Bill,
I can't give you specific references to where Wayne Simpson and
Derek Hum's data is/will be published given the usual summer
hiatus plus the fact that the PanAm games are here and have
completely disrupted all activity at the university (which is the
athletes village) but, from what you say, their data/conclusions
differ from those you report. They use SLID datat to which they
had special access to non-published micro data as 'guests' of
Stats Canada. Both are highly skilled and reputable
econometricians and their conclusions refute those based on 1991
census data. Given the quality of the data I would tend to accept
the SLID data over the Census data for this kind of analysis.
However, I hardly qualify as an expert in this field so I have to
concede to what I know. Since I have discussed the method and
data with Derek and Wayne, I have to trust their judgement.
Therefore, I would support their findings which (given my limited
econometric expertise seems impecable to me) suggests that
there is little/no visible minority discrimination to Canadian born in
Canda with the exception of black males -- with the proviso that the
data does not allow for an analysis of aboriginals -- of whom I think
we all agree there is discrimination against.
I will send you references to their work when normality returns to
the U of Manitoba.
Paul
Paul Phillips,
Economics,
University of Manitoba
Date sent: Thu, 15 Jul 1999 14:11:49 -0700
To: [EMAIL PROTECTED]
From: Bill Burgess <[EMAIL PROTECTED]>
Subject: [PEN-L:9209] Re: income by 'race'
Send reply to: [EMAIL PROTECTED]
> A week or two ago I disagreed with Rod Hay's claims that there "appears to
> be no widespread, identifiable systemic racism, at present" in Canada
> (except against Natives), and that there are no economic studies which
> "find any significance for 'self-identified race' in determining income".
>
> I had results by Shapiro and Stelcner on hand which showed that unlingual
> francophone men earned 9% less than unilingual anglophone men in Quebec in
> 1991 (8% when adjusted for education), and I cited them as evidence to the
> contrary. Rod dismissed this study for being based on language rather than
> 'race', though I still don't see why.
>
> Paul Phillips described a study by colleagues Simpson and Hum as providing
>
> >fairly strong evidence against the existence of *systematic income
> discrimination* based on visible 'racial' physical characteristics.
> It found that
> >With the exception of black males, there was no negative income
> >discrimination against Canadian born visible minorities in Canada.
> >Indeed, there was some support for positive returns to Asian and
> >Indian (from India) subgroups.
> >However, this was not the case for visible minority immigrants
> >which, as a group, received lower incomes -- but only for the first
> >generation.
> (Paul noted that > Unfortunately, the SLID data does not separate out
> aboriginals...However, I have little doubt that if it were possible, the
> results would show discrimination against our native population.)
>
> I have been on vacation, and if Rod has posted the studies that show 'race'
> is not a significant factor in determining income, I missed them. However,
> I did come across a recent study by Pendakur and Pendakur [P&P], who write
> that in the last 5 years a "surge of research" has found "earnings and wage
> differentials among ethnic groups that cannot be attributed to differences
> in observable characteristics such as age and education. Although suitable
> cautious, these authors conclude that discrimination may play a negative
> role for some ethnic groups" ("The color of money: earning differentials
> among ethnic groups in Canada" in the August '98 _Canadian Journal of
> Economics_).
>
> P&P argue their own study extends and strengthens the evidence of
> discimination. They distinguish between Canadian-born and immigrant
> workers, men and women, within the various white and visible-minority
> categories, and 'control' for major characteristics like education and
> labour market experience.
>
> Using 1991 census data, P&P find that "conditional on observable
> characteristics, Canadian-born visible-minority men face an earnings gap of
> 8% and Aboriginal men a gap of 13%, in comparison with Canadian white men."
> Canadian-born British-origin men 'earned' 17% more than Canadian-born Black
> men, 13% more than Canadian-born Chinese men, 10% more than Canadian-born
> Balkan and Greek men, and 23% more than Aboriginal (single-origin) men.
>
> In the case of immigrant-origin men, visible-minority men earned 16% less
> less than Canadian-born men, while immigrant white men earned 2% less.
> Within the immigrant men category there was substantial variation: no
> substantial earnings penalty for Northern and central European immigrant
> men, but Black immigrant men earned 22.2% less than Canadian-born
> British-origin men, Latin American and Philipino immigrant men at least 20%
> less, immigrant Vietnamese and West Asian men at least 18% less, Greek
> immigrant men 17% less, South Asian and Arab immigrant men at least 14%
> less, and Chinese immigrant men 1% less.
>
> The differences are less in the case of Canadian-born women. P&P did not
> find any earning gap between Canadian-born visible-minority and white
> women. However, Canadian-born Aboriginal and Greek-origin women do face
> large and significant gaps compared to Canadian-born British-origin women.
> Among immigrant women, Black, Vietnamese, and West Asian women face earning
> differentials of more than 10% compared to Canadian-born British-origin
> women, but most other immigrant women do not.
>
> PP's 'suitably cautious' conclusion is that "the 8.2 per cent earnings
> differential we found between Canadian-born white men and Canadian-born
> visible minority men is very important. Unlike the gap between immigrant
> and Canadian-born workers, which may erode over time as immigrants
> assimilate into Canadian labour markets, we are not as hopeful that the
> earnings gaps found within the Canadian-born population will disappear over
> time. Further, although some earnings differences between immigrants and
> the Canadian born may be due to preferences, we believe that Canadian-born
> white workers and Canadian-born visible-minority workers have similar
> preferences, so that earnings gaps are primarily due to differential
> opportunities. Thus, these earnings gaps found among Canadian-born ethnic
> groups suggest that economic discrimination may play an important role in
> Canadian labour markets."
>
> I'm pretty weak on regression analysis, but this does seem to be a thorough
> study. They ran separate regressions for metropolitan and non-metro
> residents. The 'independent' variables included full/part time labour
> status, weeks worked, province, household type, occupation, industry,
> schooling, and knowledge of English and French. Labour market experience is
> counted by years since completion of school or since immigration to Canada.
> They test for the effect of educational qualifications being from outside
> Canada and find this makes little difference. (An immigrant
> visible-minority man who completed his education in Canada may expect to
> earn 16.2% less than a Canadian-born white man, even though both have
> Canadian education). One dissapointment is that it does not address workers
> who are totally unemployed.
>
> I think this study reinforces my claim that race is a significant factor in
> determining income in Canada, but I'd like to hear any opinions on its
> validity, e.g. as compared to the Simpson and Hum study Paul reported on
> (is it easily available?), or other studies. Are there comparable results
> in the U.S. or elsewhere?
>
> I think Paul and Rod argue that affirmative action divides the working
> class while I think it is necessary to unite workers, but in any case it is
> important to establish the scale of earning differentials in Canada. Is
> this really the "kinder and more gentle" country our ruling class likes to
> talk about?
>
> Bill Burgess
>
[PEN-L:9223] Re: Re: income by 'race'
ts99u-1.cc.umanitoba.ca [130.179.154.224] Thu, 15 Jul 1999 21:38:43 -0500
