David I. Levine, "Editorial: Overt Discrimination by Multinational Firms," Industrial Relations, 37, 2, April 1998, pp. 121-125.

The extent and even the presence of racial and sexual employment discrimination in the United States today remains an area of active research and dispute. A number of authors claim that gaps in labor market outcomes are due to differences in skills and tastes, not to labor-market discrimination (Herrnstein and Murray, 1994; Juhn, Murphy, and Pierce, 1991). Other authors use regression analysis, field experiments (Fix, 1993), or the rare natural experiment (Goldin and Rouse, 1997) and find evidence consistent with continuing employment discrimination.

Until relatively recently, the presence of discrimination was not in question. Employment advertisements in the United States routinely listed gender and ethnic qualifications. "No Irish need apply" was an infamous of the 1800s. Until the mid-1960s, when they were first outlawed, similar statements remained in many employment postings. The presence of these "smoking guns" left no doubt that discrimination persisted.

The 1964 law making such advertisements illegal in the US (and similar laws in Europe and more recently and less strongly in Japan) have helped reduce discrimination. At the same time, it is more difficult to identify discrimination when it is no longer legal to announce it.

In many nations explicit discrimination remains legal. In a paper appearing in this issue, John Lawler and Johngseok Bae examine discriminatory want ads in Thailand, a nation where such advertisements remain legal. They constructed a novel data set by reading a sample of want ads in the English-language paper in Bangkok. They then examined which multinationals are most likely to use discriminatory announcements. They find that roughly half the Japanese-based multinationals, a third of those from Europe, and a fourth of those from the US explicitly request only men or only women need apply for a given job.

The list of companies with explicit declarations of discrimination include many of the most prominent global players, including companies such as Nissan from Japan, Siemens from Europe, and Nike from the US.

As a rough test of the extent of these findings to other settings, we examined a week=s worth of recent newspapers from Malaysia (The Star) and Singapore (The Straits Times, both for the week of May 19-23, 1997; the data are in the published appendix). The data are subject to many problems. Some multinational companies use headhunting agencies that usually do not identify the company for whom they are hiring. Such ads were not included in the sample. An additional complication is that several ads listed jobs that required different sexes; these ads were counted twice. Finally, identification of foreign-owned firms is not easy, so some of the listed firms may not be multinationals, and some multinationals may have been omitted.

The results were roughly consistent with those in Thailand, although the samples were small. Again, major global corporations appear on the list of discriminators. For example, Sony, the medical products giant Baxter Healthcare Corporation, and the disk drive maker Maxtor all requested that only females apply to be production operators. About 16 percent of the 62 ads in Singapore, and 18 percent of the 22 ads in Malaysia were explicitly discriminatory. In results not included in the Appendix, local Singapore firms were much more likely to discriminate in their want ads than were the multinationals that are included in the table. In addition to sex discrimination there was significant racial discrimination in Singapore; some ads specifically noted that only Chinese need apply. A casual inspection of the one day's ads in The Japan Times (May 19, 1997) found that in spite of a formal policy discouraging discrimination, the newspaper has separate sections for positions only for females and for position open to both males and females. (There is no section for male-only positions.) Of the 263 help wanted ads in English, 67 were in the female section and 196 were in the male and females section.

The presence of "smoking gun" discrimination suggests a number of research possibilities.

• How would the list of discriminatory employers change in a sample of more nations or more years of data? What predicts the patterns of discrimination by industry, occupation and region?
• Do those companies with discriminatory ads have more segregated workforces, or does the lack of explicit discrimination not actually matter?
• How would these results differ for the locally-owned businesses advertising in local newspapers, instead of multinationals advertising in the English-language press?
• Does the presence of explicit employment discrimination increase profits (perhaps due to customer or employee-employee discrimination) or decrease profits (perhaps due to employer or manager discrimination)?
• How do these results differ across time and across nations?

As noted above, it is legal for these companies to place advertisements in these newspapers with gender and racial qualifications. Moreover, in the presence of discrimination by customers, other employees, or other important groups, such discrimination can also maximize profits. At the same time, most of these companies that operate in the U.S. proclaim they are equal opportunity employers within the U.S. Some customers, employees and other stakeholders may be interested in knowing which multinationals vary their policies toward equal employment opportunity as the law permits.

Acknowledgement: John Lawler kindly provided his raw data, and Jimmy Torrez compiled the Singapore, Malaysian, and Japanese data.