Racial Attitudes And Social Spending--Part 1

by: Paul Rosenberg

Sat Sep 19, 2009 at 14:30

The specific racist attitudes involved are: (1) blaming blacks for their continued lower incomes and lack of wealth, and (2) denying that discrimination plays a dominant role in holding blacks back.  Of course, some will deny that these are racist beliefs, but white supremacists have always insisted on the sole right to define everything. Suffice it to say that both beliefs are race-based, and both are factually false.  If race-based lies that help preserve white privilege aren't racist, then one has to wonder, "What is?"

That said, one can--and should--legitimately distinguish between having some racist attitudes--particularly given their currency in mainstream political discourse--and being a racist.  Especially if one is interested in so identifying such attitudes in order to help get rid of them.

This is particularly true for those who may think that discrimination is a problem, but who minimize its impact.  Indeed, such a mindset among whites is part of the construct of "color-blind racism," as described by Eduardo Bonilla-Silva in Racism without Racists: Color-Blind Racism and the Persistence of Racial Inequality in the United States.  At the same time, such a mindset among individual blacks may be key to individual success, even as it helps preclude broader advances for the black community as a whole.  It seems both accurate and fair to describe this as internalized racism that is nonetheless individually beneficial--precisely the kind of double-bind that has been a recurrent feature of the psychic baggage that blacks have been made to bear since first being kidnapped and brought to America.  While there is a great deal more one can say about how to view such beliefs, the point of this diary is to focus on their political significance, which turns out to be quite considerable.

More than one-third of all Americans hold both these beliefs, while a similar number hold one of the two beliefs.  Only about one quarter reject both.  The following chart--drawn from the concluding section of this diary--shows the widespread impact of such attitudes.  From a battery of seven spending items, support for increased spending on 4 or more items is half as high among those who hold both racist beliefs (30.1%) as it among those who reject both (62.2%):

A further point:  The number saying that we're spending too much on these seven items, net (i.e. those who indicated "too much" on more items than they indicated "too little" on), was four times as large among those who express the racist attitudes (14.6%) compared to those that reject them (3.6%).

On the flip is refresher on the context that this diary originally derived from, followed by a detailed examination of the impacts of these beliefs on nine social spending priorities covered in the General Social Survey. Seven of those come from one sub-sample, and the combined data from that group is examined at the end of the diary.  A followup diary will look at this same data in terms of the differences between the White South--the core of the GOP today--and the rest of the nation.

(1) In Chris's May 19 diary, "The Public Option and The Grand Arc Of American Politics", Chris wrote:

It is not a coincidence that United States public sector social spending stalled at around the same time that the modern conservative coalition came together under Richard Nixon in 1968 and 1972.... While we wanted more social services, we eschewed increasing our public sector social spending over the last 30-35 years because we didn't want that social spending to go to everyone. More specifically, the majority of the country wanted more social services, but the white majority didn't want to pay for social services for racial and ethnic minorities. As such, we continued to increase social spending, but we did so in the private sector, where people had to pay on their own, rather than in the public sector, where people collectively paid for each other.

(2) In my August 9 diary, "Racial Divisions vs. Public Social Spending" I reported on the findings of the 2001 Brookings Institute paper, "Why Doesn't the United States Have a European-Style Welfare State?" by Alesina, Alberto, Edward L. Glaeser, and Bruce Sacerdote, which included two charts showing that social spending decreases as racial/ethnic diversity increases.  The frist chart compared countries around the world (see original diary), and the second compared states:

[click to enlarge in new window]

In this diary, I'll be focusing on the relationship between attitudes towards blacks and support for various forms of social spending. The attitudes concern whether blacks lag in income and economic achievement generally because of internal or external factors (described more fully on the flip.  As an example of the effect we see--in line with what we'd expect from the data in the earlie diary, here are the nationwide results for spending on "welfare" (from the General Social Survey):

A solid majority of those who blame blacks for their lack of economic success think we're spending too much on welfare, but this number shrinks to less than a quarter among those who blame external factors, while a larger number thinks we're spending too little on welfare.

The underlying data for the above chart is as follows:

Spending on welfare is the most racially-charged issue there is.  The term "welfare" has been the subject of demonization for decades.  What's remarkable is that attitudes are not nearly so harsh when questions are asked in different ways.  For example, in the late 1980s, the GSS included a question module that asked about how much assistance should be given to welfare recipients in various different scenarios.  The results were astonishing: 98% of those questioned said that people should get more than they were then receiving--many of them saying two or three times more.  But the standard GSS question about welfare spending primarily reflects the commonplace demonization of welfare.  

The impact of blaming blacks vs. blaming outside forces (explained more fully in the section below) is striking in terms of support for welfare spending. The belief we are spending too much is twice as high among those who blame blacks compared to those who blame outside forces (53.3% vs. 24.6%), while the belief we are spending too little is just half  (15.9% vs. 32.4%).

It's also worth noting that the percentage of people identifying external reasons for black's lower income (14.9%) is less than 1/3 of those blaming internal reasons (47.8%).  These figures vary from questions to question, as not all questions are answered by every respondent, but the range for this question is fairly typical.  These figures will gain more significance in Part 2, when I compare attitudes in White South with the rest of the country.

While the differences are quite extreme in this case, they can also be found on issues you might never expect--such as spending on improving and protecting the environment, as we'll see below.  We also look at spending on Social Security, solving the problems of big cities, improving the conditions of blacks, improving the nation's education system, improving and protecting the nation's health, assistance to the poor and assistance for childcare.

But first, some explanation.

The Basis of "Internal" vs. "External" Attributions

The racial attitudes are derived from the first and last of four GSS questions:

266. On the average (negroes/blacks/African-Americans) have worse jobs, income, and housing than white people. Do you think these differences are:

a. Mainly due to discrimination?
b. Because most (negroes/blacks/African-Americans) have less in-born ability to learn?
c. Because most (negroes/blacks/African-Americans) don't have the chance for  education that it takes to rise out of poverty?
d. Because most (negroes/blacks/African-Americans) just don't have the motivation or willpower to pull themselves up out of poverty?

Very few people answer "yes" to (b), while (c) is potentially ambiguous.  But (a) clearly points to external reasons for lack of black success, while (d) clearly lays the blame on blacks.  Thus, those who say "yes" to (a) and "no" to (d) were classified "all external," those who did the reverse were classified "all internal", and everyone else was classified "mixed."

More Racially-Coded Spending Items

As noted above, attitudes toward "welfare" are highly distorted by demonization, and more concrete questions tend to produce very different, much more sympathetic responses. Here, we take a much simpler approach, we simply eliminate the word "welfare" and substitute "assistance to the poor", which is still perceived as racially coded, but not as harshly:

The 19.2-point difference in those thinking we are spending "too little" between those who blame blacks and those who do not is even larger than the 16.4-point difference for "welfare", but the overall level of support is much higher (68.4% vs. 23.1% for all respondents), so it's a significantly smaller drop in percentage of the support-recorded as "Int/Ext" in the chart above.  The 76.6% figure represents a drop of almost 24% (100%-76.6%=23.4%), compared to a drop of over 50% for welfare. The percentage drop is a good yardstick to use when comparing different items that may have significantly different levels of baseline support.  It allows us to compare how much difference it makes to blame blacks for their situation, regardless of the general level of support for a spending item.

Big cities are also associated with blacks, and the 16-point drop--from 56.4 to 40.4--translates into an almost 30% drop (100%-71.6%=28.4%).  In both these cases--the poor and big cities--it bears mention that there are actually more non-blacks than blacks involved in the target population.  Surprise! Surprise!  Demonization is catching.

But the purest test comes in asking point blank about assistance for blacks:

Here's there's a whopping 37.4-point difference in those saying we're spending too little, which translates into a nearly two-thirds drop (100%-35.5%=64.5%) from 58.1 to 20.6.

On the other side--those saying we're spending "too much", the raw numbers are lower, but the percentage swing--in the opposite direction--is much higher, a drop of more than 80% (100%-18.8%-81.2%) from 28.4% to 5.3%

This is clearly a much bigger difference than we get for assistance to the poor or for big cities.  It confirms the basic premise of this diary, refined from the August 9 predecessor, that having recipients of a different race is a driving factor in decreasing support--provided one holds them responsible for their condition.  Having recipients of the same race dilutes this effect.

We now turn to a range of questions that one would not ordinarily think have much to do with race-spending on health care, child care, education, the environment and Social Security.  We discover that race actually impacts the willingness to spend on all these issues, although it is a very small influence on just one: Social Security.  We consider them in order of declining racist impact.  First up is childcared.

The nearly 1/4 drop (100%-75.4%=24.6%) in support between those who don't blame blacks and those who do puts support for childcare in the same general range as support for big cities and support for the poor.  While this may be surprising-even shocking--to some, it's indicative of how central racism is to welfare state opposition. It's also indicative of how deeply anti-child American political culture is.  It's no surprise that our child poverty rates look more like a Third World nation than they do like Germany or Japan.

Of course this ties into racism, since the child poverty rate is so much higher among minorities-and perhaps more tellingly, minority child poverty is far more visible in the media.  Still, it indicates a level of virulent antipathy sharply at odds with the myth of colorblind America.

If the issue of childcare indicates a racial component that many folks would not think of, thie same can also be said about the environment, since low-income minority communities are far and away the most polluted.  Still, this fact is deeply buried by the American media that it's somewhat surprising to see how much impact racial attitudes have on spending to protect the environment: The drop is almost 20% (100%-81.8%=18.2%).

Surprisingly, education is less affected than spending on the environment.  The drop is a few points lower, roughly 15% (100%-84.7%=15.3%).

The figures on health spending are quite comparable to education-a drop of 14% (100%-86%).  More telling, perhaps, is the fact that those saying we're spending too little show more than a one-third drop (100%-63.5%=36.5%) in the opposite direction.

Racial impacts do not entirely disappear with Social Security, but they are dramatically lower than with all other items in our sample.  The 4% (100%-96%) drop in support is less than a third of the drop in health care spending support.  Still, when one looks at the more sensitive indicator-those saying we're spending "too much", the drop in the other direction is over 12 1/2 %--one in eight (100%-87.3%=12.7%).  This drop takes place within is a small segment of the population, but it's a potent reminder that race never really goes away as a factor in attitudes about social spending.

Combined 7-Item Index

Of the nine items we're looked at, seven of them come from the same sub-sample, and can be combined into a comprehensive index.  For purposes of greater insight, we divide them into four categories-a high-spending support group that supports increased spending for 4 or more items, another spending support group that suppors increased spending for 1-3 items, a "no-change" group that supports keeping spending as it is, and a "cut spending" group that thinks we're spending too much on one or more programs.  All these descriptions are based on net results.  If someone thinks we're spending too little on 6 items, but too much on one item, that's support for more spending on 5 items net.

As described toward the beginning of this diary, this combined index demonstrates a sharp difference between those who hold all external explanations for blacks' low income and those who hold all internal explanations. Support for increased spending on 4 or more items is half as high among those who hold both internal, racist explanations (30.1%) compared to those who hold as it among those who hold external explanations (62.2%).  At the same time, the number saying that we're spending too much on these seven items, net, was four times as large among those who express the racist attitudes (14.6%) compared to those that reject them (3.6%).

While the measure of support for 4-7 items may be largely attributable to the more racially-charged spending items, this seems significantly less plausible as an explanation for the even larger discrepancy among those saying we're spending "too much".  This comports with the generally pattern we've observed above, which shows that these racial attitudes significantly affect levels of spending support for all items, except for Social Security.

In short, although the race of the recipients--and the salience of this recognition (accurate or not!)--is a prime factor in determining the intensity of opposition, the attitudes themselves have an impact that carries over into virtually the entire realm of social spending issues.