Debate  |  Volume 157

Implicit Race Bias and the 2008 Presidential Election: Much Ado About Nothing?

Debate - Implicit Race Bias and the 2008 Presidential Election: <i>Much Ado About Nothing</i>?

Dr. Gregory Parks and Professor Jeffrey Rachlinksi debate Professor Richard Epstein on the implications of President Obama’s election on race-based policies such as affirmative action and antidiscrimination laws.  Dismissing claims that a “post-racial America” has arrived, Parks and Rachlinski argue that the presidential campaign simply demonstrated that modern racism “operates not as an absolute barrier, but as a kind of tax on members of racial minorities.” Because a typical minority job applicant does not have access to the “enormous resources” available to Obama, affirmative action and antidiscrimination laws are still necessary. Epstein, on the other hand, argues that while Parks’ and Rachlinski’s analysis of the campaign is flawed, the debate itself is a “sideshow.” Instead, “any private individual or firm can hire a person for good reason, bad reason, or no reason at all” and the market, not public force, should end unsavory hiring practices.

Continue to full article.

Opening Statement — Gregory S. Parks & Jeffrey J. Rachlinski

Barack Obama, Implicit Bias, and the 2008 Election

Law Clerk, District of Columbia Court of Appeals; Professor of Law, Cornell Law School.

The election of Barack Obama as the forty-fourth President of the United States suggests that the United States has made great strides with regards to race.  The blogs and the pundits all assert that Obama’s win means that we now live in a “post-racial America.”  But is it accurate to suggest that race no longer significantly influences how Americans evaluate each other?  Does Obama’s victory suggest that affirmative action and antidiscrimination protections are no longer necessary?  We think not.  Ironically, rather than marking the dawn of a post-racial America, Senator Obama’s candidacy reveals how deeply race affects judgment.

With notable exceptions, conscious or explicit racism was not part of the 2008 campaign.  But social psychologists argue that unconscious or implicit biases have a powerful effect on how people evaluate each other.  Much of this work is documented at  Implicit racial bias is widespread; the vast majority of adult Americans, for example, more closely associate white faces with positive imagery and black faces with negative imagery.  Implicit bias induces dangerous assumptions; white Americans more readily associate black Americans with weapons and white Americans with tools than the opposite pairing.  Implicit bias is crude and ugly; white Americans associate apes with black Americans.  White adults also more readily associate the concept of American with being white, and showing white adults subliminal images of the American flag increases their antiblack bias.  These findings particularly show the contrast between explicit beliefs and unconscious associations:  African Americans are obviously American, but they seem less so to most adult white brains.

Furthermore, implicit biases influence how people evaluate others.  White interviewers who harbor strong anti-black unconscious biases make less eye contact with black job applicants, exhibit hostile body language, and report that these interviews are uncomfortable.  White interviewers who do not harbor such biases do not exhibit the same effects.  And implicit biases have a documented neurobiological component.  Those who evidence a strong association of white with good and black with bad use a part of their brain associated with the fear response (the amygdala) to process black faces.  And at least one study also shows that unconscious racial biases can affect how people vote.

But did this landscape of unconscious bias affect the course of the 2008 election?  Researchers have struggled to demonstrate the influence of unconscious biases in the real world.  Ironically, several aspects of the election of the first black President of the United States provide that demonstration.

First, throughout the campaign, criticisms abounded that Obama was unpatriotic or insufficiently American.  These attacks began early, when a news story that he failed to place his hand over his heart during the singing of the national anthem at an Iowa fair gained traction.  They continued as his detractors complained that he declined to wear an American flag pin on his lapel.  The absence of a flag on Obama’s lapel was a small wonder when he was a little-known candidate, given the ability of American imagery to prompt negative associations among white Americans.  Associations between being black and being foreign helped make Obama vulnerable to such charges.

So deep is the connection between black and foreign in many Americans’ minds that one early study, conducted in the spring and fall of 2007, showed not only that that voters more closely associated Hillary Clinton with American imagery than Barack Obama, they more closely associated Tony Blair with American imagery than Barack Obama.  Thierry Devos et al., Is Barack Obama American Enough to Be the Next President?  The Role of Ethnicity and National Identity in American Politics, available at  In addition to conflating President Obama’s race with a lack of authentic Americanness, critics also alluded to his middle name, “Hussein,” or alleged that he was Muslim or an Arab as other indicators that he was, as Pat Buchanan often termed, “exotic.”  It was perhaps no surprise that Senator McCain’s campaign theme was “Country First,” which takes fair advantage of McCain’s war record, but also implicates that Obama fails to put country first in the same way.  Unconscious racial associations between black and foreign helped make McCain’s campaign theme seem to be a desirable strategy.

Second, the campaign was not entirely free of explicit racial references, many mimicking the studies of associations between black people and apes.  A white Georgia bar-and-grill owner began selling t-shirts at his establishment depicting the image of Curious George, a cartoon monkey, with the slogan “Obama in ’08.”  In June, a Utah company began making a sock monkey (doll) of Obama.  During the fall, a man at a McCain rally carried a monkey doll with an Obama sticker wrapped around its head.  At various points, both Democrats and Republicans used milder racial slurs to refer to Obama.  Senator Clinton surrogate, Andrew Cuomo, used the phrase “shuck and jive” in an indirect reference to Obama’s campaign strategy.  Republican congressman Tom Davis, in discussing how Senator Obama would have difficulty handling the immigration debate, described this issue as a “tar baby.”  Even when charging Obama with being an “elitist”—a charge that would seem to be inconsistent with stereotypes about black Americans—many of his detractors used the more racially tinged word “uppity.”

Third, the primary elections exhibited what has been called the Bradley Effect—the tendency of polls to overestimate support for a black candidate in an election against a white candidate.  See Anthony G. Greenwald & Bethany Albertson, Tracking the Race Factor, Pew Research Center, Mar. 14, 2008, (providing the source of the data reported here).  Although commentators denied that the Bradley Effect occurred, the pattern that emerged during the spring primaries was clear.  States with small percentages of black voters that held primaries (California, Massachusetts, New Hampshire, and Rhode Island) exhibited the Bradley Effect.  By contrast, polls were basically accurate in states with black populations near the national black population of 12.3%:  Illinois, Indiana, Ohio, Tennessee, and Texas.  A reverse Bradley Effect—whereby pollsters underestimate support for Senator Obama—occurred in Alabama, Georgia, Mississippi, North Carolina, South Carolina, and Virginia, all of which are 19% or more black.  Of the eighteen states with open primaries and available data, only Wisconsin was inconsistent with this trend.

The pattern of polling error suggests strongly that voters either lied to pollsters or changed their minds at the last minute.  White voters flinched at the last moment, unwilling to pull the lever in favor of the black candidate.  Black voters, did the opposite; finding themselves unable to resist the prospect of voting for a viable black candidate when the time came to cast their ballots (or turned up at polls in numbers greater than expected).  That this pattern did not persist in the fall is an interesting and promising development.  But no pollster who assesses the spring primary data carefully will advise a future black candidate to ignore the possibility.

Fourth, the election was marked by deeply racially stratified voting.  Obama won among black voters by 91 percentage points; among Latinos by 36 points; among Asians by 27 points; but he lost among white voters by 12 points.  ABC News, How They Voted:  Exit Poll Full Results, (last visited Jan. 31, 2009).  The spring Democratic Party primaries (which obviously control for political party preferences) were even more stratified.  Exit polls showed that Obama never fared better among white voters than black voters.  See, e.g.,, Exit Polls, (last visited Jan. 31, 2009); Washington Post, Entrance and Exit Polls,  (last visited Jan. 31, 2009).  Although he won overwhelmingly among black voters everywhere, only in Iowa, Illinois, Vermont, Indiana, and North Carolina did he win among white voters.  After the news reports of his former pastor, Reverend Wright, surfaced, he performed even worse among white voters.  He lost white voters in Pennsylvania, Ohio, and Kentucky by 26, 30, and 49 points, respectively.  All of this occurred even as less than 10% of voters indicted to pollsters that race influenced their vote, suggesting that voters might not understand their own motives well.

The campaign was thus a reflection of how contemporary racism works.  Modern racism does not produce an overt smoking gun marking its influence; one has to look fairly carefully to find its influence.  It operates not as an absolute barrier, but as a kind of tax on members of racial minorities.  It facilitates certain negative assumptions through an invisible influence.  McCain, after all, did not face a fair fight.  Obama’s success came in large measure from raising enormously more money than McCain and from the specter of an unpopular Republican President presiding over a horrific financial crisis that induced great demand for the kind of government intervention more closely associated with Democrats.  And of course, implicit and explicit biases against older Americans’ abilities are common as well.

Obama navigated the racial waters well.  He spent a great deal of time and money creating positive imagery to combat the negative associations that are so common.  For most of the spring campaign, his message was one of raw, positive optimism, unadorned with details.  Wisely so, as studies of implicit racial bias suggest that details concerning resumes and qualifications are influenced by unconscious associations.  Once Obama created his own set of associations, he was rarely seen without a bevy of American flags behind him.  Although campaign leaders now report that they only rarely discussed race, they ran a campaign well-suited to combating unconscious bias, just as McCain ran one well-suited to taking advantage of it.

But, of course, Obama had an army of strategists and pollsters backing his lengthy job interview with America.  The ordinary black job applicant faces the same racial environment without such assistance.  Affirmative action and antidiscrimination laws can hardly be said to be unnecessary in a world in which the enormous resources Obama had available are necessary to combat bias.  The 2008 campaign thus teaches us that America is not so virulently racist as to reject a black applicant for a serious position outright.  The nature of the campaign, however, shows that race continues to play a complex and profound role in how Americans judge each other.  The post-racial America may be on its way, but has yet to arrive.

Rebuttal — Richard A. Epstein

The Good News on Race Relations

James Parker Hall Distinguished Service Professor of Law, University of Chicago Law School.

Gregory Parks and Jeffrey Rachlinski have written a highly provocative, but exceedingly lop-sided, essay with the ominous title, Barack Obama, Implicit Bias and the 2008 Election.  In it, they offer some grudging acknowledgement that the election of an African American President marks something of a milestone in the history of race relations in the United States.  Obama received, for the record, more popular votes than any other candidate who has ever run for high office in the United States.  He won by a respectable margin of seven percentage points, 53 to 46 and his 66 million–plus votes gave him an edge of about 8.5 million over John McCain.  Obama was able to attract and hold deeply committed supporters of all races and creeds.  He raised, month in and month out, huge sums of money online.  He was able to call on an army of volunteers who scoured the landscape in close states, doing everything to secure his victory.  The mood at his election-night celebration in Grant Park, Chicago, can only be described as euphoric.  His inauguration was only slightly less so.  It did not take a deep statistical examination of the crowds at either event to realize that they were a cross-section of the American population by race.  And it did not take deep psychological analysis to see the near worshipful looks of happiness and pride on the faces of everyone in attendance.  Barack Obama stands as an iconic figure.

As a long-time resident of Hyde Park (who was in New York during the entire campaign), I can name many of my well-to-do white friends who took to the highways and byways to campaign for Obama in the strong conviction that he would present a public face for the United States that would allow us to regain the affection and respect of people all around the globe.  The Obama adoration that runs through Europe, Latin America, and Asia is not a subtle form of implicit or unconscious racism.  It is exactly the opposite.  It is an explicit and conscious affirmation that Barack Obama has the personal qualities to lead the United States back to its former glory.

In the face of all this nonstop adulation, I find it odd, almost incredible, that Parks and Rachlinski think that the appropriate way to examine the mood of the nation is to offer a list of the worst racial episodes of the past campaign.  In doing so, they commit multiple mistakes and omissions.  Here are three:  First, they ignore all the ugly but unsuccessful efforts to link Obama to Bill Ayres in order to paint Obama as a man who consorts with terrorists.  Ayres, of course, is white.  Second, they ignore the widespread praise that Obama earned for his speech defending himself from the charge of being too cozy with the Reverend Jeremiah Wright, from whom he eventually—and rightly—distanced himself.  Neither effort to bait or inflame the American electorate against Obama had any lasting effect.  Third, and most egregious, they ignore the literally millions of kind and generous actions by people of all races that propelled Obama to his hard-fought and well-earned victory in the last election.

Indeed, in explaining why Obama won, I would put grace under fire at the top of the list.  McCain was clinging to a small lead in the polls until the financial meltdown hit Wall Street in late September and early October.  McCain’s response was to go into panic mode, and to suggest that the first debate be canceled so that he and Obama could return to Washington to address the problems of the nation.  Obama then got off the best line of the campaign when he said, in response, that the President of the United States ought to be able to multitask.  This perfect putdown of McCain showed how, without uttering McCain’s name, Obama could portray his opponent as frazzled and panicky, while keeping his dignity and cool.

Nor was it just a flash in the pan.  Obama kept that image up all through the debates.  His demeanor, especially when he was not speaking, was flawless.  He did not gesticulate or fidget, but kept a calmly skeptical gaze on McCain as the over-the-hill Republican lurched back and forth on the stage, desperately trying to score points.  The contrast between the dignified Democrat and the rambunctious Republican was not lost on television audiences.  The polls reported that Obama won the debates hands down.

The concerns that Parks and Rachlinski raise about the treatment of Obama are odder still in light of their puzzling silence on the vicious treatment directed nonstop toward George W. Bush, John McCain, and especially Sarah Palin.  In this regard, I am not of course referring to the dead-on impersonations of Palin that catapulted Tina Fey to fame on Saturday Night Live.  Rather, I am thinking of the posters depicting Palin as a vampire that greeted me each day as I walked down West 15th Street in New York City.  In addition, the endless abuse and epithets hurled toward George Bush for his Iraqi and domestic policies revealed a hatred that would have generated an instant outcry if directed toward Obama, which thankfully it was not.  Whatever implicit resentments some people harbored toward Obama, his race insulated him from the kind of ugly and explicit charges routinely hurled at white Republican candidates.

In light of these complex political cross currents, we should think long and hard before attributing much, if any, weight to the so-called Bradley Effect, which posits that many white Americans are prepared to say that they will vote for a black candidate, but are unable to pull that lever in the polling booth.  There is of course much debate over whether the Bradley Effect actually played a role in the defeat of Mayor Tom Bradley of Los Angeles by George Deukmejian, his Republican rival in the hotly contested 1982 California gubernatorial election.  The alternative explanation was that the polls stopped too soon to pick up the Deukmejian surge, which drove home the substantive differences between the two candidates on budgetary and economic issues.  Likewise, the polls also missed the effective Deukmejian campaign to collect absentee ballots.  Lots of people fretted about the Bradley Effect after Hillary Clinton thumped Obama in the California primary, but Obama easily carried California against McCain in the general election, 61% to 37%, with a plurality of nearly 3 million votes.  The far more likely explanation for these numbers is this:  some people, white or black, may have an implicit racial bias, but what really counts is that they have no desire to defend their bias once it is called to their attention.  If anything, their conscious actions may well overcorrect for their implicit preferences, which could work to the advantage of candidates like Obama.  Indeed, African American candidates everywhere have, in recent years, been consistently able to make large inroads among white voters while white candidates, especially Republicans, find it notoriously difficult to attract black voters at all.  One very obvious explanation for these trends is that the Democratic platform, with its strong social justice component, appeals to African American voters more than it does to white voters.

Parks and Rachlinski also misfire by failing to place the question of implicit racial bias in its larger social context.  For starters, the possibility of implicit sexism with respect to both Hillary Clinton and Sarah Palin may have struck a more responsive chord in the last election.  Yet even that is small potatoes compared to the truly ugly campaigns relating to Proposition 8 in California, over gay marriages, where both sides revealed all too many explicit biases, reflecting credit on no one.  And, if I were to look for other hot-button issues, I would turn first to immigration, where the anti-outsider campaigns often have an explicit ugliness that was wholly absent in the presidential election.

In sum, the evidence from the past election deserves a much more positive interpretation than Parks and Rachlinski give it.  But what about the political agenda that motivated their remarks in the first instance—the strong boost for affirmative action and antidiscrimination protections?  Their obvious fear is that Americans will let down their guard on these fronts now that Obama is about to take over the White House.  My reaction is to disagree with them on both counts, but for different reasons.

Long before the current election, I wrote a book called Forbidden Grounds:  The Case Against Employment Discrimination Laws (1992).  The gist of my argument was that competition in labor markets affords workers far greater protection against racial discrimination than any government program, which could easily end up, through public enforcement actions, creating more discrimination than it eliminates.  Seventeen years later I see no reason to change my negative judgment on these laws, which are both ineffective and costly to enforce.

The key analytical point is the stark contrast between discrimination and the use of force.  Aggression exposes any individual to the tender mercies of the person who likes him or her the least.  But even in a market that is rife with discrimination, the economic fortunes of members of the disfavored groups are determined by the attitude of those who like them the most.  The most that people with racial hatreds can do is to refuse to hire people whom they hate.  In this environment, the implicit biases of some do not matter much.  What counts are the favorable attitudes of others.

This approach has powerful implications for the treatment of affirmative action.  The modern cast of mind demands special justification for private firms and institutions to engage in affirmative action programs.  The standard approach is to give a harsh indictment of American racial practices to explain the deviations from the color-blind antidiscrimination laws that now sit on the statute books.  With each passing year, tales of overt discrimination a generation or more ago supply ever weaker justifications for today’s affirmative action programs.  Appeals to racial diversity do not quite pick up the slack because many different groups can claim a part of the new affirmative action programs.

To a classical liberal/libertarian like myself, these debates are a sideshow.  The key point is that any private individual or firm can hire a person for good reason, bad reason, or no reason at all.  Hence they can engage in affirmative action programs even if they reject every single claim that Parks and Rachlinski make about the dire effects of implicit bias on the white-American psyche.  And that is much the better way to go about our national business.  It is a mistake to require private parties who wish to engage in affirmative action to issue a harsh public denunciation of past practices of dominant social elites to justify their action.  There is absolutely no need at the time of one of America’s great racial achievements to urge, yet again, that more public and private action is needed to exorcise our innermost demons.  It should be quite enough to let people who want to start affirmative action programs do so.  And if other organizations want to start white-only programs, let them do so. Freedom of association is the operative principle.  We are strong enough as a nation not to treat private offense to the associational preferences of others as a reason to shut them down by public force.  Better that they should just die on the vine because people have consciously decided, one at a time, not to do business with them.  Or do Parks and Rachlinski really believe that various hate-groups will take over the nation if the antidiscrimination laws are repealed and private affirmative action is allowed?  I am eager to hear their response.

Next Debate 

Democracy, Political Ignorance, and Constitutional Reform

Ilya Somin and Sanford Levinson