa d end).
The pape uses 2016-17 da a om he Household Labou Fo ce Su ey and s anda d
Oaxaca-Blinde decomposi ion me hods. The pu pose o he pape is o in es iga e he
ex en o which e hnic g oup di e ences in demog aphic, p oduc i i y- ela ed o job
cha ac e is ics may be con ibu ing o he e hnic pay gaps. T easu y has p oduced his
pape as one o i s con ibu ions o He kai kei aku inga, he Go e nmen ’s Māo i
economic de elopmen s a egy.
E hnic wage gaps a e he ocus o he pape . Fo a b oade pe spec i e on he
employmen and economic ou comes o he New Zealand Paci ic popula ion, see K ieble
and Gampe le (2016). Fo a b oade pe spec i e on Māo i employmen , see Minis y o
Business, Inno a ion and Employmen (2017).
Sec ion 2 desc ibes he me hods. Sec ion 3 p o ides summa y s a is ics on e hnic g oup
di e ences in wages and wage- ela ed cha ac e is ics. Sec ion 4 gi es he esul s o he
decomposi ions. Sec ion 5 discusses he in e p e a ion o he indings, hei limi a ions,
and di ec ions o u he esea ch.
Figu e 1: Ra io o Māo i and Paci ic people’s hou ly wages o Pākehā hou ly wages
Sou ce: Household Labou Fo ce Su ey, S a is ics NZ published es ima es.
70.0
72.0
74.0
76.0
78.0
80.0
82.0
84.0
86.0
88.0
90.0
2009 2010 2011 2012 2013 2014 2015 2016 2017
Pe cen age
Ra io o Paci ic peoples' a e age wage
o Pākehāa e age wage
A e age Hou ly Ea nings Median Hou ly Ea nings
70.0
72.0
74.0
76.0
78.0
80.0
82.0
84.0
86.0
88.0
90.0
2009 2010 2011 2012 2013 2014 2015 2016 2017
Pe cen age
Ra io o Māo i a e age wage
o Pākehāa e age wage
A e age Hou ly Ea nings Median Hou ly Ea nings
AP 18/03 | S a is ical Analysis o E hnic Wage Gaps in New Zealand
2
2 Me hods
2.1 Da a sou ces
Ou main da a sou ce is he June 2016 and June 2017 qua e s o he Household Labou
Fo ce Su ey (HLFS). Income ques ions a e asked in June qua e s only in he HLFS.
We pool da a om he la es wo yea s o he su ey in o de o inc ease sample sizes.
We don’ use da a o ea lie yea s because o changes o he ques ionnai e be ween
2015 and 2016, al e ing some o he key a iables.
The In eg a ed Da a In as uc u e (IDI) is a seconda y da a sou ce, which is used solely
o de i e a measu e o ime spen in wage o sala ied employmen du ing he pas 5
yea s, o each pe son in he s udy popula ion.
2.2 S udy popula ion
The s udy popula ion includes all wage and sala y ea ne s aged 15-69 yea s who
p o ided esponses o he su ey ques ions on hei ea nings, di ec ly o h ough ano he
membe o hei household.
Responden s whose ea nings esponses we e la e impu ed by S a is ics NZ – because
hey answe ed he co e HLFS ques ions bu no he addi ional ea nings ques ions – a e
no included in he sample o analysis. Fou een pe cen o eco ds we e excluded
because o he impu a ion o he pe son’s ea nings. App oxima ely 40% o ea nings
esponses in he emaining sample we e p o ided by a p oxy esponden – ha is, by
ano he membe o he amily.
2.3 De ini ions and ela i e size o each e hnic g oup
E hnic g oups a e de ined as in Table 1.
Table 1: E hnic g oup de ini ions
G oup De ini ion
Māo i People who ga e ‘Māo i’ as one o hei e hnic g oups
Sole Māo i People who ga e ‘Māo i’ as hei only e hnic g oup
Pākehā People who ga e ‘Eu opean’ o ‘Pākehā’ as one hei e hnic g oups
Paci ic peoples People who ga e any o he Paci ic e hnici ies as one o hei e hnic
g oups. This includes all o he islands o he Paci ic (Polynesia and
Melanesia).
The Pākehā g oup inco po a es 73.3% o wage and sala y ea ne s in he sample; he
Māo i g oup 12.4%; sole Māo i 6.4%; and he Paci ic peoples g oup 6.3%.
AP 18/03 | S a is ical Analysis o E hnic Wage Gaps in New Zealand
9
The mean age o his sample o Māo i wage and sala y ea ne s was abou 2 yea s younge
han he mean age o he Pākehā sample. The Paci ic peoples’ sample was abou 3 yea s
younge on a e age. The sole Māo i sample, in con as , was simila in i s age s uc u e and
a e age age o he Pākehā sample (and olde han he ull Māo i sample).
Māo i and Paci ic males we e mo e likely o be join pa en s o dependen child en, and
sligh ly mo e likely o be sole pa en s o dependen child en, han Pākehā males. Tu ning
o emales, he p opo ion o Māo i who we e sole pa en s was much highe han he
p opo ion o Pākehā, while Paci ic women we e mo e likely o ha e dependen child en in
bo h pa ne ships and as sole pa en s.
The egional p o iles o he e hnic g oups a e di e en , wi h Māo i being less likely han
Pākehā o li e in he Auckland, Welling on and Can e bu y egions and mo e likely o li e
elsewhe e in he No h Island, and Paci ic peoples being a mo e likely o li e in Auckland
han anywhe e else. Two- hi ds li e in Auckland.
The quali ica ion p o iles also show some qui e la ge e hnic g oup di e ences, wi h
Pākehā holding highe quali ica ions han he o he wo g oups on a e age. Fo ins ance,
36% o women in he Pākehā sample held a deg ee, compa ed wi h 23% o Māo i women
and 19% o Paci ic women.
Coun ing he numbe o calenda mon hs in he pas 5 yea s in which income om wages
o sala ies was ea ned (any amoun ), we ind ha Pākehā wo ke s had somewha g ea e
employmen con inui y han Māo i o Paci ic wo ke s. The male means a e 48.6 mon hs
o Pākehā, 46.1 mon hs o Māo i, and 44.6 mon hs o Paci ic peoples. The emale
means a e 47.2 mon hs o Pākehā, 43.3 mon hs o Māo i, and 41.7 mon hs o Paci ic
peoples.3
Tu ning o cu en job cha ac e is ics, we conside hou s wo ked because he e is a
co ela ion be ween pa - ime hou s and lowe wages o wo ke s in gene al, bu he
esul s show e y li le di e ence be ween he mean weekly hou s o he di e en e hnic
g oups. Job enu e (ie ime spen in he cu en job) is posi i ely co ela ed wi h wage le el
on a e age, and he enu e measu es do show some small di e ences, wi h Māo i and
Paci ic employees ha ing lowe job enu e on a e age.
Occupa ion is a key de e minan o ea nings, and he s a is ics on he p opo ion o each
e hnic g oup in each 1-digi occupa ional g oup indica e ha Pākehā we e much mo e
likely o hold manage ial o p o essional jobs, and much less likely o be in he wo leas
skilled occupa ional g oups (machine y ope a o s and d i e s and labou e s), compa ed
wi h Māo i and Paci ic peoples. The indus y s a is ics show ha Māo i we e mo e likely
han Pākehā o wo k in he p ima y indus ies, in manu ac u ing, o in anspo , pos al and
wa ehousing, while Paci ic peoples we e mo e likely o wo k in manu ac u ing o in
anspo , pos al and wa ehousing.
3 Mon hs o sel -employmen a e no coun ed, because mon hly da a on sel -employmen income a e no
a ailable.

AP 18/03 | S a is ical Analysis o E hnic Wage Gaps in New Zealand
10
4 Con ibu ion o di e ences in
cha ac e is ics o he e hnic wage gaps
Table 4 and Table 5 se ou ou es ima es o he con ibu ion o di e ences in measu ed
cha ac e is ics o he hou ly wage gaps, using he Oaxaca-Blinde decomposi ion me hod.
No e ha hese decomposi ions a e ca ied ou using he na u al log o wages a he han
wage le els. This is he s anda d app oach, adop ed because aking he log o each
pe son’s wage makes he dis ibu ion o he wage a iable less skewed o he igh , so
ha he mean is close o he cen e o he wage dis ibu ion and close o he wage
ea ned by he ypical wo ke .
In Table 4, coe icien s om a wage eg ession using he Pākehā sample we e used o
weigh he mean di e ences in cha ac e is ics. In Table 5, coe icien s om a wage
eg ession using people o all e hnic g oups a e used o weigh he di e ences in
cha ac e is ics. I Pākehā a e ewa ded mo e highly o each cha ac e is ic, hen he i s
weigh ing app oach will end o lead o highe es ima es o he ‘explained’ po ion o he
o al e hnic wage gap.
The esul s o decomposi ions using pe sonal cha ac e is ics only a e p esen ed in he
uppe sec ion o he ables. The a iables included he e a e age, highes quali ica ion,
pa en al s a us, egion and employmen con inui y (measu ed by he numbe o mon hs o
wage o sala ied employmen in he pas 5 yea s). Age is a p oxy measu e o employmen
expe ience. Highes quali ica ion is a p oxy measu e o job- ela ed skill le el. Pa en al
s a us is a p oduc i i y- ela ed cha ac e is ic because i can a ec indi iduals’ p e e ed
hou s o wo k o willingness o wo k in pa icula jobs and loca ions. Region o esidence
is a p oduc i i y- ela ed cha ac e is ic because labou is mo e aluable in egions o highe
labou demand. The employmen con inui y measu e is ano he measu e o employmen
expe ience.
Focusing on Table 4, he i s i e a iables (age, highes quali ica ion, pa en al s a us,
egion and employmen con inui y) oge he explain 59% o he Māo i-Pākehā wage gap
o males and 68% o he Māo i-Pākehā wage gap o emales. The di e ence in
quali ica ions makes he la ges con ibu ion, wi h smalle con ibu ions coming om age,
egion o esidence, employmen con inui y, and pa en al s a us in he case o emales.
Māo i employees a e younge , less likely o li e in highe paying egions such as
Auckland, Welling on and Ch is chu ch, and i emale a e mo e likely o be sole pa en s.
They had sligh ly less employmen con inui y han Pākehā (as shown in Figu e 3).
The explana o y powe o hese ‘pe sonal’ a iables is lowe when sole Māo i a e
compa ed wi h Pākehā ( e lec ing, among o he hings, he ac ha he sole Māo i g oup
is olde han he o al Māo i g oup). I is subs an ially lowe when Paci ic peoples a e
compa ed wi h Pākehā. These a iables explain 16% o he Paci ic-Pākehā wage gap o
males and 21% o he Paci ic-Pākehā wage gap o emales. The di e ence in
quali ica ions makes he la ges con ibu ion o he wage gap be ween Paci ic peoples and
Pākehā, wi h a smalle con ibu ion coming om di e ences in age ( e lec ing he ac ha
Paci ic employees a e younge ). The egional loca ion o Paci ic peoples has a nega i e
sign in he decomposi ion ( educing he o al explana o y powe o he model), because
Paci ic peoples a e disp opo iona ely loca ed in Auckland, a egion in which Pākehā
wages a e highe han a e age. This esidence pa e n should lead o highe wages o
Paci ic peoples, bu i does no . Two hi ds o he Paci ic peoples in he sample we e li ing
in he Auckland egion, compa ed wi h 28% o Pākehā and 21% o Māo i.
AP 18/03 | S a is ical Analysis o E hnic Wage Gaps in New Zealand
11
Table 4: Decomposi ion esul s using Pākehā wage eg ession coe icien s as weigh s
Māo i
Sole Māo i
Paci ic peoples
M
F
M
F
M
F
Log hou ly wage di e en ial, compa ed wi h
Pākehā 0.191 0.137 0.216 0.142 0.248 0.197
Model 1: Pe sonal cha ac e is ics
Age 0.027 0.009 -0.003 -0.010 0.022 0.017
Pa en al s a us -0.008 0.007 -0.007 0.007 -0.019 0.000
Region 0.020 0.018 0.031 0.023 -0.057 -0.053
Quali ica ions 0.066 0.049 0.081 0.055 0.082 0.063
Employmen con inui y* 0.007 0.011 0.004 0.007 0.011 0.015
All pe sonal cha ac e is ics
0.113
0.093
0.106
0.082
0.039
0.042
% Age 14.3 6.8 -1.5 -7.3 9.0 8.4
% Pa en al s a us -4.1 4.8 -3.3 5.3 -7.8 -0.1
% Region 10.6 12.9 14.4 16.3 -22.9 -26.6
% Quali ica ions 34.6 35.7 37.7 38.4 33.0 31.8
% Employmen con inui y 3.5 8.0 1.9 4.9 4.3 7.5
% Explained 58.9 68.3 49.2 57.5 15.6 21.1
Model 2: Pe sonal and job cha ac e is ics
Age 0.020 0.005 -0.002 -0.005 0.018 0.010
Pa en al s a us -0.007 0.003 -0.006 0.004 -0.016 -0.002
Region 0.015 0.014 0.023 0.019 -0.041 -0.039
Quali ica ions 0.043 0.034 0.052 0.037 0.053 0.044
Employmen con inui y 0.003 0.004 0.002 0.003 0.004 0.005
Hou s 0.002 -0.002 0.002 -0.003 -0.002 -0.005
Job enu e 0.003 0.005 0.001 -0.001 0.005 0.005
Type employmen ela ionship 0.001 0.000 0.000 0.000 -0.001 0.001
Occupa ion 0.049 0.028 0.063 0.029 0.076 0.057
Indus y 0.009 0.010 0.014 0.010 0.001 0.003
All cha ac e is ics
0.139
0.102
0.148
0.093
0.098
0.080
% Age
10.7
3.6
-1.1
-3.4
7.2
5.2
% Pa en al s a us
-3.4
2.2
-2.8
2.7
-6.2
-1.0
% Region
7.7
10.6
10.4
13.4
-16.5
-19.6
% Quali ica ions
22.3
25.2
24.0
26.4
21.5
22.1
% Employmen con inui y
1.4
3.1
0.9
2.0
1.6
2.7
% Hou s
1.0
-1.2
0.9
-2.3
-0.7
-2.3
% Job enu e
1.6
3.4
0.6
-0.7
2.1
2.5
% Employmen ela ionship
0.4
0.2
0.2
0.2
-0.4
0.5
% Occupa ion
25.8
20.5
29.2
20.5
30.5
28.8
% Indus y
4.9
7.2
6.3
6.7
0.3
1.5
% Explained
72.5
74.6
68.5
65.3
39.3
40.5
* Employmen con inui y means mon hs o wage o sala ied employmen in he pas 5 yea s.
AP 18/03 | S a is ical Analysis o E hnic Wage Gaps in New Zealand
12
Table 5: Decomposi ion esul s using ull sample wage eg ession coe icien s as weigh s
Māo i
Sole Māo i
Paci ic peoples
M
F
M
F
M
F
Log hou ly wage di e en ial, compa ed wi h
Pākehā 0.191 0.137 0.216 0.142 0.248 0.197
Model 1: Pe sonal cha ac e is ics
Age
0.024
0.009
-0.003
-0.008
0.022
0.017
Pa en al s a us
-0.006
0.005
-0.005
0.006
-0.016
0.000
Region
0.014
0.014
0.021
0.019
-0.025
-0.031
Quali ica ions
0.062
0.048
0.078
0.055
0.082
0.065
Employmen con inui y
0.009
0.014
0.005
0.008
0.015
0.018
All pe sonal cha ac e is ics
0.103
0.091
0.096
0.080
0.078
0.069
% Age
12.7
6.8
-1.4
-5.8
8.7
8.4
% Pa en al s a us
-3.2
3.9
-2.5
4.2
-6.4
-0.1
% Region
7.3
10.5
9.7
13.3
-10.2
-15.7
% Quali ica ions
32.4
35.3
35.9
38.8
33.2
33.1
% Employmen con inui y
4.6
9.9
2.5
5.9
6.0
9.3
% Explained
53.7
66.4
44.2
56.5
31.2
35.0
Model 2: Pe sonal and job cha ac e is ics
Age
0.017
0.005
-0.002
-0.004
0.016
0.010
Pa en al s a us
-0.005
0.003
-0.005
0.003
-0.012
-0.001
Region
0.008
0.012
0.012
0.016
-0.015
-0.020
Quali ica ions
0.034
0.030
0.042
0.034
0.047
0.040
Employmen con inui y
0.004
0.006
0.003
0.004
0.007
0.008
Hou s
0.001
-0.001
0.001
-0.001
-0.001
-0.003
Job enu e
0.003
0.004
0.001
-0.001
0.006
0.005
Type employmen ela ionship
0.001
0.001
0.000
0.000
0.002
0.002
Occupa ion
0.053
0.033
0.068
0.034
0.083
0.065
Indus y
0.013
0.010
0.018
0.010
0.006
0.004
All cha ac e is ics
0.130
0.103
0.139
0.096
0.137
0.109
% Age
9.1
3.4
-1.0
-2.6
6.5
5.1
% Pa en al s a us
-2.7
1.9
-2.1
2.3
-5.0
-0.7
% Region
4.3
8.8
5.6
11.2
-6.1
-10.3
% Quali ica ions
17.7
22.2
19.5
23.9
18.7
20.1
% Employmen con inui y
2.3
4.6
1.4
2.7
2.8
4.2
% Hou s
0.7
-0.6
0.7
-0.9
-0.4
-1.4
% Job enu e
1.5
3.0
0.6
-0.8
2.4
2.4
% Employmen ela ionship
0.4
0.4
-0.1
0.2
0.8
1.1
% Occupa ion
28.0
24.0
31.6
24.3
33.3
32.7
% Indus y
6.8
7.4
8.3
7.1
2.3
2.2
% Explained
68.0
75.1
64.4
67.3
55.3
55.3
* Employmen con inui y means mon hs o wage o sala ied employmen in he pas 5 yea s.
AP 18/03 | S a is ical Analysis o E hnic Wage Gaps in New Zealand
13
The esul s o decomposi ions using in o ma ion on bo h pe sonal and job cha ac e is ics
a e p esen ed in he lowe sec ion o he ables. The addi ional a iables included in he
second se o wage eg essions as explana o y a iables a e hou s o wo k; job enu e;
whe he he job in ol ed a non-s anda d employmen ela ionship ( ixed- e m,
employmen agency, casual, o seasonal); 1-digi occupa ion and 1-digi indus y. These
job cha ac e is ics we e chosen because measu es a e a ailable in he HLFS and hey
a e empi ically associa ed wi h signi ican wage a ia ions in he labou ma ke .
In Table 4, he ull se o a iables included in he second eg ession model explains 73%
o he Māo i-Pākehā wage gap o males and 75% o he Māo i-Pākehā wage gap o
emales. In hese Māo i-Pākehā wage gap decomposi ions, occupa ion plays he bigges
ole, accoun ing o abou 26% pe cen o he en i e wage gap o males and 21% o
emales. Highes quali ica ion is he second mos impo an a iable. Age, egion o
esidence and indus y each make a small con ibu ion. Di e ences in hou s o wo k, job
enu e, and ype o employmen ela ionship a e o li le impo ance once he o he
a iables ha e been included.
The explana o y powe o he ull se o a iables is simila bu lowe when sole Māo i a e
compa ed wi h Pākehā.
The a iables included in he second eg ession model oge he explain 39% o he
Pākehā-Paci ic wage gap o males and 41% o he Pākehā -Paci ic wage gap o
emales. Occupa ion is he mos impo an explana o y a iable, wi h highes quali ica ion
also making a s ong con ibu ion o he o al wage gap. Pa en al s a us, egion and hou s
o wo k all make e y small nega i e con ibu ions o hese decomposi ions, indica ing ha
he dis ibu ion o hese cha ac e is ics among Paci ic peoples is sligh ly mo e a ou able
owa ds highe wage le els han he dis ibu ion among Pākehā.
By compa ing Table 4 and Table 5 we can see ha he choice o weigh ing scheme
makes li le di e ence o he decomposi ion o he Māo i-Pākehā wage gap. The
‘explained’ p opo ion is only sligh ly highe in Table 5 (see he bo om ows o Table 4
and Table 5). I is a di e en s o y o he Paci ic-Pākehā wage gap: using he coe icien s
om he pooled sample wage eg ession (ie using da a o all e hnic g oups) ma e ially
aises he ‘explained’ p opo ion o he Paci ic-Pākehā wage gap o bo h males and
emales, om less han 40% o abou 55%.
I ’s wo h no ing ha he explana o y powe o e e y ac o is in luenced by he choice o
o he a iables o be included in he model. The explana o y powe o educa ion in he
decomposi ion o he Māo i-Pākehā wage gap, o example, is educed by abou 12
pe cen age poin s when job cha ac e is ics a e also included. The e o e, his me hod does
no gi e a de ini i e, single esul o he ole o a speci ic ac o like educa ion when
unde s anding e hnic wage di e ences. The models a e mos use ul in gi ing an indica ion
o he ela i e impo ance o speci ic ac o s, such as age o educa ion, ela i e o he
o he ac o s ha ha e been included in he same model.
AP 18/03 | S a is ical Analysis o E hnic Wage Gaps in New Zealand
14
5 Summa y and discussion
5.1 Summa y
This analysis has es ima ed he s a is ical con ibu ion o a numbe o measu ed
p oduc i i y- ela ed cha ac e is ics o e hnic wage dispa i ies.
We ind ha educa ional le el and occupa ion a e he wo ac o s, among he se o ac o s
conside ed, ha ha e he la ges impac on Māo i-Pākehā and Paci ic-Pākehā wage
dispa i ies. In he ull models o he Māo i-Pākehā wage gap, di e ences in highes
quali ica ion accoun o 18–22% o he wage gap o males and 22–25% o he wage gap
o emales, while di e ences in occupa ion accoun o 26–28% o he wage gap o
males and 21–24% o he wage gap o emales. In he ull models o he Paci ic-Pākehā
wage gap, di e ences in highes quali ica ion accoun o 19–22% o he wage gap o
males and 20–22% o he wage gap o emales, while di e ences in occupa ion accoun
o 31–33% o he wage gap o males and 29–33% o he wage gap o emales.
This implies ha educing e hnic di e ences in educa ional a ainmen and/o occupa ional
s a us could help o educe he size o he agg ega e wage dispa i ies.
These s a is ical models accoun o 68–73% o he en i e Māo i-Pākehā wage gap o males,
75% o he Māo i-Pākehā wage gap o emales, 39–55% o he Paci ic-Pākehā wage gap o
males, and 41–55% o he Paci ic-Pākehā wage gap o emales. The emaining po ions o
he wage gaps could be caused by a a ie y o ac o s, including di e ences be ween e hnic
g oups o o he cha ac e is ics ha ha en’ been included in he model (such as he ield o
he quali ica ion ha is held, English language p o iciency, de ailed occupa ional skills, and
i m-speci ic skills and expe ience); di e ences be ween e hnic g oups in p e e ences o
di e en jobs because o hei non-wage cha ac e is ics, o disc imina ion.
Abou 18% o he Pākehā wage ea ne s in he s udy sample and 54% o he Paci ic wage
ea ne s in he s udy sample we e bo n o e seas. The e o e, a mo e comple e explana ion
o he Paci ic-Pākehā wage gap would need o ake in o accoun he e ec s o being an
immig an on ea nings, including he e ec s o a ia ions in English language skills,
a ia ions in social capi al and social ne wo ks, and disc imina ion agains mig an s.
These ac o s would also be impo an o any u u e analysis o Asian wage ea ne s.
5.2 Wha ha e we lea n abou e hnic wage gaps?
E hnic g oup di e ences in hou ly ea nings a e pa ly due o e hnic g oup di e ences in
p oduc i i y- ela ed cha ac e is ics and pa ly due o di e ences in job cha ac e is ics. Pa
o each pay gap emains unexplained. The ‘cu en ly unexplained’ po ion is la ge o
Paci ic peoples han o Māo i.
Because o he s a is ical impo ance o e hnic di e ences in educa ional a ainmen and
occupa ional s a us, educing he size o hese di e ences would be pa icula ly use ul.
Alongside hese pay gaps, he e is also he issue o highe unemploymen a es o Māo i
and Paci ic peoples compa ed wi h Pākehā. As a June 2017, he unemploymen a e o
Pākehā was 3.4%, o Māo i i was 11.1% and o Paci ic peoples i was 10.1%. Reducing
he gaps in educa ion a ainmen could also help o educe he unemploymen a es o
Māo i and Paci ic peoples.

AP 18/03 | S a is ical Analysis o E hnic Wage Gaps in New Zealand
15
5.3 Wha ha e we lea n abou disc imina ion?
Many Māo i and Paci ic people expe ience disc imina ion (see, o example, S a is ics
New Zealand 2016). This is backed up in a ange o s udies, desc ibed in Sec ion 5.4
below. This pa icula pape , howe e , is no designed o obus ly es o disc imina ion.
Oaxaca-Blinde decomposi ions we e o iginally de eloped as a ool o es ima ing how
much o a pa icula wage gap migh be due o disc imina ion. His o ically, some
esea che s who p esen ed Oaxaca-Blinde decomposi ions a ibu ed he ‘explained’
po ion o he wage gaps o di e ences in he a e age p oduc i i y o each g oup, and he
‘unexplained’ po ion o disc imina ion.
Mos labou economis s no longe suppo his in e p e a ion and ha e mo ed on o using
o he me hods o s udy disc imina ion. The e a e se e al easons why he ‘unexplained’
po ion o e hnic wage gaps can’ be alidly a ibu ed o disc imina ion.
Fi s , as al eady no ed, he cu en eg ession models don’ include all dimensions o
wo ke s’ skill and labou ma ke p oduc i i y, only he aspec s o which da a a e a ailable.
Fo example, we only know he highes quali ica ion o each wo ke , and no he subjec
ield, he yea and ins i u ion whe e he quali ica ion was ob ained, o whe he mul iple
quali ica ions a e held. No do we ha e measu es o o he ypes o skill ha also in luence
ea nings, such as li e acy skills, nume acy skills, and w i en communica ion skills.
Ano he impo an gap is ha we ha e li le in o ma ion on wo ke s’ employmen his o y,
bu much o he skill a ia ion ha leads o wage a ia ion is acqui ed on- he-job o e he
cou se o a wo ke ’s li e ime, and is in luenced by he o ganisa ions wo ked a and he
speci ic oles ha a e unde aken. Because he measu es o skill included in he model
a e so limi ed, he impac o e hnic skill di e ences on wages is likely o be signi ican ly
unde -es ima ed. This in u n means ha he po ion o he wage gap ha is ‘unexplained’
by p oduc i i y- ela ed cha ac e is ics is likely o be signi ican ly o e -es ima ed.
Figu e 3: Li e acy and nume acy skills by e hnici y and highes quali ica ion – people
aged 15 o 65
Sou ce: Minis y o Educa ion and Minis y o Business, Inno a ion and Employmen , 2016, Figu e 9 and Figu e 10.
200
220
240
260
280
300
320
No o mal
quali ica ions
NCEA Le el 1-3,
Ce i ica e Le el
1-3
Ce i ica e and
diploma Le el
4-7
Bachelo s o
highe
educa ion
A e age li e acy sco e
NZ Eu opean Māo i
200
220
240
260
280
300
320
No o mal
quali ica ions
NCEA Le el 1-3,
Ce i ica e Le el
1-3
Ce i ica e and
diploma Le el
4-7
Bachelo s o
highe
educa ion
A e age nume acy sco e
NZ Eu opean Māo i
AP 18/03 | S a is ical Analysis o E hnic Wage Gaps in New Zealand
16
The Su ey o Adul Skills (2014) p o ides some e idence o di e ences in he skills o
Pākehā and Māo i who ha e he same le el o highes quali ica ion. Figu e 3 ep oduces
Figu e 9 and Figu e 10 om a Minis y o Educa ion and Minis y o Business, Inno a ion
and Employmen epo (2016, pp13-14). I shows ha Māo i adul s in each educa ional
a ainmen g oup ha e lowe li e acy and nume acy skills, on a e age, han NZ Eu opeans
in he same educa ional a ainmen g oup.
Second, disc imina ion in he labou ma ke o in he wide socie y may in luence he
de elopmen o wo ke s’ skills and p oduc i i y- ela ed cha ac e is ics in a a ie y o
di e en ways. Fo example, i he educa ional ou comes o Māo i and Paci ic you h a e
pa ly due o he low expec a ions o hei eache s, o he s uden s’ own pe cep ions o
he jobs hey a e likely o be employed in, hen hei cu en educa ional a ainmen has
po en ially been in luenced by disc imina ion. Simila ly, a wo ke ’s employmen his o y and
hei cu en job cha ac e is ics (and ea nings) can clea ly be shaped by disc imina ion i
he e is disc imina ion in hi ing. This means ha he ‘explained’ componen o he wage
gap is no ee om he e ec s o disc imina ion, and disc imina ion, i i exis s, is likely o
be con ibu ing o bo h he ‘explained’ and ‘unexplained’ po ions o he ea nings gap.
5.4 O he ypes o e idence on disc imina ion
Because o he di icul ies o iden i ying disc imina ion using eg ession decomposi ions,
his app oach has been la gely eplaced in he economics discipline by o he app oaches,
such as p oduc ion- unc ion analyses in which he p oduc i i y o each wo ke is es ima ed
and compa ed wi h hei wages, and expe imen s ha a e designed o s udy he beha iou
o people in con ex s whe e disc imina ion can occu . These me hods a e e iewed in
Neuma k (2016). Expe imen al e idence is a ou ed on he g ounds ha his ype o
e idence is mo e able o demons a e he exis ence o disc imina ion, p o ided he
expe imen s a e well designed.
Re iewing he e idence om expe imen s conduc ed in he US, Newma k no es ha
much o his e idence p o ides con i ming e idence o hi ing disc imina ion, especially
wi h espec o espec o e hnici y and ace (ibid, p73). Newma k says ha ‘ his e iew
ein o ces he conclusion ha hi ing disc imina ion is pe asi e’ (p.74). I disc imina ion in
hi ing educes he likelihood o applican s om e hnic mino i ies being ec ui ed o being
ec ui ed o be e -paying jobs, hen i is likely o be one o he causes o e hnic gaps in
a e age ea nings. This could be he case in New Zealand as well as in he US.
The e is a small body o li e a u e on bias agains Māo i in hi ing decisions. This mos ly
comp ises psychological labo a o y expe imen s o sel - epo ed a i udes in su eys o
in e iews. Fo example, McKenzie (1996) inds bias in a i udes owa ds Māo i among
employe s su eyed. A u he body o wo k inds bias agains o he e hnic g oups such as
Indian and Chinese, such as Wilson e al (2005) and Wilson and Pa ke (2007).
Using a sample o 210 uni e si y s uden s, Singe and Ede (1989) es pe cep ions o
highly quali ied and less quali ied Māo i wi h and wi hou s ong Māo i accen s. Māo i
applican s we e less likely o be sho lis ed han Du ch applican s. In a simila psychology
s udy, Jackson and Fische (2007) es pe cep ions o Māo i and Pākehā in a labo a o y
expe imen . They p esen ed wo se s o iden ical esumes o 114 Pākehā unde g adua e
psychology s uden s. Iden ical high-quali ica ion esumes we e p esen ed. O hese, Māo i
we e a ed sligh ly mo e a ou ably han high-me i Pākehā. Howe e , when low-
quali ica ion esumes we e p esen ed, he iden ical Māo i esumes we e a ed lowe .
AP 18/03 | S a is ical Analysis o E hnic Wage Gaps in New Zealand
17
The e appea s o be a gap in he li e a u e: using expe imen s and da a om eal
employe s o explo e he occu ence o and easons o disc imina ion agains Māo i and
Pasi ika. Jackson and Fische (2007) w i e:
“... ela i ely less esea ch has been conduc ed on he e ec s o applican e hnici y
in w i en job applica ions (i.e. cu icula i ae o applica ion o ms), pa icula ly
compa ing Māo i and New Zealand Eu opean/ Pākehā applican s.”
A decade la e , in 2018, his appea s o s ill be he case.
The bes cu en e idence ha e hnic mino i ies may ace a highe le el o disc imina ion
han Pākehā comes om he New Zealand Gene al Social Su ey, 2014. This ound ha
26% o Māo i el hey had been disc imina ed agains o e he las yea , as did 20% o
Paci ic peoples and 27% o Asian peoples, compa ed wi h 15 pe cen o New Zealand
Eu opeans.4 These igu es include all o ms o disc imina ion on all g ounds (eg, gende ,
e hnici y, age, appea ance, o disabili y).
5.5 Fu u e esea ch
Neuma k (2016) no es ha while he e is ex ensi e expe imen al e idence on
disc imina ion in hi ing, s udying he ole o disc imina ion in o he employe decisions
(such as pay se ing o p omo ion) has been much mo e di icul , and a compa able se o
expe imen al e idence on disc imina ion in pay decisions o p omo ions does no ye exis .
He a gues ha u u e expe imen al esea ch should do mo e han simply demons a e ha
some popula ion g oups ace employe disc imina ion in hi ing. To design good policy
esponses, he na u e o and he easons o he disc imina ion need o be known. These
may a y ac oss di e en con ex s and popula ion g oups. Fo example, g oups ha a e
no iceably di e en om he employe ’s popula ion g oup may ace ‘ as e’ disc imina ion,
while g oups ha p esen a po en ial employmen issue ha is unknown o unp edic able a
he ime o ec ui men (such as mo he s o young child en o people who ha e p e iously
expe ienced a men al illness) may ace s a is ical disc imina ion (Neuma k, 2016, p.77).
Field expe imen s should ideally be supplemen ed wi h su ey o in e iew e idence on wha
employe s know, wha in o ma ion hey use, wha assump ions hey make, and how his
changes du ing he sea ch, in e iew and job o e p ocess (ibid, p76), in o de o build be e
e idence abou he basis o employe disc imina ion. This could possibly build on he me hods
used ecen ly in Sou h Auckland by he Auckland Co-Design Lab (2016) o in es iga e he
issues and ba ie s pe cei ed by employe s when hey seek o hi e young school lea e s.
Ano he challenge o he expe imen al esea ch is de e mine whe he he si ua ions in which
disc imina ion has been shown o occu a e ep esen a i e o wha happens in he es o he
labou ma ke . Conduc ing s udies in a wide ange o occupa ional and i m se ings may help.
A a b oade le el, T easu y will moni o e hnic di e ences in income and ma e ial
s anda d o li ing as pa o i s wo k on he Li ing S anda ds F amewo k. The Li ing
S anda ds F amewo k aims o achie e highe li ing s anda ds o New Zealande s, and
will in ol e de eloping a se o indica o s ha go beyond economic measu es o include
wellbeing and sus ainable de elopmen . This will include u he analysis o ac o s
associa ed wi h ela i ely good and poo ou comes.
4 S a is ics New Zealand. New Zealand Gene al Social Su ey 2014 in o ma ion elease.
AP 18/03 | S a is ical Analysis o E hnic Wage Gaps in New Zealand
18
Re e ences
Auckland Co-Design Lab (2016) ‘The A i ude Gap Challenge: A Sou h Auckland
Employmen and Skills Challenge’. h ps://www.aucklandco-lab.nz/a i udegap
Blinde , Alan (1973) ‘Wage disc imina ion: Reduced o m and s uc u al es ima es.’ The
Jou nal o Human Resou ces 8: 436-455.
Jackson, B., & Fische , R. (2007). ‘Bicul u alism in employee selec ion o ‘who should ge
he job’? Pe cep ions o Māo i and Pākehā job applican s in a NZ Eu opean s uden
sample.’ New Zealand Jou nal o Psychology, 36(2), 100-108.
K ieble, T. and Gampe le, D., (2016) Paci ic economic ends and snapsho . NZIER epo
o he Minis y o Business, Inno a ion and Employmen .
h ps://nzie .o g.nz/publica ion/paci ic-economic- ends-and-snapsho -2016
McKenzie, S. (1996). S e eo ype a i udes o selec ion in e iewe s and he implica ions
o Māo i o cu en selec ion in e iew p ac ice in New Zealand: A hesis p esen ed in
pa ial ul ilmen o he equi emen s o he deg ee o Mas e o A s in Psychology a
Massey Uni e si y (Doc o al disse a ion, Massey Uni e si y).
Minis y o Business, Inno a ion and Employmen (2017) Māo i in he labou ma ke :
Moni o ing epo , 2017. Published on MBIE websi e: www.mbie.go .nz
Minis y o Educa ion and Minis y o Business, Inno a ion and Employmen (2016)
Educa ion and Skills: Su ey o Adul Skills (PIAAC).
h ps://www.educa ioncoun s.go .nz/ opics/ esea ch/su ey-o -adul -skills
Neuma k, Da id (1988) ‘Employe s’ disc imina o y beha iou and he es ima ion o wage
disc imina ion.’ Jou nal o Human Resou ces, 23.
Neuma k, Da id (2016). ‘Expe imen al esea ch on labo ma ke disc imina ion.’ Na ional
Bu eau o Economic Resea ch wo king pape No. 22022.
Oaxaca, R (1973) ‘Male- emale wage di e en ials in u ban labo ma ke s.’ In e na ional
Economic Re iew, 14: 693-709.
Singe , M., & S. Ede , G. (1989). ‘E ec s o e hnici y, accen , and job s a us on selec ion
decisions.’ In e na ional Jou nal o Psychology, 24(1-5), 13-34.
S a is ics New Zealand. (2016). Gene al Social Su ey.
Wilson, M. G., Gahlou , P., Liu, L., & Mouly, S. (2005). ‘A ose by any o he name: The
e ec o e hnici y and name on access o employmen .’ Uni e si y o Auckland Business
Re iew 7(2).
Wilson, M. G., & Pa ke , P. (2007). ‘The gap be ween immig a ion and employmen : A
policy-cap u ing analysis o e hnici y-d i en selec ion biases.’ New Zealand Jou nal o
Employmen Rela ions (Online), 32(1), 28.