A Reply to Richard Epstein on Genetic Testing

dna6.jpgIn his first post to the relatively new Chicago Law Faculty Blog (which has turned out to be a really interesting blog by the way), Professor Richard Epstein argues against my recent post about genetic testing in the workplace. Epstein disagrees with my general view that it is better to restrict employers from using genetic information in making employment decisions.

epstein.jpgEpstein’s argument is based in part on his view that privacy is a form of misrepresentation, tantamount to a kind of fraud by concealing disreputable and harmful information. In this regard, he agrees with his colleague, Richard Posner, who makes a similar argument. If a person knows he will drop dead in a month from a fatal disease, it would be fraud to deliberately conceal this information on a life insurance application. So why not when seeking employment, Epstein asks, since employers often invest heavily in training a person?

Epstein contends that genetic information should not be treated differently than other information – there is nothing that makes genetic information any more special than other health information. He argues that “genetic information is in principle no different from any other kind of information that can be asked about a given person.”

Genetic information, however, is sometimes different than other forms of health information in that genetic testing often doesn’t reveal the presence of a particular disorder, just the fact that a person is at a higher risk than normal to develop such a disorder. People with genetic predispositions for certain disorders can be perfectly fine at present, and they may never develop the debilitating condition. Should they be treated disadvantageously because of the greater possibility of developing some future condition? In a way, they are being treated differently from others because we are better able to assess the risk for their potential health disorders. In other words, people with predispositions to develop genetic conditions often come with more information than people who might develop non-genetic conditions. The information becomes a curse, a form of baggage that can result (at least partially) in that person already being treated as if she has the condition she is predisposed for.

It is true that it is rational for employers to discriminate against people with genetic conditions. But I believe it is rational for society to limit the extent to which employers can discriminate on this basis and the extent to which employers can learn this information. That’s because the employer’s ability to make efficient business decisions is not the only social value at stake.

One countervailing social value is ensuring individual freedom and autonomy. Work for many people is more than just an income stream. It is a central part of their life. I, for one, do my job because I see it as essential to who I am, not because of the money, as I could be earning much more working for a law firm or in other position. Work isn’t just about “show me the money.” As I asked in my initial post, do we want a world like that in the movie Gattaca, where people cannot pursue their dreams because of genetic predispositions?

Another countervailing social value is privacy. In an article I published a few years ago, I took on Epstein on this point. There is a lot of information that employers might find useful in determining an employee’s health risks: her eating habits, sex life, family health history, hobbies (does she like to do risky things like skydiving?), financial condition, children’s health (after all, if an employee has an unhealthy child, this can detract from workplace productivity), and more. There are limits, however, imposed both by law and by social norms, on what employers can demand to know about employees. And for good reason. Even though the information might help the employer make a rational decision, we value privacy. Privacy certainly isn’t free – it comes at a cost, but that cost might well be worth paying.

Epstein also takes issue with the argument that without privacy, people will be deterred from undergoing genetic testing:

One argument is that people will shy away from finding out their genetic position if they are subject to the test. Don’t believe all this. By assumption, there is good reason to believe that the information that is acquired from genetic sources is of value not only to the employer but also to the employee. Suppose that a women has the gene that renders her susceptible to breast cancer, which if identified would allow for certain prophylactic choices. Does it really make sense to think that she would choose not to get that information if she had to disclose it to a prospective employer? Hard to believe that workers would take that kind of risk with their own lives.

Ultimately, the extent of deterrence is an empirical question that neither of us has the data to answer. Epstein makes his argument in terms of what the hypothetical rational person would do. And at that level, I can offer a counterargument. In his example of the woman getting genetic testing for breast cancer, he assumes that the woman would clearly not want to risk her life just to keep her job. But many people risk their lives for their jobs. Look at the soldiers in Iraq, police officers, firefighters, and others. Thus, it is not entirely clear that the rational person wouldn’t put her health at risk for her job. And people don’t always act as the rational economic actors as assumed by some law and economics scholars. There are many people who are already reluctant to find out information about their health, even when it could help them. Risking damage (or even an end) to one’s career can make this reluctance all the greater.

You may also like...

13 Responses

  1. Adam says:

    Great article, just a small point about risk-taking: HIV testing is offered anonymously because there was (and is) a substantial group of people who won’t get tested if they have to give their name. Epstein may say think that its “hard to believe that workers would take that kind of risk with their own lives,” but there’s emperical evidence that people make precisely those choices.

  2. John Armstrong says:

    I don’t think you have to look as far as soldiers, police, and fire fighters to find people willing to risk their health for jobs. Life looks a lot different with the job security of tenure.

    Say you’re unemployed (“downsized”) and you’ve got a choice to get a genetic profile. Also, assume that if anything on a certain list of propensities shows up you’re likely not going to get hired to any well-paying job. Also, if any of those conditions show up you can modify your behavior somewhat to improve your odds in the long run, though the propensity will still get you rejected from jobs. Now you’ve got a choice: maybe possibly not doing certain unusual things that would slightly improve your long-term prospects for health(everyone has benefits to eating well and exercizing, so that doesn’t count), or very likely not being able to make ends meet. No matter how inclined your body may be towards melanoma decades in the future, starvation and exposure will kill you now.

    From a public health standpoint, then, how do we remove the penalties associated with behaviors we want to encourage? Limits (if not IBM-style outright bans) on what genetic information is and is not a rational basis for disqualifying an applicant.

  3. Paul Gowder says:

    Your point about genetic information representing a mere predisposition is conclusive, or close to it, I think.

    We recognize, in this society, a fundamental injustice in permitting individuals to be injured based on statistical patterns absent extreme need. That’s why we don’t ordinarily permit racial profiling. Similarly, we don’t permit employers to discriminate against disabled people on the basis of their diagnosis alone, even when that diagnosis could impair them from doing the work, without an individualized assessment.

    (In fact, considering employment law generally, Epstein’s argument proves far too much: an old person is statistically more likely to die sooner than a younger person, yet we forbid employers from discriminating on the basis of advanced age…)

    This is also, I think, why many people experience the insurance industry as crooked and/or unjust (and particularly the compulsory auto insurance industry, whereby the state requires me, in order to drive, to do business with people who think that I’m less responsible because I’m male and not married, and charge accordingly). And why they have some justified skepticism about IQ tests. Etc. etc.

    Cf. just about everything written by my beloved Foucault, also Neil Postman’s “Technopoly,” and dozens of other works of cultural critique…

  4. Chris Beck says:

    If we were to speak purely rationally about risks and costs associated with them, then companies would be allowed to discriminate against handicapped employees because the cost of accessibility is not negligible.

    So – having mostly agreed that our society occasionally does things not based purely on the bottom line, I think it is fair to say that we can, as a society, not choose to discriminate against genetic information. If we started, it would very quickly become a very steep, very slippery slope.

  5. HeScreams says:

    I’m not well versed in employment law, but there seems to be some inconsistency in Epstien’s position. If genetic testing was used for employment decisions, does that mean an employeer could discriminate against someone with a genetic probability of contracting a disease, but not against someone who already *has* that disease?

    If a genetic predisposition for a disease kept you from working, does that mean you qualify for Disability?

  6. HeScreams,

    I believe, though I don’t want to speak for Epstein, that he would say that employers should be allowed to discriminate against both people with genetic predispositions and people who actually have the disease. This is his normative position, as the law currently puts some restrictions on the extent to which employers are allowed to discriminate based on health conditions. In fact, Epstein takes issue with anti-discrimination law more generally as it applies to private sector employers:

    I published my book Forbidden Grounds in 1992. There I took the position that current anti discrimination laws should be repealed insofar as they apply to private competitive employment markets. I relied on my libertarian bent to explain why a system of free entry and open markets can better cope with invidious forms of discrimination in employment markets than any system of government mandated and enforced norms. Prejudice in markets is always costly to the parties who practice it, so that there is a strong corrective against unthinking prejudice that works without the heavy administrative costs and logrolling mentality that often infect the political decision making process.

    Richard A. Epstein, The Legal Regulation of Genetic Discrimination: Old Responses to New Technology, 74 B.U. L. Rev. 1 (1994).

  7. Dave! says:

    I think you raise great counter arguments to Prof. Epstein’s position, but I think that with respect to employment and health concerns, the traditional economists “rational person” is fundamentally flawed.

    Because genetic testing is linked to health issues and in the U.S., at least, health insurance is typically tied to employment, the two issues cannot really be separated.

    Take, for example, the issue of a genetic test for breast cancer. Of course a “rational person” might want to know that they are at risk–to take advantage of the prophylactic measures Epstein advocates. But what if the employer doesn’t want to take on an employee because of the impact on the costs of health care to someone who could potentially get a costly disease?

    The employee risks the very employment that provides them with the health care they are dependent on for the treatment of their (potential) illness. This is hardly a set of circumstances conducive to rational decision making.

    There is nothing fraudulent or misleading about wanting to keep information that is not detectable without modern technology private. I’m sure a financial audit of every potential job applicant would reveal information that might be valuable to employers as well (such as a propensity for embezzlement) but we don’t necessarily allow or advocate that for all potential employees.

  8. A (hopeful) UC undergrad applicant says:

    Note: I agree with Epstein.

    “Privacy certainly isn’t free – it comes at a cost, but that cost might well be worth paying.”

    Yes, I agree. But why should society bear the cost? Why not have each individual decide whether to bear that cost? Each woman may decide how much information to release in return for a chance at a greater salary and greater benefits. If genetic discrimination were allowed, you would agree that a fair share of people would resist having their genes examined. Employers could capitalize on this group by hiring them at wages commensurate with their increased risks and lessened productivity and the like. The drastic scenarios imagined, specifically no job prospects for those refusing examination, relies on the fallacious assumption of a monopolistic employment market. Epstein rejects these notions in his Simple Rules.

    “We recognize, in this society…” The fact that the laws effectively prohibit Epstein’s preferences for uninhibited rational discrimination does not constitute a valid argument. Our society as you say is liable to be manipulated by every whim. Epstein’s philosophy may fly in the face of decades of legislation, but, in my mind at least, that does nothing to discredit it.

    “I’m less responsible because I’m male and not married” That is the statistical reality. If not, insurance companies would reject those tools, and continue to seek worthy ones. If males are more predisposed to crime than females, does concern for safety not demand that one focus more attention on males (as that would yield more arrests than say eyeing men and women equally)? Yes, this opens the door to racial profiling et al. But that does nothing to weaken Epstein’s position; It just intensifies (and renders even more controversial) the question.

    Dave!- I do not know the laws regarding it, but the optimal situation would be where there is no law mandating that employers provide health insurance. The sick man in your analogy would negotiate with the employer, and forgo health insurance, and/or his wages to bring the total cost of employ roughly equivalent to a similarly productive prospective hire, theoretically speaking of course. If the law mandated that employers provide health insurance, your sick man may find himself unemployed, or stuck with a salary less than what he would have obtained without the law.

    The only problem I see, is the issue of adverse selection in the case of nondisclosure. There would be a disparity between the wages of disclosed and nondisclosed people. Some of the nondisclosers would realize it in their pecuniary interest to disclose, raising the disparity for the remaining nondisclosers, akin to the adverse selection problems in health insurance (and college loans based on a percentage of future pay). But I would expect a (non-legal) solution to arise to remediate this problem.

  9. “The Force Is Strong In My Family”

    Posted by Adam In Friday Star Wars Security blogging, I was planning to start on Saltzer and Schroeder this week. But I’m going to detour a bit into genetic privacy (and Star Wars, of course). I’m inspired in part…

  10. Paul Gowder says:

    UC undergrad: the things “we recognize, in this society” have more to do with our moral intuituions than with “decades of legislation.” I suggest that permitting the imposition of injury on certain individual members of society on the basis of “statistical reality” is inconsistent with those intuitions because it creates individual injustice as the price for social efficiency. Because the institutions (insurers, states, employers, etc.) are unwilling to incur the costs of engaging in an individualized assessment of the various risks that their employees present, they accept a higher rate of errors, but fail to internalize the cost of those errors. Those errors are instead suffered by the perfectly good unmarried male drivers, the perfectly law-abiding black males, the perfectly healthy fat people, etc.

    If we really look at it closely, it’s a distributive question. The institutions making these decisions save investigative costs by imposing a risk of wrongful non-selection on their employees, customers, citizens, etc. How is this just?

  11. Genetic Testing: Further Debate with Richard Epstein

    Richard Epstein has posted a reply continuing our debate over whether employers should be able to use genetic testing information to make employment decisions regarding employees. Here are the posts in our debate so far: 1. Solove, IBM vs. NBA:…

  12. Two types of genetic testing can occur in the workplace: genetic screening and genetic monitoring. Genetic screening examines the genetic makeup of employees or job applicants for specific inherited characteristics. It may be used to detect general heritable conditions that are not associated with workplace exposures in employees or applicants. For example, employers used genetic screening in the early 1970s to identify African Americans who carried a gene mutation for sickle cell anemia. Those carrying the gene mutation were denied jobs-even though many of them were healthy and would never develop the disease. In these cases, genetic screening to identify the sickle cell trait often occurred without the consent of the individuals.

    In addition, some employers may seek to use genetic tests to discriminate against workers – even those who have not yet or who may never show signs of disease-because the employers fear the cost consequences. Based on genetic information, employers may try to avoid hiring workers who they believe are likely to take sick leave, resign, or retire early for health reasons (creating extra costs in recruiting and training new staff), file for workers’ compensation, or use health care benefits excessively. A 1989 survey of large businesses, private utilities, and labor unions found that 5 percent of the 330 organizations responding conducted genetic screening or monitoring of its workers. Another 1989 survey of 400 firms, conducted by Northwestern National Life Insurance, found that 15 percent of the companies planned, by the year 2000, to check the genetic status of prospective employees and their dependents before making employment offers. Thus, there is evidence that genetic information continues to be used to discriminate against qualified workers. The economic incentive to discriminate based on genetic information is likely to increase as genetic research advances and the costs of genetic testing decrease. Arizona PEO

  13. Dan says:

    Questioning whether employers should or should not be allowed to use genetic testing to screen potential employees is another step down a disastrous path for businesses and health policy. The important question is not whether genetic testing allows increased efficiency in the market. It is how we have gotten to the point where businesses are so willing to pervert their hiring practices that they will use a criterium highly irrelevant to job performance to screen employees.