The Metropolitan Council’s Disparities Report: The Limitations of Racial Explanations

The Metropolitan Council released a report this week statistically documenting, once again, disparities between “persons of color and White, non-Latinos” in regards to poverty rates, home ownership, employment, and levels of education in the 16-county Twin Cities region.  All four sets of disparities are among the biggest in the country.  The report relies on regression analysis, a research method I’m not nearly expert in, but about which I know two key things in this instance.

The first – this is true in all regression analyses, not just those pertaining race – is that the pertinence and comprehensiveness of the data researchers pump into equations have everything to do with the explanatory power of the data which emerge.

The second key factor has to do with studies dealing specifically with race.  Talking un-mathematically, my sense is that when regression analyses fail to map out or explain enough about a problem, a fair number of researchers assume that the unexplained portions have something to do with race, and probably not benignly.  More specifically, that any disparities between and among groups that can’t be attributed to differences in nonracial characteristics must be the results of a kind of racialism that may not be all that far away from real-live racism.   The Met Council report more than leaves this impression. I don’t share such assumptions.

This is definitely not to say racism doesn’t exist.  Of course it does.  One clear example of institutional racism has to do with homeownership, one of the characteristics the Met Council report focuses on.

For those of you who have owned one or more homes, please recall when you bought your first one.  Might you have gotten help on the down payment from your parents or in-laws?  If yes, might they have been able to make such a gift or loan, in part, because (likely being white) they never faced real discrimination in seeking jobs and building careers?  Or might they have been able to help because they had been left a respectable sum(s) of money by a deceased parent(s) or grandparent(s), who also had never contended with bigotry.  Or if they had, chances are it wasn’t as severe as that suffered by African Americans and other people of color at the time.  In other words, lower homeownership rates by African Americans currently have something to do – I wouldn’t guess how much – with discrimination faced by their forebears previously, with some of that shortchanging still at play.

But a caveat again.  This is not to say that racism, whatever kind it may be, is as determinative of present disparities in incomes, educational achievement and other aspects of life as is often claimed or implied.  Here is a relevant and central passage in the Met Council report.  It deserves to be quoted in full.

Some question whether these disparities are, in fact, based on race at all.  This line of thinking accepts that economic outcomes are worse for Black residents but rather than seeing race as the distinguishing characteristic, point to underlying demographics as the main drivers of these inequalities.  For example, younger people (of any race or ethnicity) may be less likely to be employed, show lower overall income, and are less likely to own their home.  If Black residents tend to be younger, today’s racial disparities in economic outcomes may be more the result of age than race.  Said another way, if the region’s Black and White residents had the same demographic profile, our region’s racial disparities would be drastically reduced.

Then in bold:  “However, our analysis shows that underlying demographic differences cannot explain away our region’s disparities in employment, income, and home ownership between Black and White residents.”   The implication being that something racially untoward is at work overtime.

And later:  “Even when demographic differences between the region’s White and Black residents are taken into account, large disparities in employment, income, and homeownership would remain.  This again suggests race – or factors closely aligned with race – are involved.”  (Emphasis in the original.)

Demographic Characteristics

What kind of demographic characteristics does the report use?   One of them is “level of education.”  As with all of the characteristics, the report properly acknowledges that “our analysis does not include every factor that could effect (sic) employment, income, or homeownership.”  I appreciate, moreover, how the Census data they used are not broken down the way I would like.  (No one in Washington ever asked.)  But the fact, nevertheless, remains that focusing on how many years of education a person has (“level of education”) while completely ignoring how much a person, regardless of race or ethnicity, actually has learned, is acutely misleading.  This is another way of saying that achievement, not just attainment is important, as research clearly shows that reading skills and math skills play a significant role in eventual career success and income.

So how do blacks and whites in the Twin Cities region compare when it comes to reading and math skills, not just education levels?

According 2015 numbers from the National Assessment of Educational Progress (NAEP), while 51 percent of white Minnesota eighth graders were either “Proficient” or “Advanced” in reading, the corresponding results were 16 percent for black students.  As for math, while 72 percent of white Minnesota eighth graders were either “Proficient” or “Advanced,” the corresponding number was again 16 percent for black students.   I know of no evidence that enormous gaps in essential skills like these can be reduced in any adequate and consistent way for a whole group in the quick years between adolescence and adulthood.

Or take family structure.  The Met Council report uses as a demographic characteristic “share of parents with child(ren) under age 6.”  This is an interesting measure, though it ignores completely how many little (and older) kids have been living with how many parents under same roofs.

Last I checked, 84 percent of all black babies born in Hennepin County came into this life to unmarried mothers.  The corresponding proportion for white babies was 17 percent.  Gulfs like these between the two groups have been akin for a long time now.  Given what we absolutely know about how young people growing up with two parents do much better (always on average) by every measure one can think of, black and white young people in the Twin Cities metropolitan region simply have not been entering adulthood equally equipped to get good jobs, make good money, and buy nice homes.

Many contend that differences like these reflect the reach of racism, and as with a lot of things in this life, I agree – to a degree.  But vast numbers of students doing poorly, for instance, has more to do with their not working hard than with racism of any kind, be the poison past or present.  As for massive nonmarital birth rates, a purposeful dismissal of marriage by many (not just blacks) has had as much to do with such rates as anything racially connected.  More, actually.

Also unremarked upon in the report are seriously disparate incarceration rates and their long-lasting effects in regards to jobs, housing, and a host of other things.  Many argue, of course, that such differences are a prime example of a racist criminal justice and broader system.  Much closer to the mark, I’m afraid, they’re a reflection of seriously disparate rates of criminal behavior, and not just when it comes to drugs.

Yes and without question, disparities of various sorts in Minneapolis, St. Paul and the surrounding region are worrisome and sad; I don’t discount their seriousness or danger for a moment.  But they are much more complex in origin, as well as in what it will take to reduce them, than is suggested by a Metropolitan Council report that assumes race and possibly racism are at the root of what it otherwise can’t or refuses to explain.

Mitch Pearlstein is Founder & American Experiment Senior Fellow.