THE FORUM ON TECHNOLOGY & INNOVATION

+ + + + +

THE HIGH-TECH WORKFORCE:

HOW WILL AMERICA MEET THE RISING DEMAND FOR

SKILLED WORKERS?

+ + + + +

THURSDAY,

JULY 29, 1999

+ + + + +

This transcript was produced from a tape provided by the Council on Competitiveness.

A-G-E-N-D-A

PAGE

I. Opening Remarks and Introductions 3

II. Featured Speakers

-- Richard McGinn, Chairman & CEO, 12

Lucent Technologies

-- Laura D'Andrea Tyson, Dean, 20

Haas School of Business, U.C.

Berkeley

-- Alan Krueger, Professor of 30

Economics & Public Policy

Princeton University

-- Lawrence Mishel, Vice President, 44

Economic Policy Institute

III. Roundtable Discussion and Q&A 52

P-R-O-C-E-E-D-I-N-G-S

(12:30 p.m.)

MR. ROONEY: My name is Peter Rooney, and I direct the forum for Senator Rockefeller, Senator Frist, and the Council on Competitiveness. It's my pleasure to welcome you to the fifth briefing in an ongoing series of lunchtime policy briefings on technology and new economy issues.

These briefings are brought to you by generous grants from the W.K. Kellogg Foundation, the Alfred P. Sloan Foundation, and The David and Lucile Packard Foundation. We couldn't do it without them. We're very grateful for their support.

Just a brief word about the briefing packets that you have. We have a report on information technology workers from the Department of Commerce. We have the first in their series. As many of you know, they issued a new report last month. It is already out of print. We will be mailing those to you. Everyone that attended today we'll mail you a copy next week.

I want to tell you about a couple of other things that are in the packet. As always, we have the question cards. I guess I've gotten ahead of myself here.

Let me just briefly describe the format. Two Senators have brought you a dynamite panel here today. They are going to briefly set the policy context and introduce our speakers. Then the featured speakers will each have about 10 minutes to tell you their view of the high-tech workforce issue, and then the Senators will throw the floor open to your questions and we'll have an interactive roundtable discussion.

We really want you to participate in that. We've got microphones on the floor, and I really encourage you to come to the mikes and ask your questions. We give our congressional staff the privilege of anonymity, but all of the rest of you I'd ask that you identify yourself, so the panel can have some idea of where you're coming from.

For the congenitally shy, we have these green question cards. If you could fill out these cards, hold them in the air, our staff will pick them up from you and bring them forward and the Senators will ask them -- ask the questions on your behalf, National Press Club style.

Finally, we have these blue evaluation forms. Again, we ask that you take a moment to fill out these forms and give us your feedback. We do pay attention to what you say, and your comments have helped shape today's event.

Thank you very much.

Senator Rockefeller.

SENATOR ROCKEFELLER: Thank you. Welcome to I think it is our fifth of these. Get used to them. We're going to do this for about the next 10 to 12 years, so --

(Laughter.)

-- because it's important. I think we have to understand that the issues are going to change so often, they are so crucial, and we can't fail to understand them from a public policy point of view. So this one I think is probably the one that people are the most confused about, the most concerned about, and that is: how do we train our workforce?

Everywhere that Bill Frist and I go, and everywhere that you go, you hear people talking about the inability to find people to work in information technology business. There was this instant, so to speak, a couple of years ago, or a year and a half ago, when the Silicon Valley imported 50,000 French IT workers, and everybody got mad, and they, you know, said, "We can't do that; got to give Americans a chance."

But then, if you looked under the figures, and you looked at California, you would have noticed that only 15 percent of California classrooms were wired up to even be able to access computers or internet, much less do anything useful with them. And even that, obviously you were talking about a public school system, not advanced computer training, which is needed.

So America is not ready, and there is already a sense of desperation about, how do we train a workforce? How do we find the people who are going to do what -- for an industry which is exploding and which, you know, Alan Greenspan yesterday and many other times has talked about the fundamental engine of what's going to be happening in America over the next number of decades.

So every day the number of jobs required grows, and every day the inadequacy, therefore, of our response becomes highlighted and more dramatic. So, then, what do we do in Congress?

Well, the answer to that is: not much, in general, but on this perhaps we could be helpful.

That's since you left, Laura, our level -- our productivity has gone down in Congress.

(Laughter.)

DEAN TYSON: I’ll have to come back.

SENATOR ROCKEFELLER: There are sort of two things we can do. We can seek ways to improve the workforce, or we can import workers from other places where they've already made those pains and taken those efforts.

And so we have to -- it seems to me that we have to do several things. We have to figure out how to improve our education system. Spellman College, which is a superb college, has made a decision -- I think a fascinating decision -- to sort of take African-American women and focus on math and science.

Now, I don't know what happens to Spanish and French and Frangelico and Johann Sebastian Bach in that curriculum; I don't know. But math and science is what they're going to focus on, because they figure that's the best way for African-American women to advance economically in this society.

Well, but that's at the college level, and we're still worrying about wiring up schools and classrooms for K through 12. So we have to do it at the high school level, we have to do it at the college level, we have to do it at the post-graduate level -- improve our education system.

We also have to decide that all of Americans have to have access to technology. It isn't just a matter of young people. It's a matter of adults. It's a matter of people who are in their forties and fifties who have been laid off from coal mines and steel mills, talking about where I come from, who also have a right to be retrained.

Well, America has never been very good at retraining. We talk about it a lot. It's the moral thing to do. It's the right thing to do. But we don't seem to invest in it, and we don't seem to be particularly creative in it, at least from my point of view.

So to get people who are in their forties and fifties ready to work in the IT world is not impossible. I mean, there's nothing written in stone that says that you have to be 20 or 30 in order to be able to do this work. People who are 40 and 50 are, to some degree, still capable of thinking and even doing. So can we decide that we're willing to do something about them?

And then another obvious thing is the whole question of the H-1B visa question, and last year we raised the caps on those visas from 65,000 a year to 115,000 people a year. Despite that cap, the Labor Department reports that it's wholly inadequate for what are now the needs.

It has run out or will in mid-June, and since we're in mid-July I guess I can say it has run out. And so this Tuesday a bill has been introduced to raise the cap. So all of these things are enormous.

And I'll just say from a personal point of view, I live in a state where our -- we've had basic industries and I've watched them. When I was governor for two years, we had 21 percent unemployment. That was not a whole lot of fun, I might say. But it was because people -- the coal industry, the steel industry, the glass industry, the pottery industry, all kinds of industries, were downsizing, and we as a state were unable to deal with that.

So now, can we as a government deal with it? I think we have to be able to do that, and I think we can, and we have an incredible panel of experts. I'm really, really proud of who they are, what they know, and Bill Frist is going to prevent them -- present them to you. Not prevent them.

SENATOR FRIST: Thank you, Jay.

The High-Tech Workforce: How Will America Meet the Rising Demand for Skilled Workers? As Jay mentioned, four outstanding discussants, and I will briefly introduce them. I'll go ahead and introduce each of them, so we can proceed right through with their approximately 10-minute presentations, and then we can move straight to our discussion question and answer in the usual format.

To my immediate right is Richard McGinn, who is Chairman and CEO of Lucent Technologies, a worldwide research and development and manufacturing enterprise with over 145,000 employees and sales of over $30 billion. He served as CEO of AT&T's Network Systems before helping to found Lucent. At both AT&T and Lucent, he has been on the front lines of global competition for skilled workers.

As a technology-intensive company whose commercial success absolutely depends on delivering innovative products to market quickly, Lucent invests heavily in recruiting and retaining the very best talent.

Next to Senator Rockefeller, Laura Tyson is Dean of the Haas School of Business at the University of California at Berkeley. She served as President Clinton's National Economic Advisor and Chair of the White House Council of Economic Advisors. She was a key architect of the administration policy during the first term, a thoughtful commentator on many of the issues we'll be talking about today on employment and workforce issues. Recently wrote an editorial calling upon Congress to lift the H-1B visa caps and open the gates wide to high-skill foreign workers.

To my far right, Alan Krueger is Professor of Economics & Public Policy at Princeton University and Director of the Survey Research Center at the Woodrow Wilson School of Public and International Affairs.

He is a former chief economist at the U.S. Department of Labor and co-author of an influential treatise in labor economics entitled "Myth and Measurement: The New Economics of the Minimum Wage." He has been a keen observer of the impact of technology on the workplace.

Second to the last, on the end down to my left is Larry Mishel, Vice President and Research Director at the Economic Policy Institute. He is co-author of "The State of Working America," a comprehensive review of employment and wealth and income distribution published every two years. He is an economist with expertise in productivity, labor markets, and industrial relations.

As I mentioned, we'll go through approximate 10-minute presentations by each in the order that I introduced them, and then we'll open the floor to discussion.

Richard?

MR. McGINN: Thank you very much, Bill.

SENATOR FRIST: Richard, pull the microphone down.

To everybody, these microphones end up being fairly directional, and you have to pull them down. If you can't hear in the back, send somebody up or raise your hand, because we want to make sure everybody can hear to participate.

MR. McGINN: Good afternoon. And thank you, Senators Frist and Rockefeller, for inviting me to be part of this distinguished panel that includes friends and colleagues and the opportunity to make a few comments about the growing demand for highly skilled people in this nation. In fact, we describe it as a war for talent. It's so crucial to our success.

Technological innovation continues to fuel the economic growth and global leadership of the United States. Recently, Alan Greenspan noted this in describing technological innovation as well as a loss of pricing power and modest prices of commodities around the world as the key elements that are fueling economic growth in the United States and keeping our economy on a stable footing.

We believe that the communications revolution going on around the world is being led today by U.S. companies. We have a possibility to continue that, but it's very much tied to our ability to attract and retain and to guide highly skilled, technologically competent workers. It is going to have profound impact, changes in the way in which we live and conduct our business and our personal lives.

As was noted, Lucent is the world's largest enterprise in the communications networking industry with expected sales this year approaching $40 billion. Three years ago when we were founded we had 118,000 workers. At this point, we have 155,000 workers. We have added all of those jobs, and to a great extent these are in our high-tech spaces, many of whom are engineers and scientists. Our engineers and scientists now number almost 30,000 individuals.

The rise of the Internet in this decade has sparked that revolution in communications networking. The individuals, in their personal and professional lives, are really driving this digital future with their demands, their aspirations, to live their lives and to create work environments that give them productivity.

The network of the future will be much more than today's voice or Internet network. It's going to be a vast network of networks that delivers voice, data, and video services to people where they want it and when they want it. The people involved in creating these technological wonders and these new services in research and development will be crucial to shaping this future.

This year we're investing over $4 billion in R&D, and over $300 million in pure research, in physics and mathematics and computer sciences. Not many other companies do that separate from cooperation with government efforts, but we believe it's an insurance policy for our future.

For U.S. companies to stay competitive in this global marketplace, we need to have a constant supply of these highly trained, highly sought after individuals. And clearly this will place a premium on the nation's pool of technological masters and Ph.D.s.

In just one area -- software-- which already constitutes the largest portion of people at Lucent's Bell Laboratories, we'll need an ever larger number of highly trained, superbly talented people because software is one of the most people-intensive of all of the technical disciplines that we use.

We're already finding the supply of engineering and science graduates is not only not meeting our needs, but is, in fact, diminishing. And demand for computer scientists and software developers is far outstripping its supply.

Given this shortage of people, and we're trying to add about a thousand scientists, net add scientists and engineers every single year, to fill these high value jobs. What we're doing beyond simply being more aggressive in recruiting, and, in fact, paying out options to meet the needs of individuals paying off student loans in many instances -- there's a dirty secret involved as well.

We are being forced to go outside the United States and to hire people and to employ them in their countries. So of the people, of the 118,000 that has gone to 155,000, many of those jobs are now outside the United States. We are now employing nearly 3,500 engineers in the U.K., in Germany, in China, in Japan, in Taiwan and Korea. Not for market access, but because we can't find the people here to employ.

So we're losing the value added here, we're employing the people elsewhere, and it's a long-term problem for us that unless things change will only get worse.

One of the projects or programs that we're doing to try to adjust the long-term issue, because I believe it's really endemic throughout the United States, not only or solely in a high tech, is called Project Grad. It's a multi-school reform program where we're working with the State of New Jersey and the school system to improve student performance, to lower dropout rates, and, as well, to raise the ability of people to become contributing members of society.

A cornerstone for the program is the offer of scholarships for high school students to go on to college, as long as they pledge to achieve at a basic level of classroom performance, graduate on time, take college prep math courses, and participate in two summer-based summer training institutes. And the teachers have to agree to implement this, and they, too, have to go to training.

We in the Ford Foundation have embraced this and have gone to one school in Newark, New Jersey, one high school, and eight feeder schools and put in $15 million of our money to support this. We are now looking at additional schools where we can do this, and we would love to see other businesses join in these public-private cooperative efforts to address the needs of these children.

We're talking about graduation rates of approximately 40 percent or less in these days in a school system that is cheek by jowl right next to New York City.

We have also found a number of programs designed to develop scientific and engineering abilities in minorities and women, including a cooperative research fellowship program for minorities and a graduate research program for women, where we invest millions of dollars to support promising young people.

This week I met with 80 secondary school students from around the world -- 40 from the United States, 40 from outside the United States, in what we call "The Lucent Global Scholars Program." And this is a weeklong summit of extraordinary young people. Of the people in the United States, the average SAT scores for the 40 students was 1,590 out of a perfect score of 1,600. These are extraordinary people. I don't know who the laggards were at 1,480 or something like that.

(Laughter.)

But it was an extraordinary group. Their questions about technology, about life sciences, about deep space research, and as well about business, were on a par with any polished professional, 10, 20 or 30 years beyond their ages.

I had the opportunity to spend a few hours with these people, and it's the kind of a place where you turn out the lights in the room and there's still a tremendous glow. Extraordinary people. But there are not enough of these people, there are not enough being trained, not enough available, to meet the needs to keep this economy growing.

We recruit today at 80 colleges and universities in the United States, and yet even with that we can't find enough people to fill tremendous jobs with great compensation. We have summer intern programs. This summer 1,100 interns are working with us around the United States in R&D programs, and we're looking for more and we honestly can't find them.

Lucent's R&D and personnel needs really drive two broad classes of partnerships with universities. First, we have a variety of R&D partnerships between our Bell Labs researchers and university faculty and graduate students. And, second, we have a variety of partnerships with professors and administrators to help shape and enrich the curricula, enabling courses to better prepare students for productive lives.

Today, our technicians, and our manufacturing associates as well, have high-tech content jobs, and this will only be increasing as what we do in terms of creating technology becomes more and more part of the fabric of our society.

Fundamental research in basic science is one of the traditional strengths in universities, and because it is fundamental research varied companies who may ultimately be competitors can work together to define basic problems and work on similar opportunities.

The nation needs to intensify its investment in this collection of future growth technologies if we are expected to stay strong over the long term.

The vast majority of engineering jobs in a company such as ours deal with subjects that invariably change and technologies that even disappear. That obsolescence will not slow down. What it means is always staying on the leading edge of technology, lest you see your competitive advantage as a nation dissipate.

America's competitiveness in the future will be more dependent on science and technology than at any time in our history, and I believe that this is especially true in the communications industry where information technology continues to advance rapidly.

The challenge for all of us in business and in government, and in academia, is to make our people and our R&D efforts the most productive in the world. If we do that, we have a chance to maintain our competitive advantage as a nation and to create a higher standard of living for all of the people, not only a selective few.

Thank you.

SENATOR FRIST: Thank you very much.

Laura?

DEAN TYSON: Well, I think hearing the perspective from one major company in the United States confirms the evidence we get at the economy-wide level, and the evidence at the economy-wide level is that something dramatic is underway and has been underway in the American economy.

This has been an expansionary period, which has been marked by higher than usual investment, and half of that investment in producing durable equipment going into the information technology industry.

This has been an expansion which largely because of the diffusion of technologies and the creation of new ones has been marked by a much higher, unexpectedly higher productivity growth, which looks to have approximately doubled relative to trend, which is a major accomplishment that anyone who would have predicted that just a few years ago would have been considered to be an extremist.

This has also been an expansion which, because of the flexibility of the new economy, we have managed to enjoy very tight labor markets for quite a long period of time without any of the predicted acceleration in labor cost pressures that one would normally expect.

So all of these aspects of the macro-performance of the U.S. economy I think have been importantly influenced by the continued innovation in the technology area, particularly in the information technology area, and the diffusion of that technology very rapidly.

Now, I am seated next to Larry Mishel, and I'm going to let him do the job I think of making the case that this strong macro-performance of the U.S. economy has been good for the American workforce. It has taken a while, certainly. The progress has been slower than anticipated in many respects.

But I think you'd have to say by now, if you look at things like the pattern of real wage growth, if you look even in the wage distribution, you see that wages are rising in real terms, not just for the top of the wage distribution but now really for the bottom 20th and 10th percentile, areas where for years and years Larry's work and others' work demonstrated that those wages in real terms were actually not just stagnating but declining.

Another piece of evidence that finally we're seeing benefits for the large majority of American workers is that the increasing income in equality, or the increasing dispersion in wages which really was a 20-, 25-year trend, at least seems to have been quelled temporarily.

I don't want to go and say it's been reversed. What I'd like to say is that the numbers for the last few years have taken that trend and stopped it, and maybe we can, if we work with the new economy, make sure that that good news continues for the future.

Now, I think there are some lessons from the current expansion, and also from the experiences of individual countries -- individual companies, excuse me -- which suggest what policies we have to pursue.

The lessons from the expansion, at least from the point of view of the well-being of the American worker, is the most important thing we can do for the American worker is to figure out a set of policies which allows the U.S. economy to continue to run at a high pressure, low unemployment rate, because that is really what has gradually led to real wage increases. That is what has gradually led to a tremendous increase in the number, and now in the quality of jobs -- high pressure, low unemployment rate economy.

The second thing we need to do is to do everything we can to continue to foster increases in the growth of productivity, or at least maintain the tremendous increase that we have experienced over the last few years. And that really comes through adequate investment in human physical and knowledge capital.

And then the last thing I would say is that one of the things that the evidence shows is that the technological change which is feeding the improved performance of the economy also has the character of being what economists tend to call "skill bias." It's not neutral technological change.

It's technological change, which increases the demand for the worker with a Ph.D., for the worker with a college degree, for the worker with a master's degree, much faster over time than the demand for workers who have not finished high school or workers who have just finished high school.

So it's biased in the skill demands that it requires, and that's really what has driven much of the inequality increase on a trend basis over the last 25 or 30 years. And it shows up in things like high school/college wage premia, which I think Larry is going to talk about.

So if those are some of the things we get out of the lessons of this high performance new economy, what are the policy implications? It seems to me that in the long run we know what the implications are. It's education, education, education; training, training, training; wiring, wiring, wiring, because I do think it's very disturbing that 15 percent of the schools in California -- only 15 percent of the schools -- are wired to do this.

So I think we know what the answer is, and I do think there are very interesting private-public partnerships of the sort that Rich mentioned. I know of some that SYSCO is doing out in California, as one company. There are several other companies who are involved in working with high schools, really, trying to identify students in high schools who are willing to start in high school with the math and science preparation that are required to get them into a college degree.

But this kind of training for many of the shortages that we are now talking about in the high technology information industry -- this kind of training has very long gestation lags. I mean, we are talking about people with masters in electrical engineering and computer science. We are talking about Ph.D.s who do Ph.D.s in computer science with an expertise in software engineering.

We are not going to change -- we cannot change -- the supply of such people overnight. We particularly can't change it when the evidence shows that, again, on a trend basis, at least through '98, the number of Americans who are getting B.A.s in engineering, B.A.s in computer science, B.A.s in mathematics, Ph.D.s in engineering, Ph.D.s in computer science, Ph.D.s in mathematics, those numbers were declining.

If you look at the University of California Ph.D. programs in information technology areas, what you will see is a continuation of a trend which has been with us for quite some time, and that is the number of American students on the decline, the number of foreign students on the increase.

Now, we can -- there are many reasons for this, and some of them, unfortunately, go back through K through 12 training, through fourth grade reading scores and fourth grade mathematics scores, and eighth grade reading scores and eighth grade mathematics scores.

By the time you are in high school, you have lost a tremendous number of students who might have wanted to have these kinds of careers. But, again, that's a huge gestation problem. It takes a long time to change that.

So what do you do in the short run? Now, one thing companies do -- and you heard it for Lucent -- is they have to go abroad and find people to fill the gap. There is a gap, and either you find someone or the production doesn't get done. You find someone or the innovation doesn't get done.

So one way to find someone is to go outside of U.S. borders. Another way to find people is to bring people into the U.S. on a temporary basis to fill the gap. And that's, of course, what the H-1B visa is designed to do. It is a three-year visa, extendable for another three years. It is temporary.

We have adjusted it over time, including by the work of the last Congress, to make sure that there are ways to verify that it is not displacing American workers, it's not exercising downward pressure on the wages of American workers, it is filling a gap.

Frankly, it's very important, but it also gets much more political attention than I understood. I wrote a piece for Business Week, which Business Week did not allow me to circulate here. I guess they all want you to go out and buy the copy of Business Week. But I wrote it because I had read a study by a professor at Cal. Berkeley; Annalee Saxenian and she had done a study of the role of Silicon Valley's immigrants to the development of Silicon Valley.

I read the study. I thought it was a very good study. I then read that we had hit the cap for this year's H-1B visas in mid-June. And the cap, as all of you probably know, was raised from 65,000 to 115,000. So we hit the cap.

I wrote a piece saying that it seemed clear, given the projected growth in information technology, given that we had hit the cap, given that this study shows how important immigrants have been to the development of Silicon Valley, that it was time to reconsider the cap. That, in fact, why did we have a process, which set an arbitrary limit?

It was entirely arbitrary. The original number was set in 1990 during a recession in the high technology industry. So the original number had very little sort of basis or foundation, and the new number has very little basis or foundation. It's just a number.

Why do we have a number? This is a temporary situation, and it's a situation in which the companies have to verify that they are filling jobs that cannot otherwise be filled. I think we should really do away with the cap, or certainly extend it so significantly as to make it non-binding in the foreseeable future.

Now, I think one of the things I want to emphasize in closing is that what this study shows very clearly is that if you think about Silicon Valley you cannot tell the story of the generation of wealth, the generation of ideas, the generation of whole new industries, without telling an immigrant story.

A third of the technical, professional jobs in the Valley are filled by immigrants. A quarter of the companies that have been started in Silicon Valley have included a founder who was an immigrant.

The immigrants are not just bringing skills. They are also bringing wealth creation and entrepreneurship. And, incidentally, they are also connecting links back to their countries, so that trade flows and markets and various intricate production relationships which have made our industry the global leader have been developed with the help of these individuals.

So it strikes me that in the short run there is clearly an argument for allowing firms like Lucent to respond to their need through a temporary visa problem, and we should accept their arguments that this need is something that they cannot fill otherwise. And in the long run, moreover, these immigrants who end up staying -- and some of them through a variety of ways do -- have created a tremendous amount of wealth and have become part of the U.S. economy.

So I feel strongly that this is the right way to handle the problem in the immediate future. This in no way denigrates the importance of education, education, education, or training, training, training. After all, I am now the dean of a business school. I certainly believe in education.

Thank you.

SENATOR FRIST: Thank you very much, Laura.

Alan?

MR. KRUEGER: I'll stand over here because I find it impossible to talk without using an overhead now, an effect of technological change I think.

To a large extent, I think my observations will reinforce some of the previous comments.

SENATOR FRIST: Alan, pull that microphone down just a little bit. Again, they are very directional, so --

MR. KRUEGER: Is that better?

Let me begin by saying technological change has been with us for some time. And, in fact, I think it has been skill-biased, as Laura pointed out, technological change -- biased in favor of more highly skilled workers. This is a technical economic jargon. I'd like to think of it as going from the Flintstones to the Jetsons.

(Laughter.)

When you watch the Jetsons on TV, if you remember that show, there were not too many manual laborers. And that's the general direction I think that changing technology has taken since the end of World War II.

Now, there have been some changes in supply, in particular because the baby boom added so many better-educated workers to the labor force that this wasn't always so apparent. But this is something that the U.S. has dealt with for some time. There may be some acceleration the last couple of years, especially in the high-tech sector.

What I'd like to do is go over some of the trends in the demand for skilled workers, and then talk about supply side, how education responds, immigration, and an under-appreciated source -- workers who have severe disabilities who, by and large, have been locked out of the digital revolution.

This figure documents that between 1984 and 1994 the occupations that have grown the most are those that require the most skill. This simply divides all of the occupations in the U.S. economy into three groups based on the average education level in the occupation. And you can see the fastest growth has been in the high education occupations.

Moreover, that's what is predicted to continue for the next decade. The Bureau of Labor Statistics makes projections of occupational employment. The last one they did was for 1996 through 2006. And you can see that professional occupations are predicted to grow at 27 percent, compared to 14 percent employment growth overall.

Moreover, if one looks at the top three fastest-growing occupations, BLS believes they are all in the computer sector. Database administration, computer support, computer scientist, computer engineer, systems analyst -- all predicted to grow by over 100 percent.

Interestingly, most of the rest of the top 10 is rounded out by health care services. That's the second main user of the H-1B visas.

In terms of the accuracy of these projections, by the way, about 80 percent of the time the BLS gets the direction right if you compare their 10-year projections. So it's difficult to predict what's going to happen in the future, but I think it's a fairly safe bet that there will be increased demand in the high-tech fields in the next decade.

As for evidence that this looks like it's being driven by demand, let me show you the following. This shows you on the left the total share of hours worked in the U.S. economy worked by college graduates, in 1980 and in 1994. And in this 14-year period the total share of hours worked by college graduates increased from about 20 percent to almost 30 percent of all of the hours worked.

So at the same time that we've had more work being done by college graduates, their relative wage, which is on the right here, which Laura alluded to earlier, has increased to historically high levels. This suggests that in spite of the fact that we've had an increase in relative employment of well-trained or better-trained workers, demand has far outstripped that increase in supply and that's what's causing wages to rise.

It does appear, however, that the U.S. supply response, in terms of educational payment, is pretty flexible, pretty responsive. This figure shows you each year, from 1967 through '93 -- if you continue this to the present it looks similar -- the fraction of 18- to 20-year olds who are enrolled in college -- that's what the solid blocks are -- as well as the rewards of going to college, the premium for college graduates relative to high school graduates.

And you can see while we've had this sharp increase in the payoff to attending college, we've also seen a big supply response. We've seen record levels of students enrolling in college. In 1997, which is the most recent year for which data are available, there were 9.2 million 18- to 24-year olds who were enrolled in college. That's 400,000 more than in 1996 and both of those are at record levels.

So to give you an idea of the magnitudes, that's quite a bit larger than what we're talking about when we talk about H-1B visas, where the caps are around 150- and 120,000 or so.

H-1B visa -- by the way, the requirement for an H-1B visa is a bachelor's degree or higher or equivalent experience. So I think at one level it does make sense to compare them to the pool of college graduates. I'll come back to that in a moment.

Nonetheless, we've had this rather rapid increase, and when one looks at graduate school enrollments one sees the same type of pattern -- increased enrollments. The increase in college enrollment, however, did not occur across the board. It has mainly been a phenomena for middle class and higher income class families.

The lower income families have not displayed an increase in enrollment rates, and government policy the past few years has focused mostly on the middle. If you think about the tax breaks that were given for college enrollment, those are going to be of less value to the families at the bottom of the income distribution who have been lagging behind in terms of college enrollment.

Another indication of the responsiveness of the workforce to changes in demands, changes in what's being done at work, is the fraction of workers who use a computer. This shows you, from 1984 through 1997, the percent of workers who use a computer directly at work. And that increased from 25 percent of the workforce to a little bit over half of the workforce -- a rather remarkable shift in the way that work is done.

Most of the people who use a computer at work learn how to do it on the job because most people were already through with school by the time the computer revolution occurred. And it's also the case, as has been mentioned that people who have the types of skills to use new technology tend to do better in the labor market.

I think -- and I agree with Laura on this -- that a strong case can be made for raising the skills of the workforce, and, in particular, looking at -- looking towards immigration for doing it. Demand has been shifting in the direction of more highly skilled workers. Supply has been adjusting, but it is, not surprisingly, slow to adjust.

I think there is also something somewhat unique about higher technology. It's much more important to be first, I think, in the communications revolution than it is in other industries. This is what economists often call "network externalities." The first company to establish the standard has a big competitive advantage, and I think that carries over to the nation as a whole. And I think there are advantages to trying to use immigration policy to address this issue.

But I also think that once you think about workforce demands in the context of overall immigration policy, not just H-1B visas, and in terms of overall immigration policy the U.S. has been moving more and more towards less skilled immigrants.

This figure shows the difference between the average education level of recent immigrants, those who have been in the country for five years or fewer, and natives. And you can see in 1960 the average immigrant had almost half a year more of education on average than the average native. By 1990, the average immigrant is trailing the average native by 1.3 years or so. There has been a rather rapid shift, large shift, in the skills of immigrants.

Now, this is based on Census data from a study by George Borhaus. We don't have the year 2000 Census data. The early indications suggest that this might be beginning to improve a little bit, but it's still the case that newly arriving immigrants on average have lower skills than the native population.

And I think it makes sense to think of the demand for more highly skilled workers and the demand for higher technology workers, in the context of the overall immigration policy.

Now, H-1B visas is one way of addressing the skill shortages. On the other hand, I would be cautious about expanding the program without knowing more about the way it works. It is kind of amazing how little we know about the H-1B visas. Let me give you some examples.

We don't know where the workers who get H-1B visas are working. We don't know what sectors they are in. We do know from the Department of Labor, which does the initial check on the applicants, that about half of those who are applying are in computer-related fields.

On the other hand, the number of applicants greatly exceeds the number of immigrants who are coming over, not necessarily because of the caps but because companies apply in order to be in the queue, and then they determine whether or not they actually are going to need the workers down the road.

So there may be three or four times as many applicants as there are actually bodies who are being processed or sent over to INS.

We don't know of those who actually come what their skills are. We don't know how many of them are the Ph.D. types, very highly trained, which -- and I agree with Laura -- it certainly takes a long time to get them through the pipeline, or how many are more run-of-the-mill types of B.A. or equivalent type workers.

I don't think we know very much about how many Americans are available to fill the types of positions that are being filled by those on the visas. We don't know how much they are being paid. We do know, for the applications, what the employer says the prevailing wage is, and those tend to be fairly middle class salaries.

We don't know how fast wages grow for those who come in after they're here. We don't know how much training they're getting. And I think of most concern, one should be concerned that employers have more influence over those who are here on temporary H-1B visas than they do over the typical American worker.

I presume that most of those who come here on H-1B visas are hoping to get green cards. In order to get a green card; they would need their employer to sponsor them. This gives the employer a lot more influence over them than the employer might have over other employees, or at least that's a certain that I would have.

It seems to me that these are the kinds of questions that are very well suited for the GAO to study.

Now, I think there are some ways of strengthening the H-1B visa program, which I would think would make sense if we are considering expanding it as well. For example, the visas last for three years and they can be extended another three years. One can consider after three years giving the employee the option of changing employers.

The employer won't have to -- the new employer wouldn't have to go through the whole visa application process. I think that would give the employee more mobility. I think it's probably also economically efficient. It may be that his skills are better used elsewhere in the economy, and I think that reduces the ability or the risk that employees might be exploited who are here on H-1B visas.

Some other possibilities would be to target the visas more to the areas where they're in demand. And, again, we don't know. We don't know how many visas are going into high-tech or into other fields. And another possibility would be to add a labor market test at the front end.

As I understand it, when employers apply to the Labor Department, there is a one-page form, which essentially asks them what the prevailing wage is. It's not quite the same type of demanding process that one would need for a green card, to demonstrate that there are no other Americans who could qualify for the position.

Now, I would personally prefer to think about this issue in terms of overall immigration, in terms of permanent immigration, and perhaps expanding the number of permanent immigrants with higher levels of skills.

Now, the green card process is much slower. It could take one year, up to four years or so, to fill a position. That's not going to help a company in an area, which is changing very quickly. So one way to get around this would be to try to expedite certification under the green card process. Also, to target them more to fields that are in demand.

There is a kind of curious way we do immigration in the U.S., because the H-1B visas, which are temporary, are not based on current demand being high necessarily in the process for application but the green cards- to get a green card it’s necessary to establish that there is no other American who could fill the position--

(End of Tape 1, Side A. Beginning of Tape 1, Side B.)

MR. KRUEGER: People with disabilities tend to be much less likely to use computers than those who don't have disabilities.

These data are from a study I did with Doug Kruse, who is a professor at Rutgers, of people who have spinal cord injuries. And we interviewed individuals who had a spinal cord injury and someone who they used to work with prior to their injury, and you could see that those who have spinal cord injuries, 41 percent of them use a computer somewhere -- at home or at work or at school -- compared to 54 percent of those without spinal cord injuries.

Those without spinal cord injuries I mentioned was somebody who we paired them up with, who they nominated, someone who they thought was like them prior to their injury. This gap is about as large as the racial gap in use of computers, and I think there's a straightforward explanation for it.

Only 20 percent of those with spinal cord injuries, with severe injuries, are employed full-time. Only an additional nine percent are employed part-time. Since most people learn how to use a computer at work, these individuals were not at work when they had the chance to learn how to use a computer.

And the results -- if we look at other severe disabilities, the results are pretty similar. I think the spinal cord injury group has been studied the most thoroughly. And it's a real tragedy because people with disabilities are the ones, I think, in American society who would benefit the most from this form of technology. It will help them to overcome the disabilities that they have.

And this final slide shows how that seems to be the case for those who have spinal cord injuries. Here we looked at the average weekly earnings for people who don't use a computer, those with spinal cord injuries and those without, and there is an enormous gap in terms of their weekly earnings -- an almost $200 gap per week.

And then for those who do use computers, the individuals with spinal cord injuries actually earn a little bit more than those without them. And, again, this is using the paired sample, where we matched individuals with their former co-workers as a way of trying to control for other differences between them.

So it seems to me that there are some other sources that we could use in addition to immigration, and I think it makes sense to think of increasing the skills of the workforce through immigration in terms of permanent immigration.

SENATOR ROCKEFELLER: Okay. Larry is going to come next. But first I have to make an apology for Senator Frist and myself. We are here totally bipartisan. We take no point of view on anything. We moderate, we facilitate, and we bring everything out.

However, we are now going down to do absolute pitched battle against each other on a Medicare prescription drug benefit.

(Laughter.)

So we'll both be gone for a few minutes, and be tolerant. In the meantime, you'll do much better listening to Larry.

And Peter will handle the questions until we get back, if we get back.

(Laughter.)

MR. MISHEL: Well, let me say that I'm for a low unemployment, high-pressure economy. I'm for science and technology development. I'm for more education and training, and I'm for wiring schools and community centers and every place we can through a public investment program.

I remain a skeptic, however, about raising visas for -- to solve the IT worker shortage. It seems to me that what we hear a lot is basically, "This is a problem. It's the fault of the schools. It's the fault that the workers don't have the right skills. Maybe the fault of different kinds of government policies." But we don't hear anything about what the companies themselves have been doing for 10, 15 years.

And it's my belief that the industry itself, in how it develops its own human resource systems, is largely to blame for the current shortage that it faces.

Now, let me start with the overall picture, at least to state my view on it, of, you know, is there a skill deficit, a new economy, and all of that sort of stuff. I guess my view is that we don't have so much of a skill deficit or an education gap, but we do have a wage gap, and we have a wage problem.

And that wage problem arises because we have managed, in our society, to rearrange who gets what so that people who don't have a college degree get less. And people who get a college degree have not actually done that fabulously well, except for the last few years. They have just sort of improved a bit.

So I don't necessarily buy the New Economy interpretation of the labor market trends or that there is really a strong excess demand that continues to grow for more skilled, more educated workers. At best, that was a story for the 1980s. It's not a story for the 1990s.

But let's just go down to the level of the IT workers, and I want to present some information about that.

There is no doubt in my mind that there has been a very strong employment growth in engineering, in information systems, in computers, scientists and programmers, and of that nature. That would lead one to think that there is a very strong growth in demand, and I think there has.

My skepticism about this story is what I've seen in the wage trends over the 1990s. So let me bring them on.

These are the wages of workers, who are age 23 to 29, have a four-year college degree. The guys are in red and the women are in blue. So these are the young college graduates and the kind of earnings they had when they left college.

Now, what's kind of interesting, when you look at the 1990s, or when you look at the red line, the men, you'll see that the wages for new college graduates was actually falling since the mid 1980s, and fell precipitously up through 1995, around 10 percent or so.

Okay. We've seen in the last few years, with the period of low unemployment, it has gone up a lot. But there's a lot of explaining we have to do about the early '90s.

Even among women -- and women college graduates have been sort of the premier earners over the last two decades. That's one group that has fared very well. Among young women just coming out of college, getting a job, even their wages were sliding downwards in the early 1990s up through 1996, but in the period of low unemployment have gone up.

Now, you should note that throughout the 1990s the unemployment rate for college graduates was probably around three percent or less. It has never been, you know, a high unemployment area. So what we are observing is that even with relatively low unemployment workers were not able to press their wage demands with employers.

But you might say, "Well, this is college graduates. That includes all of these history majors and political scientists," and all that sort of stuff, not the business school people, not the engineers. But we can look at the occupation distribution -- and this is a graph where it's an index, so 1989 equals 100, meaning when the line goes down they are doing worse than 1989. When the line is going up, they're doing better.

So the red line is for everybody, and you can sort of see that young college graduates, which here, by the way, is defined as people with one to 10 years' experience to get the sample sizes I needed but -- so I could do an occupation. But you can see that the green, which is the math and computer science, I mean, their wages were doing as poorly as all young college graduates, up through 1995-'96, and, in fact, aren't doing that much better since.

If you look at engineers, which is a story we haven't really focused that much on here, engineering wages have been falling for a long time, and they're still not even back up to their 1989 level. So there is a pool of people with technical skills that you would think you'd be able to draw on.

Now, another source of information is, what are employers bidding for college workers as they come out, based on the surveys done by the National Association of Colleges and Employers? And these are three occupational groups that are tracked -- systems analysis and design in green, computer programming in red, information systems in blue.

So there's two -- what we've been talking a lot about is the upturn and haven't talked at all about is the downturn. And both are true.

If you looked at engineers, engineers is actually a much more pessimistic graph than this about wages. It would be driven down further and would have risen far less.

So how would one interpret all of this? Well, one of my problems with the story that's told is it goes like this. We are the technology industry. We are really important. We have driven the new economy in the 1990s. We lead the country. We lead the world.

We need some workers, very, very highly skilled, highly educated workers. And our choices are the following: either you let us bring people from overseas here, or we'll have to take the work somewhere else. But the option of we need to improve our working conditions, our wages, our benefits, seems like it's written off the table. We can't do that because it's a global economy.

Now, when I've talked to my friends that sometimes I debate on trade issues, I say that I don't know why the free traders among the economists and everybody else does not talk loudly and say, "Hey, high-tech, you're giving globalization a bad rap." I mean, how can it be that a global economy means that we can't have fast-growing wages among people in the leading industry with the most skills?

If that's true, if people, ordinary citizens hear that, why are they to believe that somehow globalization is going to improve their well-being, when, in fact, the most privileged, you know, education elite, technical elite were told, "No, there's too much global competition; we can't raise wages now."

It is true that wages have been going up of late, but it also seems to me that the companies, when they could get away with it in the early '90s, and for periods probably before that, they were not paying a premium, and they were not improving the distance between these kinds of workers and other workers.

And so those people who want to live by the market die by the market, in my view. The industries -- not Lucent, I don't think, but I think there's a -- well, Lucent has a longer history and comes out of a different tradition.

MR. McGINN: Yes, we're three years old now.

MR. MISHEL: Yes. Well, they were --

(Laughter.)

-- somewhere else before they were Lucent, attached to a long-established company.

You know, there is issues of an industry with a spot market mentality that wanted to hire people temporarily, wanted to hire people as independent contractors, who are willing to attract people by giving bonuses.

But, you know, when you hire people with a bonus, an economist interpretation of that, or even any human resource professional, would tell you, you give a bonus to the people you're hiring because you don't want to have to give more to the people that have been there two, three, five, 10 years. Right?

I mean, if you really wanted to attract wages, you would be raising the salary structure in its entirety, and that would be a good indicator of the degree to which there is excess demand, in my view.

Now, the way that you -- so I would like to call attention to the need for firms to develop what we call "internal labor markets," mechanisms to bring people in at entry-level wages, develop their skills over time, to think about how they're going to use, you know, older workers, disabled workers. And I fear for letting an outlet through the immigration is -- in fact, will not facilitate companies doing that.

Now, last I guess there has to be an explanation about why we don't think the market is going to work to solve this problem. Alan presented the information that showed that people respond to price changes in the wage structure to go to college.

From what I understand from people that are in universities, there is now a rapid growth of people studying, you know, computer programming, computer science, etceteras. Those people are going to hit the market soon. It doesn't take -- we're not talking about mostly Ph.D.s. The IT worker shortage is not about Ph.D.s, I don't think. It's mostly about college graduates.

I'm sure there is some small problem there, but it's not in the 60,000 a year, you know, doubling the H-1B visa arena.

So I guess it seems to me that I don't want to solve this industry's problems by letting them an outlet through the immigration when I think there is a lot of internal thinking they need to do to be able to dramatically improve the position of the workers they hire.

Thank you.

MR. ROONEY: Thank you very much.

We'd love to see people come to the microphones. We've gotten some of your questions on cards. This issue is always highly polarized, and we see that in the questions that some of you are asking.

I think this is a question that's probably directed to Rich McGinn. The question wants to know: what is the role of certification versus the role of traditional degrees for information technology technical professionals? And would Americans benefit from more certification assistance?

MR. McGINN: Peter, I'm happy to answer that. But since my microphone -- my directional microphone is tilted down, I have to make a preamble comment following up on Larry's comments.

I felt as if for --

MR. ROONEY: Please. Go ahead.

MR. McGINN: I felt as if for a few moments I was watching a Fellini movie, quite honestly. Just to give you a point of reference, Larry, and I'm sure that data is statistically valid, to a very great extent -- and that is not a statistically valid comment -- to a very great extent, we are not raising salaries at all. We're not adding to salary grades.

Most of the increases in compensation are taking place in performance-based bonuses and stock options. And when we talk about higher salaries to the young people today, they don't want it. They want a good salary, but they would prefer to take whatever you're talking about of salary increases and leverage it into performance-based options with up side based on their performance, and options in the company.

So total compensation costs to us are up dramatically, and it is not simply at the engineering entry level for people with B.A. degrees in engineering.

To give you an example -- in data networking, our sales force in data networking is almost all electrical engineering or master's in engineering for entry-level sales jobs. And they are on a compensation plan with options, so you don't get a chance to see the full impact of their wage value based upon their performance.

I think that focusing only on salaries does not give you a true picture of the cost of what's going on here.

Now, having made my preamble comment, to a very great extent we are looking at traditional degreed individuals who have taken certain core curriculum courses, and it is not because we don't do training -- we spend almost a billion dollars a year on training our own employees inside the company -- but, in fact, we are looking to get people who have already demonstrated competence by virtue of their -- the courses they've taken and, as well, the grade point in those courses.

And frequently in the United States we even go to specific professors who have a track record of having counseled and mentored strong students, and look for the best of the best in that department in that particular expertise to recruit that person in.

And we wind up giving people, as I said, even out of college at a university at the master's level, sometimes excusing their college loans as part of the hiring program to get them on board, because that's how intense the hiring process is.

MR. ROONEY: Larry, do you have any response? Before we move on, do you want to take a shot, or shall we go to the next person?

Any of the rest of you on the panel, feel free to jump in here.

MR. MISHEL: Well, one, the options and stocks don't really I think do very much to explain the wage decline of the early 1990s. I don't think people were asking for lower wages in order to get stock. They may not have -- you know, maybe the -- the anecdotal evidence is what we're hearing. I guess I'd like to see more than anecdotal evidence about that and how much that really adds up.

And the performance bonuses -- when we look at the economy-wide numbers, or even by industry, and you add up all of these other kind of bonuses, such as in the employment cost index, it really doesn't add up to a very large part of the compensation package. Maybe it does at particular companies, in particular jobs, at a particular time. But I'm not sure -- I guess I -- you know, I remain a skeptic, and I'd like to see more evidence on that.

MR. ROONEY: Rich, I'd like to just follow up on the questioner's question, because you folks are not just recruiting the top level scientists; you've got manufacturing plants where you are bringing in assembly line workers who also have to be technically literate.

I mean, when I was working for Senator Lieberman, I visited one of your plants in Pennsylvania, and you folks did a marvelous job in taking the sons and daughters of steel workers and really turning them into very competent, very empowered technical professionals. And I'm wondering whether Lucent can make more use of certification programs for those types of positions.

MR. McGINN: When I said "traditional degrees," I didn't only mean four-year degrees. A lot of our manufacturing associates and our systems technicians come in with two-year degrees now. And so I consider them traditional degreed individuals, and that gives them a tremendous leg up on coming in and getting good positions in our manufacturing facilities or in our installation and technical support organizations, which are so key for us for delivering product to our customers.

So we see that as a normal part of the progress here, and one that's really indispensable for us.

As far as conversion of others, we do have programs in the company for trying to train workers to take on new jobs as technologies change. We think that's a natural and normal approach, and one that has to be done. But, as well, we need other sources of input that are already degreed and then we can build from that point forward.

MR. ROONEY: I don't want to seem as though I'm picking on you, but I do have another question from the floor for you. When you bring H-1B visa workers into Lucent, where do they go?

MR. McGINN: We have -- out of the 120,000 plus employees of Lucent in the United States, we have 1,200 H-1B visa holders, or one percent of our U.S. population. So the numbers that are outside the U.S. that are technical dwarf that particular number. The vast majority of them go into our Bell Labs operations.

MR. ROONEY: Great. We do have a question from the floor.

AUDIENCE MEMBER: Good afternoon, everybody. I'm Dr. Ronnie Lowenstein. I have two hats. I represent Congressman Major Owens from Brooklyn, and I'm also the president of a nonprofit called The Education Technology Think Tank. And we promote technology for the underserved communities. We'll be doing a two-day technology forum at the Congressional Black Caucus in September. It's the third time we've done it.

Several things were brought up that are of interest to me, and I'd like each person, if at all possible, to respond. And which -- and that deals with addressing the potential population in the underserved, including the disabled. By the way, there's a community options, Microsoft-supported program, that focuses on pass it on with disabled and IT.

And I want to -- I'm going to ask about how the economic and the education policies can address the underserved. And Secretary Daley has talked, and he physically said, "We have this great need here, and we have the population that doesn't have jobs and is unskilled here. Let's bring it together." So I'd like to know specifically what -- how you think this can be done.

And before that, I just want to thank you. Lucent is the number eight company in -- Fortune magazine listed the 50 top companies that are addressing diversity and the -- Lucent is number eight. And I've spoken to you. I called up your Sue Ann Roth, and I just think that you're doing a wonderful job.

So in addition to sort of responding how you all think we can bring the economic policy and the issues of diversity with the moral imperatives together, we have here some -- a company that is doing it. And my hat is off to you.

MR. McGINN: You're very kind. Thank you.

DEAN TYSON: I can start, if you -- I don't -- what I tried to emphasize in my remarks is I think we have to distinguish cyclical phenomenon from long-term phenomenon. And I get to the education issue that way.

One of the things that Larry pointed out was that real wages were declining through 1995, and so maybe this is -- that is partly -- and engineers did go through a sort of very -- a surplus period in the early 1990s, in part because we had the layoff of lots of people, or the movement of lots of people out of defense, and because engineering is not engineering is not engineering, because in all of the discussions here we're sort of amalgamating things like IT workers, which include two-year associate degrees from community colleges and Ph.D.s in electrical engineering. These are not the same people. They're not the same skills.

So when you ask the question about how to educate people, and sort of policy for this, I mean, really, there is an issue of education to what level. I emphasized in my remarks the -- probably because I was very influenced by this study, which I think is a very good study -- the immigrants that were discussed in the study were immigrants who basically were doctoral-level people. These were master's and Ph.D. people.

They are people who would take our society, takes -- the gestation period is at least five years, if you have the B.A. And to get into an undergraduate program which gives you the skills required to get a Ph.D. in electrical engineering at one of the nation's major universities is -- you probably have to have a very good reading score when you're in fourth grade.

So I was trying to look at a very high skill area where, in fact, the education policy has to go way back in through -- to K through 12, has to ask the question, which Alan Krueger mentioned in his comments, which I think is very important: what can we do to change the reality that the percentage of students enrolling in college varies dramatically by the income of the family?

And the most disturbing numbers about education, from my point of view, are not that -- it's good news that we have an increase in the number of students who are going to college. But when you look at it and see that the increase is not coming from the bottom 20th percentile of the income distribution, it's not coming from the bottom 30th percentile of the income distribution, then you see there is a real structural problem.

And it goes to the schools in the communities that a lot of these families live in. And I think that that's why, in a way, while I can on the one hand say, you know, we definitely need to do something -- we definitely need to ease the cap on H-1B visas because I don't see it as a large enough issue to have the influence on wages that Larry is talking about.

I agree -- Larry's data is right. It's absolutely right. I have no disagreement with that data. I don't think H-1B visas have anything to do with it, either on the down side or on the up side, frankly. But I think it is the case that for the long term we have to worry about the K through 12 education issues.

Now, do I have a panacea for that? No, I do not. And I think what we have is a number of examples, including what Rich talked about at Lucent, what I know is going on at SYSCO, where companies are trying to go in with either community colleges or local governments, and identify kids in middle school and work with them up through high school to the point where they might then choose the community college option, which they wouldn't have chosen before, the B.A. option, or, in some rare circumstances, will actually get into advanced degree options.

That's the most important thing we can do long run, I think.

MR. ROONEY: Does anyone else on the panel want to comment? Go ahead, Al. Please, Alan.

MR. KRUEGER: Thanks. Let me start with the disabled. I mentioned them in part because I think public policy has had an impact on whether the disabled work. There are very strong disincentives for working for someone who is on disability insurance or receives private disability insurance.

If one goes to work, one would lose Medicare, one would lose disability benefits. Plus, if technologies change in the meantime, it might actually be worth it for the person to go to work, if the person had the skills. But the only way you can get the skills is through work, to some extent. So there is a catch 22.

I think one -- and I think public policy can address that. One possibility would be to try to encourage rehabilitation, and to reward the rehabilitation centers. Senators Kennedy and Jeffords have something called "Ticket to Independence," which I think is based on work by Monroe Berkowitz and Rutgers, where the basic idea is that if you take someone off of the -- if you take someone who is on disability insurance, who has an extraordinarily low chance of working subsequently, which is the case for almost everybody on disability insurance, and then if they go through a rehabilitation center which places them in a job, then the rehabilitation center would get half of the savings on the disability insurance benefits for up to five years or so.

And I think one could also make changes to Medicare to make it possible for them to maintain their coverage if they go to work and are not provided with coverage through an employer.

So on the disabled I think there are some things that public policy can do in addition to training, which is not just specific to the disabled.

In terms of more general education, I think one of the things we've learned is we have to start early, that investments in education seem to have the biggest payoff at the lowest grade levels. So I would think about focusing on the early years.

For the college attendance rates, I think that's in part, as Laura pointed out -- in part has to do with inadequate schooling background, and it also has to do with money and access to higher education.

One last point I was going to make, I was going to comment on one thing Laura said -- that H-1B visas are too small to influence the wages. If that's the case, then it's not clear what the benefit is either. You know, if they are too small to have an effect on the wages, then it's not clear they're going to have much economic effect at all.

I suspect that the immigrants who are starting up -- the startups are not on temporary visas. It's probably more likely they are permanent.

DEAN TYSON: I agree with that. They're ones that either were permanent or may be later.

MR. McGINN: Peter, if I may?

MR. ROONEY: Yes, please. Go ahead.

MR. McGINN: Three quick comments on your multiple topics. First, the underserved -- it's a special area of interest. In fact, my wife is working for the past six months out of a joint venture with a new communities organization in Newark, New Jersey, and had been working with another firm to try to set up a joint venture to deal with getting training and jobs in high-tech for disabled individuals.

It's, in a sense, a bit of an ironic situation in that the very technology we produce actually has the potential of getting them more opportunities to be well-served, if you will, to get full employment. Just the notion of help desks and call center support, and that software is in place at broad band networks, you don't have to go to a physical location.

You can log on and log off from your location and perform the job very well without the cost of a van and extra support to move around, should you choose to do it that way. You may want to be out of the house. That's a separate issue. But there are lots of opportunities to do that, enabled by the very technology.

And, as well, she's working with the State of New Jersey disabled organization to see what government resources can be brought to bear to help these fledgling ventures to get off the ground.

Secondly, as far as diversity is concerned, just a generalized plug, we think of it as a competitive weapon or advantage. That may sound counterintuitive, but we have a belief that different people with different ideas, even different passports, both genders, races, and different experiences, put together in a soup actually deliver better results, higher performance, faster than terrible sameness of, if you will, an all white male organization that has had, to a great extent, similar experiences over time.

So we believe it actually helps us to do things rather than it being something we're doing for a moral imperative, or, in fact, something that is a disadvantage. It has the benefit of being a good thing and the right thing to do, but primarily, as capitalists, we're driven by the value that it brings for us.

Lastly, I should tell you that as far as the educational system is concerned, and kind of a post-script comment to Laura, I met recently with a number of students from Newark, New Jersey public schools who are part of this Project Grad Program, a K through 12 program. It starts out at kindergarten, to work their way through.

There were about nine children ranging from juniors in high school down to about fourth grade. The most eloquent of those in the group, the juniors, we were asking them why it's important, and they summed it up very well. And all of these children were African-American children.

One young man who has a 3.9 grade average in the Newark public schools said, "Well, I can tell you, it's something that I've always been excited about. But the fact of the matter is, it's all about the money. The only people in my neighborhood who have money are the people who sell drugs. Most of them are dead already. I think I've got a better chance of getting out of here if I do this, do it well, take your scholarship money, somebody else's scholarship money, get on to school, and make a good living for myself. Then I'll have the money."

Now, we may think that's not a high moral imperative perhaps, but I can tell you what: this person is going to succeed, because he -- or others, she -- see a way out of the cycle of despair that they're facing, and they're going to go for it. And if we had more and more of these efforts going on, we'd have more and more success for a different group of underserved people.

DEAN TYSON: Can I say one clarifying thing?

MR. ROONEY: Go ahead.

DEAN TYSON: I just want to say one clarifying thing about immigration and whether it has a big effect or not, because I do want to agree with Alan that my view about immigration is we should be thinking about focusing on permanent issues and how we change the skill mix of permanent immigrants.

My view of the H-1B visa issue is, yes, the effect actually is small, whether it's on wages or on anything else, because the number of people we're talking about is, you know, a tenth of one percent of the professional population.

So I don't -- I think it's unfortunate, in a way, that the H-1B visa issue gets the kind of controversy and political attention that suggests it's a major factor one way or the other. The issue really is: long run, what kind of permanent immigration policy do we follow? And what kinds of skills policy do we follow within that?

But I don't see the down side to having the H-1B visa in the meantime be a cyclical outlet which allows companies to deal with shortages at home rather than search for labor abroad.

MR. KRUEGER: I agree with that completely. I would just think there are some protections that might help with the H-1Bs.

MR. MISHEL: Well, Laura, I'm not sure that makes total sense to me, because if there is not a substantial shortage measured for the sectors involved, or the occupations involved, then we wouldn't -- you know, there wouldn't be this screaming about it. There wouldn't be all of this public policy attention.

And it seems to me that the visas, you know, very well may have an effect on the sectors involved. And it's not about professionals all across the board and --

DEAN TYSON: No.

MR. MISHEL: -- but about, you know, the sectors involved. And, you know, it can't both be true, that there's a big problem for those sectors, and that if we deal with it it won't -- it might not affect the domestic employment and their wages.

MR. ROONEY: We have just time for one more question. We've got one from the floor here.

AUDIENCE MEMBER: Yes. I'm Duncan Moore from the Office of Science and Technology Policy. And one of the issues I don't think has been talked about here at all today -- I know we are really on the H-1B visa issue and IT workforce issues, but I'm much more concerned about K through 12 science and math teachers.

Nobody has even mentioned that, and the reality is that in the next 10 years we are going to retire two million teachers of a three million teacher workforce. That's two-thirds of our teachers. The supply side of teachers right now is 100,000 per year. So anybody can multiply 100,000 by 10 and come up with one million, and find out we've got a shortage of about one million teachers that's coming through.

And the interesting point is isn't -- when we look at, you know, social studies teachers, because there are plenty of social studies teachers -- in fact, there as a wonderful article I'd refer you to in the Post last Sunday on the whole issue of recruiting teachers here in the greater Washington area.

But the reality is we have a two percent unemployment rate among those with a college degree in any field. It makes no difference whether it be religion or engineering or science, or whatever. It's only two percent, and it has been two percent for the last 24 months, within two-tenths of a percent.

So the question is: if there's a shortage of science and math people, which all the industry wants, how are we going to encourage young people to go into teaching at K through 12? Because the failure to do that is going to have a huge implication for us in the year 2010.

I know we're all bemoaning the fact that the science and math scores of our students, you know, fourth, eighth, and twelfth grades are going down. They're going to continue to go down. That is the reality until some sort of intervention occurs which makes some changes in science and math education in K through 12.

And that's why some of us at the White House and some of us through Senator Frist's office have been working very actively to find ways in which local companies and local school boards will jointly recruit science, math, and technology teachers as a local phenomena.

That is, a perspective teacher comes in, is interviewed by the School Board, is interviewed by a local company, and the company guarantees the teacher a summer job, because one of the problems we know in science and math teachers, and teachers in general, it is a nine-month contract. It is not a 12-month contract.

And we need to find better ways to get some of these good people into science and math education in K through 12, because if we don't solve that problem, the workforce issue problems and IT is just going to go on and on and on, and we're going to have a very big problem.

Thank you.

MR. ROONEY: My guess is everybody here agrees with that. Does anybody want to jump in with a wrap-up comment?

MR. McGINN: As large a problem as you state, I think you understate it. I think it's even more profound than that because the opportunities for these same people who have those kinds of skills that would be relevant to the TIMMS data that shows how we're doing versus other countries around the world, their job opportunities are so strong that they can earn twice as much money by either not going into teaching or leaving teaching to get jobs in industry.

And I think it's going to get worse rather than better, and the market forces may not be enough to deal with that, as you were talking about, of attracting people with a full-year job. I think it's a profound problem. I agree with you.

MR. ROONEY: Well, I want to thank the panel. I think this was an outstanding group here today. I want to thank all of you for coming.

(Applause.)

We're going to be taking a break in August, but we'll be back with you in September with a fall lineup. And we'll fill your mailboxes and e-mail boxes with plenty of announcements, and we'll look forward to seeing you back here again in September.

Thank you.