Do employers discriminate against the disabled?

Persons with disabilities earn less and are less likely to be employed, a phenomenon often explained away as the result of lower productivity due to the impact of functional limitations. In an important new book, economists Marjorie L. Baldwin and William G. Johnson review economic studies utilizing labor-market data that suggest employer discrimination may contribute to the poor outcomes for workers with disabilities.

Regardless of their workplace productivity, disabled people earn less and are less likely to be employed than their able-bodied co-workers, according to a review chapter, authored by two Arizona State University researchers and published in an award-winning book.

"This is the first time that discrimination of the disabled has been studied in the same economic framework as discrimination of Hispanics, women and other minority groups," said co-investigator William G. Johnson, who is the director of the Center for Health Information and Research at ASU's Ira A. Fulton School of Engineering. "We had no empirical knowledge if disability discrimination was taking place and what its impact was; now, we have actual measures of it."

The book, "Handbook on the Economics of Discrimination," was recently named one of the "Outstanding Academic Books" for 2006 by the American Library Association's review journal. Johnson and fellow economist Marjorie L. Baldwin, director of the School of Health Management and Policy at the W. P. Carey School of Business, jointly researched and wrote a chapter called "A critical review of studies of discrimination against workers with disabilities."

It's all about productivity or not!

Baldwin and Johnson wanted to test the prevailing wisdom — never proven empirically — that disabled workers earn less and have other job-related woes simply because their health conditions make them less productive, more likely to miss work and unqualified for many positions. And if they are the first to go in a staff cut, that's simply because they're among the least productive workers, the argument goes.

"People who take other forms of job discrimination seriously are often dismissive about disability discrimination, figuring the ill effects are deserved by poor performance," Baldwin explained. Investigating this topic is tough, partly because the term "disability" is defined differently. Until the mid-1970's, the medical model — defining disability by the nature and severity of a person's health condition or impairment — was dominant.

Newer models define disability based on functional limitations or non-health related influences on labor supply. Then there is the fact that disability spans so many conditions, from epilepsy to paraplegia, deafness to morbid obesity, making it hard to compare apples to oranges. The new study isn't the first collaboration for the Phoenix-area economists. Johnson published his initial paper on disability discrimination in 1984, and he and Baldwin have partnered on several projects since.

Soon after the Americans with Disabilities Act took effect in 1992, officials from the National Institute on Disability and Rehabilitation Research invited Baldwin and Johnson to present their findings to staffers. To the economists' surprise, NIDRR staffers strongly disagreed that a worker's disability-related functional impairment can hurt his or her productivity. The staffers insisted that productivity problems come only from unenlightened co-workers, bad workplace policies or discriminatory bosses, not the limitations of a disabled worker.

Any functional limitations can and should be compensated for, their reasoning went. Today, many disabled-rights advocates share this belief system, Johnson said. Baldwin and Johnson are convinced that the data support the idea of productivity differences, but they still wanted to know if, as they suspected, even disabled workers whose productivity matches non-disabled peers still sometimes suffer prejudicial treatment.

So they sifted through more than a hundred studies on disabled workers before settling on a dozen containing a rarity — actual labor-market data arranged in economic models. Depending on the study, wage gaps from zero to 21 percent were noted between disabled and non-disabled workers. Differences in employment rates were as high as 25 percent.

The studies used statistical methods to control variables such as workers' on-the-job experience, education and the above-mentioned functional limitations — all potentially the cause of wage differentials. With those variables stripped away, the studies compared disabled and non-disabled workers on a level playing field.

The result: "the evidence strongly suggests that disabled workers are systematically paid less, unrelated to their functional limitations," Baldwin said. The researchers also found that "wage discrimination creates disincentive effects that discourage some persons with disabilities from working." They suspect that other, as-yet-unidentified factors fuel additional worker drop-outs.

Acknowledging discrimination

Baldwin provides a couple of caveats. First, she and Johnson reviewed previously published studies — several of them their own — and as a result, had to rely on data already gathered according to particular methodologies. For instance, males were more often represented in the studies than females. Second, the studies reviewed dealt mostly with functional losses associated with physical disability, rather than cognitive losses.

It's possible that results would be different if more cognitive data were studied. "We were studying what was already noted. Is it perfect research? No, but if we waited for the perfect data to be gathered, most research would never get done," Baldwin said. "What makes this chapter important is that we are looking across a range of studies, using different methodologies and data, and the conclusion is we can't explain it all away without acknowledging discrimination."

Johnson next wants to look at "measures of prejudice:" how social attitudes about the relative "acceptability" of specific disabilities impacts wages and other workplace factors. For instance, is someone with schizophrenia discriminated against more severely than a person who was crippled by polio? Is deafness treated more fairly than end-stage liver disease? "We saw some indications of prejudice measures when reviewing the disability studies most recently, but the sample was small," he noted.

"Assuming that differential reasons can be separated, then it might be possible to design policies to remedy this type of prejudice." In addition to finding evidence of job discrimination based on disability and unrelated to productivity, Baldwin said there's another lesson learned from their project.

"ADA, since 1992, has not had a whole lot of impact in terms of improving wages and employment rates. And that's important, because when disabled workers can't find employment, they go on government assistance," she explained. "As a society, we need to find better ways for disabled people to enter and stay in the workforce."

Bottom Line:

• Disability spans so many conditions, from epilepsy to paraplegia, deafness to morbid obesity, making it hard to compare apples to oranges.
• Many disabled-rights advocates share the belief system that productivity problems come only from unenlightened co-workers or bad workplaces policies, not the limitations of a disabled worker.
• Baldwin and Johnson wanted to test the prevailing wisdom — never proven empirically — that disabled workers earn less and have other job-related woes simply because their health conditions make them less productive, more likely to miss work and unqualified for many positions.
• Evidence suggests that disabled workers are systematically paid less, unrelated to their functional limitations. This creates disincentive effects that discourage persons with disabilities from working.
• Since ADA's inception in 1992, not much has happened in terms of improving wages and employment rates for the disabled.