Vol. 46, No. 3
Research by a Nobel Memorial Prize-winning economist sheds light on how affirmative action both helps and hurts minorities, how working mothers decide to quit their jobs and how training dollars are wasted.
Last fall, University of Chicago economics professor James Heckman, 56, was named co-winner (with Daniel McFadden) of the Nobel Memorial Prize in Economic Sciences for his contributions to econometrics—a field combining economics and statistics. Econometrics provides the standard of proof for microeconomics research, which studies everything from household spending and investment by firms to the organization of industries, labor markets and the effects of public policy.
HR Magazine interviewed Heckman about the implications of his research for human resource management.
HR Magazine: In selecting areas to research do you look for practical problems to study?
Heckman: Yes. I think that is the source of all scientific progress. Almost every problem I’ve started with was motivated by a practical problem. I tried to organize it in some kind of intellectual framework. Then starting from that, went on to develop models and methods. I don’t view economics as just a theoretical exercise.
HR Magazine: Why should HR be interested in econometrics? What is its value?
Heckman: Statistical analysis enables you to produce an objective summary of what’s happened in the past. It’s a tool that helps you build from a series of case studies a body of knowledge about what has worked previously and predict what will work in the future.
HR Magazine: HR managers often feel overwhelmed by the reams of data and reports they are asked to compile. Is there a point of diminishing returns?
Heckman: The danger is that HR collects a huge amount of information, but it is not processed systematically or accumulated using proper statistical methods. The information stays in the mind of the current HR manager, who uses it to make seat-of-the-pants judgments. That’s fine, but the data doesn’t accumulate in the same way it would if there was some ongoing statistical analysis.
Affirmative Action, Reaction
HR Magazine: You attribute the failure of certain groups of people to gain access to the workplace at least in part to discrimination they experience early in life. What do you mean?
Heckman: If there is discrimination, it’s in the family and environment you’re born into. A growing body of scientific research shows that young children developing in neighborhoods with high levels of violence and stress actually have their brain growth inhibited, leaving them with a major disadvantage later in life. We’ve known that the inhalation of toxic chemicals, eating lead and that sort of thing, is a serious retardant to brain growth. But the similar impact of social influences is just emerging.
Actually, I’m happy to find this out because it means that we can act to eliminate these differences. For example, if you take the viewpoint that [racial] differences are genetic, you will have a very pessimistic view of society. You are going to believe that certain groups of people are always going to be disadvantaged.
HR Magazine: So in the debate over nature versus nurture, you’re strong on environmental influences.
Heckman: Everything we know now says nature plays a role, but it’s powerfully reinforced by nurture. In each generation—we’re talking every 20 years—the IQs of the children go up. What it means fundamentally is that you are getting an incredible amount of input from the larger society and that is affecting intellectual growth.
HR Magazine: How do you relate what you’re saying to affirmative action and diversity programs as we know them today?
Heckman: When I study civil rights and affirmative action, I break those two topics completely. I look at the history of the litigation and the history of civil rights activity from the 1930s to 1965 as a period when discrimination and segregation in the South were major problems. Laws prohibited blacks from being part of the larger society and inhibited all kinds of development. That’s very pernicious. Getting rid of it was an extremely useful and valuable step, not only in the political sense of changing society, but economically. It lets firms get workers they couldn’t hire before because of social and community pressures. The Civil Rights Act of 1964 was pro-business, working to promote the employment of blacks. Between 1965 and 1975 we see a huge success in the economic status and well-being of black Americans.
HR Magazine: Isn’t affirmative action a natural continuation of the earlier civil rights legislation?
Heckman: My feelings about affirmative action are equivocal. It’s played a role in creating the black middle class, but at the same time it’s created a certain amount of dissension and bitterness in the larger society—especially when the market has been declining, as it has been over the past 15 years, for low-skilled workers. American society is extremely open to advancing minorities. If you look at television and study the popular culture, there’s no doubt most Americans want African Americans to succeed. On the other hand, when the success is perceived to sit outside the rules of the normal workplace—whether it’s blue collar or white collar—animosity is created. It’s led to a kind of racial politics in the workplace, which I think is unfortunate.
HR Magazine: But hasn’t affirmative action helped more minorities and women to enter and advance in the workplace?
Heckman: Clearly, when a company is told to hire more blacks or women and there is a mechanism in place to enforce it, it has led to an increase in minorities in specific organizations. But overall, affirmative action hasn’t translated into widespread improvements for blacks as a group. That’s partly because when one sector is targeted for hiring or promotion, it takes workers from other sectors, resulting in a game of musical chairs.
HR Magazine: Do you see unanticipated problems resulting from affirmative action?
Heckman: I’m concerned that because it’s harder to fire a black or woman, firms are going to be more reluctant to hire them in the first place. What you get is a mixed blessing. Firms are compelled to treat people fairly and when they are forced to hire, they will. But the employee becomes less desirable economically from the point of view of the firm. You’re dealing with persons with a legal halo around them. You assume the added expense of dealing with litigation and the cost of terminating an essentially bad employee. For example, if you fire a black and don’t also fire a white for the same position, you’ve opened the firm to potentially costly claims of discrimination. With the additional costs of affirmative action, the overall demand for labor goes down. Affirmative action may change the racial composition and may help blacks relative to whites, but it creates a zero-sum game. One gains and the other loses.
HR Magazine: Critics of affirmative action claim it encourages a form of reverse discrimination. What are your views?
Heckman: There is the fear of reverse stigmatization under affirmative action—that blacks who were hired under this policy can’t possibly be as good as whites. A famous economist, a refugee from Poland, gave me an example that demonstrates the impact of government policies on stigmatization, plus and minus. In the 1920s there was profound anti-Semitism in Poland and, as a result, only the most brilliant Jewish students could get into medical schools. It created a luster; everybody knew that if you had a Jewish doctor, this doctor was superb. He’d had to jump through 100 hoops and then some to earn his degree. Then in the 1930s, the Polish government tried to reverse the stereotype; they started admitting only dumb Jews to medical school. If you were smart and you were Jewish, you couldn’t get into medical schools.
Today, there is a general perception that haunts blacks that, for some, success comes because they are black, not because they’re good. That leads to a discounting of some very competent blacks. It’s Poland in the 1920s in reverse. We perceive the black is incompetent because the scale was tilted at the early stages. There’s a general perception there that people are not willing to admit is a problem. If you appear to dilute qualifications, it may actually shift demand against the group you are trying to help.
HR Magazine: Are there other reasons why affirmative action is missing the mark?
Heckman: One of the mysteries is why there are so few competent blacks in certain areas of the workplace. I would go back one step: why are so few blacks and Hispanics going to college relative to whites? Contrary to popular opinion, the availability of college funds for tuition for 18-year-old minority students is not the reason. The limiting factor is college readiness—the ability to actually benefit from a college education.
Our society has gone out of its way in terms of affirmative action policies in education to promote aid to disadvantaged groups. Any minority who has graduated high school can get government grants to go to a community college and then possibly go on to a four-year school. Right now, you have Pell grants [government tuition grants] on the table that are going begging.
Leaving School Early
HR Magazine: High school and college students with high-tech skills are being offered attractive financial and educational incentives by organizations eager to recruit them. Should these young people be encouraged to complete their education? What are the economic implications if they don’t?
Heckman: Those individuals run a risk, although the scenario you are describing is somewhat unusual. Today, it is difficult to get a good high-tech private sector job without having a least something like an associate degree level of skill. In the market at large, having a college degree matters a lot. It gives people more general skills that are useful in a greater set of circumstances. It gives them more mobility. To the extent that people out of high school are going directly into companies, my assumption is they’re moving into tracks where they are going to get access to community college education, or at least substitutes for it.
For students who plan to complete only two years, if anything, there is a slight premium to the community college over a state university. Community colleges offer focused training programs, you get an associate degree, you learn a set of skills, and frequently you’re attached very closely to the workplace.
HR Magazine: The German apprenticeship model, which combines on-the-job training with study, frequently is cited as a model that the United States should consider. Do you agree?
Heckman: The German apprenticeship model is no great shakes. It has many good features but it suffers from inflexibility. In German society, mobility is very limited and these apprenticeship programs reinforce this characteristic. If you train as a plumber and then decide you would rather be a mechanic, you’re screwed. You have to start all over again.
Also, German apprenticeship programs enable firms to pay workers about a third of the wages that they would normally get for a task. It allows companies to avoid paying union wages. In most industrialized countries, the unemployment rate of workers 18 to 20 is higher than for workers 20 to 24. Germany is the only country where the rate rises for workers 20 to 24. Basically, a lot of the apprentices are living at home. Their parents are supporting them; they’re getting paid sub-minimum wage, below the union scale. It’s just a low- wage subsidy to employers.
HR Magazine: You studied decision-making by married women regarding when and how much to work. What did you conclude and what are the implications for HR?
Heckman: In my research on white women (who provided the broadest sampling opportunities), I’ve found that women who could have earned the highest wages often were not in the workplace. This finding runs against most economists’ intuition and the commonly held stereotype. We typically think of the people who aren’t working as low-wage people and they aren’t. Many professional women choose not to work for wages. What they’re working on instead is child-rearing, building a home or something outside the home they value more highly than paid work. When these women do choose to work for wages, they often take breaks resulting in career absences or seek opportunities with reduced workloads.
HR Magazine: If high-earning professional women are more likely to leave the workforce than their male counterparts, won’t it make investment in their training and career development more of a risk for an employer?
Heckman: When I began my research, many observers believed that it was risky to hire women because they couldn’t be trusted to stay in their jobs. They predicted economically unsupportable turnover rates. But this is an example of where statistical analysis is important for HR. In my study, I found there is a huge number of women working full time who are reliable workers. You can count on them; they’ll be there tomorrow.
I found a pattern where most women who worked, about 80 percent, worked most of the time. Women who didn’t work, about 20 percent, almost never worked. Therefore, a manager might want to take into account that some women are more reliable than others. The task, if you are going to invest in a woman worker, is to forecast whether the potential hire is within the 80 percent.
HR Magazine: How?
Heckman: For older women, look at work histories; for younger women, professional degrees and career choices they have made. By choosing somebody who has gone to college and has made a professional commitment to some kind of specific training, you can at least improve your chances of picking a winner. The danger of this, of course, is that you must act within the framework of anti-discrimination laws. But by using statistical methods you can make good predictions as to who is going to stick around and who isn’t.
Training Older, Younger Workers
HR Magazine: HR invests liberally in training and continuing education for employees well beyond traditional school age. What do your findings suggest about these interventions?
Heckman: The training literature and the whole training discussion has been marred by failing to recognize that we know it is very difficult to teach older workers new ideas. Typically, they are harder to train and have a lower incentive to take training. They are frequently [going to be] in the firm for another few years—say 10 years—whereas younger workers might see that training will pay off for 20 or 30 years and they therefore have a much bigger economic incentive to take the training.
The training literature and the whole training discussion has been marred by failing to recognize that we know it is very difficult to teach older workers new ideas. Typically, they are harder to train and have a lower incentive to take training. They are frequently [going to be] in the firm for another few years—say 10 years—whereas younger workers might see that training will pay off for 20 or 30 years and they therefore have a much bigger economic incentive to take the training.
The more able the older worker, the easier it is. But ability and age both contribute to the ability to benefit from training.
Professionals—doctors, lawyers, engineers and more able workers—who are 50 to 60 years old still have enormous incentive to learn the latest techniques, whether it’s word processing or a technique for diagnosing a disease. For these people, participation in training programs is very high.
But for blue-collar workers and for low-skilled, low-ability individuals the amount of post-40s learning is typically very low. Also, when you’re older, the fact that you are going to have limited horizons to cash in on your learning typically discourages many people from participating.
HR Magazine: Should HR take these findings about age and learning into account when allocating training resources?
Heckman: All things being equal, learning is harder for older workers. So if there is a mistake, it’s that we’re trying to invest training dollars in older workers who either don’t want the training or for whom it’s not beneficial.
HR Magazine: Is there a similar argument for younger, low-ability or underprepared workers?
Heckman: Yes. The case is similar for younger workers of low ability—without much preparation or schooling, they just cannot benefit much from training. Often they can’t read or write. Given the depth of the problem, there are limits as to what firms can do. It seems there’s a failed family education situation essentially, which makes these workers—I wouldn’t say untrainable, but trainable at very high cost.
HR Magazine: It could be done, but economically it’s not a productive use of resources?
Heckman: No, compared to other uses in training. There is an opportunity cost of capital. If I spend money on a non-productive training program, I’m forgoing profit that I could earn elsewhere. It’s a market test. Private training organizations cannot expect to remedy the failures of 15 or 20 years of bad schools. Even if possible, it’s not going to be a wise investment for most firms because it has a general feature to it. If you train workers to read and write, it increases their mobility, giving you less incentive to invest in that worker.
HR Magazine: Let’s back up a step. Are you suggesting that people are set in their ways years before they enter the workforce, and that by then it’s too late for them to change?
Heckman: As an economist, I’m caught in the role of a bad guy. I wouldn’t devise a policy that prevented people from changing, but I also would think that it would be extremely naive to think you’re going to have massive changes of human beings after a certain age. An analytically correct statement is that you can’t change cognitive abilities—straight IQ—after a certain age. At age 10 or so things get pretty settled.
We see people who are changing when they are 60 years old, so it’s not like change is impossible. But it gets harder as one ages. And certain tendencies for things like getting into trouble on the job, fighting, going to jail—those traits tend to get fairly well established in the mid-teens. They are very difficult to change. Of course, sometimes people do change; they find religion, they get married, they settle down. So I’m not going to say at age 18 everybody is fixed for life. But I do think that there is a very high level of persistence.
For example, in the armed forces, there’s been a substantial body of work looking at how ineffective people with high school equivalencies are in the military. The reason is that GEDs that come in the door are smart, but they’re dropouts, people who couldn’t finish high school. Typically, they’re the ones that get into trouble with the military police, the ones most likely to get dishonorable discharges.
HR Magazine: If early intervention will mean better prepared workers and managers down the road, do you think HR should foster workplace-based child development programs?
Heckman: We need to work harder to impress Americans with the importance of the early years and the benefits of child care programs But for now, if companies just independently initiated these programs, they would probably be a waste of time and money because parents wouldn’t value them. There’s going to be education required on the part of workers and companies about the benefit of child care programs that, along with caretaking, include a substantial component that focuses on developing reading, writing and non-cognitive skills. Eventually, I think workers will be willing to pay for these programs, and under current tax laws there may be some savings for them. In the short run, however, a company that sponsors a program isn’t going to capture anything directly in the cash register, except through lower wages.
HR Magazine: From your perspective, are government regulations placing too much burden on HR?
Heckman: HR must be frustrated because of the time it spends dealing with government regulations—affirmative action, occupational safety and health and other kinds of government policies that impact directly on the workplace. That burden is a serious limitation of current government policy. Of course, it also creates jobs, so HR shouldn’t be too unhappy.
HR Magazine: But sometimes the costs of government regulation are overstated. For example, when the Office of Technology Assessment, a congressional research agency, went back and looked at the actual cost of implementing OSHA rules, it found the costs were significantly less than OSHA predicted. Time and again, through innovation, industry made the safety adjustments for much less than estimated.
Heckman: We typically underestimate the ingenuity of the companies and their managers to adapt to situations. Under affirmative action, for example, many people predicted there would be a great disaster by putting unqualified blacks and women into positions of responsibility. And there is some evidence that quality and productivity levels may have been impaired. But nobody has been able to find any substantial effects of these programs that get anywhere close to validating the dire prognostications. Productivity and cost statistics do not show an impact and I suspect it has to do with this very ingenuity you’re talking about.
HR Magazine: As you look into the future, what additional observations or advice do you have for HR managers?
Heckman: Managers have to think creatively. They must realize that as long as the economy continues to boom and technology continues to change, the labor market for skilled workers is going to get tighter. Whether it’s Brazil, Argentina, China, Eastern Europe or the United States, the bias is still towards skilled labor. Promoting skill formation and providing flexible choices in workplace arrangements will be a central issue within organizations and in the country as a whole.
Finally, on a practical level, the experiences in the workplace transformation that have been the most successful seem to bring units together that are typically opposite from each other. For example, unions need not be viewed in an antagonistic role. Studies, like those on steel plants in Pittsburgh, show that company-based unions can work cooperatively with management to respond to technical change.
James Heckman is the Henry Schultz Distinguished Service Professor in the University of Chicago Department of Economics and a faculty member in the Irving B. Harris Graduate School of Public Policy Studies. He directs the Harris School’s Center for Social Program Evaluation.
Much of Heckman’s work has focused on the impact of different social programs and the methodologies used to measure their effects. He has researched areas such as education, job training programs, minimum wage legislation, mothers’ decisions to work and earnings, child care effects, anti-discrimination laws, civil rights and tax policies.
He earned a bachelor’s degree in mathematics from Colorado College and master’s and doctorate degrees in economics from Princeton University.
Heckman’s study of men in their mid- to late-20s who earned a high school equivalency degree (GED) in the 1980s found that they are not much more economically successful than high school dropouts.
Robert J. Grossman, a contributing editor of HR Magazine, is a lawyer and a professor of management studies at Marist College in Poughkeepsie, N.Y.