The Joy of Economics:  Making Sense out of Life
 Robert J. Stonebraker, Winthrop University
 

Discrimination:  It Isn't Easy Being Green

             It isn't easy being green.
                                              ...Kermit

            The Muppet Movie, one of history's great cinematic achievements, opens in a deep swamp.  The camera pans to Kermit the Frog perched on a lily pad strumming his guitar and singing:

                        It isn't easy being green;

                        Having to spend each day the color of the leaves.

                        When I think it could be nicer being red, or yellow or gold

                                    or something much more colorful like that.

            Kermit's lament does turn upbeat.  He ends with:

                        I am green and it'll do fine, It's beautiful!

                        And I think it's what I want to be.

            Kermit is correct; it isn't easy being green.  Although he dreams of being a famous Hollywood producer, Kermit's film adversaries refuse to take his ambitions seriously.  After all, frogs belong skewered in a frying pan, not sitting behind a camera.  Kermit does persevere.  With luck and considerable help from a cast of mildly maniacal friends, Kermit sheds his stereotypical frog mold and breaks into Hollywood. 

            Alas, other stereotypes are not so easily shattered.  It isn't easy being green.  And it isn't easy being African-American or Hispanic or female.  For many Americans, equal opportunity is still more dream than reality.

Job discrimination     

            The traditional U.S. response to perceived inequities is to pass a law.  So we have.  We have passed a long series of civil rights and anti-discrimination statutes over the last century.  Discrimination in the workplace is illegal and no one would ever violate the law.  Right?  In these days of affirmative action and preferential hiring policies, minorities enjoy equal access to the labor market.  Right?

            The traditional answer, at least by many middle-aged white males with secure, well-paying jobs (like me), has been: "yes".  Hard evidence answers: "no".  Economists Genevieve Kenney and Douglas Wissoker have studied the extent of job discrimination faced by young Hispanic males.¹  They culled through a list of volunteers and formed two-man teams, each with one Hispanic and one Anglo male.  The team members were carefully matched with respect to age, size, amount of education, job history, and personality.  Team members differed only in race.  Once the teams were selected, members were sent out to apply for a series of low-skill, entry-level jobs advertised in Chicago and San Diego newspapers.  Team members applied for a total of 360 employment positions.

            The results were striking.  Applying for the same positions less than one hour apart, Anglo team members were 4.4 percent more successful in being allowed to fill out application forms, 30.0 percent more successful in getting interviews, and 53.6 percent more successful in being offered employment (see Table I)

Table I

            Success Rates In:                              Anglos             Hispanics        % Difference

            Being allowed to complete

                application forms                             95%                   91%                             4.4%

            Obtaining interview                              65%                   50%                           30.0%

            Getting job offer                                  43%                   28%                           53.6%

            Although Hispanic applicants were less successful across the board, the amount of discrimination varied depending upon the type of neighborhood or firm or interviewer involved.  As we might expect, the apparent discrimination was more severe in white and high-income neighborhoods as opposed to Hispanic and low-income neighborhoods.   Interestingly, discrimination also seemed more pronounced when candidates were interviewed by males rather than by females (are men more prone to bias?) and when they were seeking jobs in local rather than national firms (might national firms have more to lose if they are found running afoul of discrimination laws?).

            This does not necessarily mean that the interviewers and employers are biased.  Perhaps employers perceive their customers to be biased.  If officials at Burger King or K-Mart suspect that their customers prefer to deal with Anglos rather than Hispanics, their profit motive would push them to discriminate against Hispanics, even if they themselves were racially and culturally ambivalent.  Profit-maximizing employers in a competitive environment must hire with customer preferences in mind, not their own.

            We can check this out.  If perceived customer bias is the root cause, the observed discrimination should vary with the amount of customer contact.  If my customers are biased, it might be profitable to grant prejudicial favor to Anglos for front-door cash register positions.  What about the man working alone in the basement counting inventory after hours?  If an employee never sees a paying customer, customer bias should be immaterial.  No dice.  Customer interaction is irrelevant.  The observed discrimination in jobs with little or no customer interaction was just as great as that with jobs involving significant interaction.  Employer bias is the apparent culprit.

What about names?

            In a more recent study, economists Marianne Bertrand and Sendhil Mullainathan studied the impact of applicant names.²  Although many names are clearly race-neutral, others are strongly identified with a particular race.  For example, research shows that most people assume that females with names such as Allison, Jill and Emily are White, while those named Keisha, LaTonya and Tamika are African-American.  Similarly, people typically assume that males named Brad, Todd and Brett are White, while those named Jamal, Rasheed and Darnell are African-American.  

            Do names make a difference in the job market?  Bertrand and Mullainathan ran a test.  They developed a file of fictitious resumes and randomly assigned either "White" or "African-American" names to each. They then mailed samples of the resumes to employers that were advertising job vacancies in Boston and Chicago newspapers.  Their question was simple.  When resumes indicate that applicants are of similar quality and have similar types of education and experience, will an application from a Jamal be treated differently than one from a Todd?

            Not surprisingly, the answer is: "yes."  Applications bearing "White" names were 50 percent more likely to get a callback from the prospective employers than those bearing "African-American" names.  On average, applicants with "White" names needed to send out about 10 resumes to get one callback, but those with "African-American" names had to send out 15.   Interestingly, that 50 percent gap was nearly identical for both males and females in both Boston and Chicago.  The racial gap persisted for all types of jobs studied: executives, supervisors, sales representatives, sales workers, secretaries, and clerical workers.   The gap persisted even for employers that specifically advertised themselves as Equal Opportunity Employers.  In short, for every type of job and every type of employer, names mattered. 

What about beauty?  

            It isn't easy being green, or female or African-American or Hispanic or a member of any other ethnic or linguistic minority.  What about ugly?  Is discrimination confined to race and gender?  If racism and sexism exist in the marketplace, what about lookism?  Well-wishers often assert "it's what inside that counts."  Do job markets agree?  Apparently not.  Raw physical beauty bestows a major employment edge on both men and women.

            Drawing on three surveys encompassing more than 7,000 respondents, economists Dan Hamermesh and Jeff Biddle tried to statistically isolate the impact of beauty on earnings.³  In addition to gathering data on employment and earnings, the interviewers rated the respondents' physical attractiveness.  Respondents were categorized as either strikingly beautiful or handsome, above average for age, average for age, below average for age, or homely.  Overall, 34 percent of the respondents were rated as striking or above average while only eight percent were rated below average or homely.  [The interviewers must have been generous.]

            Even after adjusting for such factors as education, experience, age, race, and occupation, the authors find that beauty has a significant impact on earnings.  The impact is not as great as that of race or gender, but it is significant nonetheless.  Respondents rated as striking or above average earned a wage premium of about five percent, while those rated as below average or homely suffered a wage penalty of about nine percent.  The total earnings gap between the two groups was about 14 percent.  Interestingly, the impact of physical appearance was somewhat greater for males than for females, though the difference was not statistically significant.

            Of course, attractive workers might be more productive in some industries.  Fashion modeling is an obvious example, but there are many others.  If customers prefer dealing with good-looking employees, attractive sales representatives might bring in extra revenue for the firm and attractive waiters/waitresses might lure more paying clients into a restaurant.  In these cases, higher productivities could justify higher earnings. The authors do find some evidence that attractive employees gravitate to those occupations in which physical beauty is most likely to matter, but the quantitative importance of this occupational sorting seems quite weak.  Employer bias, again, seems a likely suspect.

Solutions?

            Economists have no magic bullet to slay bigotry; but our arsenal is not empty.  Remember that discrimination takes a bite out of employers' profits as well as the pride and pocketbooks of those on the receiving end.  Indulging our prejudice can be expensive.  If firms pass up minority applicants to hire less-qualified white males, their bottom lines ultimately will suffer.              

            Managers in protected environments can afford lost profits, but those in dog-eat-dog markets cannot.  A monopolist can hire Uncle Louie's inept son-in-law and survive.  A perfectly competitive firm cannot.  Less efficient workers mean higher costs that rival firms are poised to exploit.  Firms facing cutthroat competition live continually on the edge, knowing that the slightest strategic lapse might unlock the door holding the competitive wolves at bay.

            Competitive firms in competitive markets have no freedom to discriminate.  When survival depends upon productivity, bias takes a back seat.  Such firms seek the best workers regardless of race and gender or, in the long run, they pack it in.  More and better competition will not prevent bigotry, but it will certainly constrain firms' ability to act on it. Discrimination is less likely to occur in competitive markets than in non-competitive ones.

_____________________________________________

Notes:

1.         Kenney, Genevieve M., and Wissoker, Douglas A., "An Analysis of the Correlates of Discrimination Facing Young Hispanic Job-Seekers," American Economic Review, volume 84, number 3, June 1994, p.674 (10).

2.         Bertrand, Marianne and Mullainathan, Snedhil, "Are Emily and Jane More Employable than Lakisha and Jamal?  A Field Experiment on Labor Market Discrimination," American Economic Review, volume 94, number 4, September 2004, p.991 (23).

3.         Hamermesh, Daniel S. and Biddle, Jeff E., "Beauty and the Labor Market," American Economic Review, volume 84, number 5, December 1994, p. 1174 (19).

_______________________________

Testing Yourself

To test your understanding of the major concepts in this reading, try answering the following:

1.         Describe the results of the Kenney and Wissoker experiment.

2.         Explain the difference between customer bias and employer bias and how a researcher might determine which is the more likely cause of discrimination.

3.         Explain how the amount of competition in a market might affect the prevalence of employer discrimination and explain why.

Return to Contents