 So, for those of you who are less familiar with EPI's work, we are a think tank that delves deeply into issues regarding the U.S. labor market and in particular what the available evidence shows about what's going on with workers in the United States. That leads me to why we're here today. I'm honored that we have Michael Tidalbaum here. He's written a new book, which is about the Science and Engineering Workforce, and how it came to be, how it's evolved over time, and what its prospects are for the future. I should also mention, after the talk, we'll be selling the books at a discounted rate in the back. And I guess before I say anything else, I'll just say a couple of words about Michael's biography. Michael is a Senior Research Associate at the Labor and Work Life Program at Harvard Law School, and was also the Vice President of the Sloan Foundation for a long time, I believe. He's a demographer with research interests that include the causes and consequences of very low fertility rates, the processes and implications of international migration, patterns and trends in science and engineering labor markets in the U.S. and elsewhere. And he's the author or editor of 10 books and a large number of articles on these subjects. Previously, he was a faculty member at Princeton and Oxford and served as a vice chair and acting chair of the U.S. Commission on Immigration Reform, which is also known as the Jordan Commission, named after the late congresswoman, Barbara Jordan, congresswoman and civil rights leader as well from Texas. And I guess I should just say a quick sentence or two about the Jordan Commission because I think its work was so important and we just probably have short memories and don't talk about it very often anymore. But the Immigration Act of 1990 created the commission and mandated it submit an interim report in 1994 and a final report in 1997. It held public hearings, conducted fact-finding missions and expert consultations in order to identify major immigration related issues which were facing the United States at the time and to propose solutions in those reports. The membership of the commission was distinguished and had an impressively brilliant staff. They had funds and worked for five years and amazingly almost all the recommendations that they published were unanimous. I think many of the recommendations are still instructive today but that's probably a discussion for another day. But suffice to say that such an august body like that would benefit the United States now in terms of figuring out the best ways forward in terms of immigration because our discussions and debates often are lacking evidence. Now back to why we're here, the science and engineering workforce and the question is the U.S. falling behind. The part of the workforce that is actually made up of science and engineering is actually quite small in terms of the overall workforce but nevertheless it's a very vital part of that workforce. It's important to innovation, the advancement of technology and medicine and even national security. There are a number of actors involved with a number of their own interests. These include universities, software companies, immigration lawyers, the government and the defense industry and that's just to name a few. Perhaps we should actually be a little bit surprised that no one has actually compiled all the disparate pieces of evidence about the science and engineering workforce and put it all in one place but luckily Michael has done a wonderful job of dissecting this for us including the actors, the arguments, the evidence and studies and also the spin. He does this with an amazing clarity and in simple enough terms that a non-policy wonk can understand easily but it's also so stuffed with great historical information, careful research and it's extensively footnoted and it's incredibly fair to all sides. As a result it has a lot to teach even the most veteran policy wonks among us. I'm tempted to say more about the book but I won't because I'm not gonna do a better job than the author will but after Michael's presentation there will be two great panelists as well who will be commenting on the book and after that we'll get into a discussion and a Q and A with the audience so I'll just say a quick bios of the other panelists. To Michael's left is Bob Charette, Robert Charette. He has nearly 40 years of experience in a wide range of international technology management positions. He is recognized as an international authority, pioneer and author regarding information systems and technology risk management. He's a self-styled risk ecologist interested in the intersections of technology, society and politics. He's president of the ITABHI Corporation and a managing director of the Decision Power of an Institute. He's a long time member of both ACM and IEEE. He's a lean system society and cutter of consortium fellow as well as being a member of the IEEE's Computer Society's Golden Core. He's an award-winning contributing editor to IEEE Spectrum Magazine and writes his popular risk factor blog. To his left is Dr. Jonathan Rothwell who's a senior research associate and associate fellow at the Metropolitan Policy Program at Brookings. His research covers a variety of topics on the sources of regional and national economic growth and prosperity with a focus on human capital and innovation. He has written Brookings reports on the labor market for education and skills, the economic consequences of patents and science and technical knowledge, the clean tech industry and how land regulations create income segregation and income segregation and unequal access to high-performing public schools. So with that, let's jump right into it and the floor is yours, Michael. Well, thank you very much. It's a real pleasure to be here and I want to thank the Economic Policy Institute and the Discussants for organizing and joining in this effort and Daniel Costa in particular, thank you. I was thinking about your brief description of the Jordan Commission and actually, I think all of our recommendations were unanimous with the exception of two that were unanimous, less one. So it was a nine-member commission and it was nine, zero, or almost everything except for a couple that were eight to one and that was a shock to me, I must tell you because the issues of immigration are not, as you can probably tell from the current debate, they're not very, there's not a lot of common ground on some of these issues. Okay, so this is not about immigration, although immigration is a part of this discussion and the question that I tried to address in the book is whether the US is falling behind in science and engineering and you're gonna hear some terms that are confusing like science and engineering, natural science and engineering, STEM, et cetera. So I'm gonna try to clarify, but they get confused and they're confused in the literature and they're confused in the politics and the advocacy on these subjects. So I hope if I confuse them at some point, somebody will say, oh, you just confused those things. Is the US falling behind? Well, some very eminent and influential people think so. The top quote is from the president and the other two quotes are from two reports, very influential reports, one from 15 business associations led by the business round table and the other from a distinguished special committee, an unusual committee appointed by the National Research Council at the request of four prominent members of Congress. I won't read them to you, but I think you can see that that is a pretty strong set of statements about the US falling behind. So my goal was to try to assess those kinds of concerns. Sorry? Oh, it should be quiet. Yes, I can't type very well. Yes, not a quite crisis, a quiet crisis. Thank you. These concerns are not new. It's, there's a long history and you'll find a chapter in the book on the history of such concerns. They have a cycle to them, 10 to 15 years, typically. First, the alarms about shortages of scientists and engineers are sounded by somebody and it's been different groups that have sounded these alarms in the past. The government responds sometimes quickly, sometimes very slowly, but always by boosting supply of scientists and engineers. And then after a lag and it varies, the delay varies, there has almost always been a bust that followed the boom. There's one exception in the five cycles which I'll highlight and you'll see why there wasn't a bust. Round one, after World War II, the power of science, basic science, physics, chemistry and electronic technologies was evident to everyone in the course of that awful war. And the Department of Defense and the Atomic Energy Commission in particular were concerned there was a shortage of physicists. And they supported very heavily physics research and physics graduate education. And if you look at the curve of PhD physicists which is actually in the book, you'll see it looks like, I don't know, like Montblanc or something. It goes up very sharply and it peaks and then it goes down a little bit and then later on it goes plummeting. There wasn't a boom, I'm sorry, there wasn't a bust in 57. It looks in those data as if there was about to be a bust around 56, 57. And indeed the Department of Defense and the Atomic Energy Commission which were a few years earlier supporting something like 95% of physics research in US universities were pulling back in their funding. They decided they'd reconsidered. But what happened was round two. Round two was the Sputnik induced round 1957. Sputnik one was launched. And then a month or two later Sputnik two was launched. And there was a panic, a political panic which led to the foundation very quickly of NASA, of the passage of the National Defense Education Act. And a few years later the challenge from President Kennedy that became the Apollo Moon Launch Program. The Apollo program was represented a huge allocation of money detailed in the book and a huge demand for science and engineering talent. Round three, well I should have said there was a bust at the problem with that round was that the Apollo Moon mission succeeded spectacularly in 1969 and enthusiasm for this massive funding of that kind of R&D waned pretty rapidly and there was a bust in the 70s. Round three, starting in 81 and going through the early 90s was the Reagan administration defense buildup. There was a report in 83 from the Department of Education that highlighted what it called a nation at risk. Some of you may have read that report. It's worth reading, it's got great rhetoric in it. A rising tide of mediocrity was one of the most famous phrases. The war on cancer which had started earlier was starting to be funded generously and the National Science Foundation and Office in the National Science Foundation produced projections of a looming shortfall of 675,000 scientists and engineers, PhD scientists and engineers by 2006. That waned in the early 90s. The Soviet Union imploded, the superconducting supercollider project was canceled and all of a sudden there was a bust. Round four, after the end of the Cold War, was different. It was not generated by existential national security or any kinds of concerns or competition with the Soviet Union, but rather by booms in important industries and I list them here. And it bust, as many of you in the room know, pretty uniformly all of those industries went into a sharp decline around 2001. And round five was overlapping a bit, but different. It was government initiated and it was driven by the decision, a remarkable decision to double the large budget of the National Institutes of Health to double it within five years. A decision that annually was never clear that it was gonna happen but actually did happen and the budget of NIH was doubled between 1998 and 2003 and then flat or declining depending on how you want to read the numbers. You'll see some numbers later, which has led to what most people in biomedical research would call a funding crisis for biomedical research even though the budget of NIH is twice as large as it was in 1998. So the question for discussion is whether we're now in round six. Now there are many ways to fall behind. You can fall behind in education and K to 12 is a good target for concern or maybe you're not graduating enough scientists and engineers. You can fall behind in research, basic research, applied research development or you can fall behind in workforce shortages which would impede innovation that would otherwise be made. Let me deal with each of these quickly. On education I think we all know that the US has real problems with its K to 12 science and math quality. It suffers from this shortage as a result of workforce shortage of STEM workers or scientists and engineers who are at the higher level. We've got a mediocre K to 12 system and we have declining student interest in science and math fields. And the common solutions for this are to fix the K to 12 science to math and I think all governors are trying to do that as is the federal government and a lot of company leaders as well. We want to encourage more majors in science and engineering and we will need temporarily if you will while we're catching up with this problem we'll need to import temporarily more such workers from abroad. And there's a little debate about this. I mean I think it is the conventional view. The business round table has said this very clearly and most business associations, the council on competitiveness has said this, the rising above the gathering storm report from the national academies has said this. It's a bipartisan view, unlike many things in Washington these days, this is bipartisan. Interest groups that otherwise hate each other and fight each other agree about this and it's echoed in the mainstream media. So here's the cover of the 2005 report organized by the business round table tapping America's potential. I put it here, is there a cursor on here? A pointer? Okay, well I can reach it. They put it on their cover that what their goal was. Double the number of science, technology, engineering and mathematics graduates by 2015 that was published in 2005, it's 2014 now. But double in 10 years, that's a major challenge and it hasn't been achieved yet. This is the rising above the gathering storm cover. Probably the most, in my view, one of the most influential, politically influential reports produced by the national academy complex in many, many years. So what's the evidence on K-12 education science math? The US performance in these subjects as measured by various comparative data collection efforts is average, it's medium. Some would say it's mediocre, the average scores. But one thing we know about the US education system K-12 is that it has unusually high levels of inequality in its performance. We've got large numbers of high performing graduates of the K-12 system in science and math. And indeed most of the major people who major in science and engineering in college come from these high performing tiers. We've also got very large numbers who are low performing. And if you average this distribution, you get a medium or mediocre or middling kind of average. And I put at the bottom here just a warning about the rankings. I don't tend to like rankings, including of colleges and universities because I think they miss a lot of important information. This one, you should just be aware that the rankings in the PISA study, the OECD study, half of the top 10 ranked countries are either very small like 37,000 people in Luton's line, they're one of the top 10. Estonia is in there at 1.3 million. Or there's city states like Singapore where I was a few months ago and Singapore is a city, it's not a, well it's a country too, but it is a city. That's 100% urban and it's 5.4 million people. It's very unusual and interesting place but it's hard to compare it to Great Britain or Germany. And also they rank cities in this ranking thing. So you've got Shanghai in there. I think it's the top ranked in the most recent comparison rankings. Macau is there and Hong Kong is there, but not China, China is not there. But you have three Chinese cities and they're all in the top 10. But having said that, it's also true there are some larger countries in the top 10. So it's not as easy as to say, oh no, those are just all irrelevant because you do have countries like Japan and South Korea and Poland and Canada that are in the top 10 and the US is not in the top 10. So I don't wanna exaggerate the thing, I just wanna make sure you understand that the top 10 include some very small places. Here's a ranking graphic. I don't know if you can see it but at the top is Shanghai, Dash, China it says. Then Singapore, then Hong Kong, China and then Taiwan, Korea, then Macau. So that just gives you a sense of you should be careful about these kinds of rankings. Well, there is a lot of strong opinion on this subject and here I put up probably too long a quote from the New York Times editorial board not so long ago, five months ago. And I put in red here, you can read it yourself but I put in red points that I'm gonna come to in the next slide. They say nearly 90% of high school graduates say they're not interested in a career or a college major involving STEM. According to a survey of more than one million students who take the ACT test. And then they say the number who want to pursue engineering or computer science jobs is actually falling precipitously. So let's go to the videotape or let's go to the numbers. This is from the ACT's study which they seem to be citing and they say interest in STEM is high and they give some numbers of 48.3%. I can't resolve that disparity. I don't know how that happened. It could be that the New York Times editorial writers were looking at data from eight years ago when there was a decline because of the collapse of those high tech industries but it's not consistent with the ACT data that is on their website, you can go and all look at it. So am I saying that K to 12 science and math are okay in the US, I'm certainly not saying that and it's got lots of problems. I already said that science and engineering majors in college come from these higher tiers and they are large and they're strong. And yet a large percent of the rest outside of these high tier, high performing students are not developing basic competence in science and math and I think the evidence is that every kid is gonna need that kind of competency to be successful in almost any career going forward. To be an informed citizen, you have to understand what people claim with data and scientific or non-scientific claims that are made about important issues. In my view, competency in science and math is equivalent to basic literacy in the 19th century in which the US actually led. It was one of the first countries to have compulsory primary free education partly because it wanted to have an educated electorate and partly because it wanted to have a literate workforce and I think these skills are essential in the 21st century and we're not delivering them to a substantial number of kids in this country. So I think they're major problems. These are just quickly some data on first degrees in natural science and engineering. It leaves out social sciences, internationally comparative. I assume you can see the graph and you can see that it's all pretty flat for the countries there with the exception of one, which is China and it's a very rapid rate of increase. One thing you should know about China which I didn't know is that when we put together natural science and engineering or science and engineering, if we combine them, we miss the fact that almost all of these very large numbers of Chinese graduates in these fields are in engineering. Something like 33% of all graduates in China are in engineering. So the number in science is not that much bigger than most other countries. So it's an engineering phenomenon and not a science phenomenon. And this is for doctorates. I think you'll see here that the US numbers, which are the dark line there, are growing. US numbers are confused because substantial and growing fractions are international students who are getting PhDs, so that may be exaggerated a bit. China is rapidly growing in the PhD domain and other countries are pretty flat. So the shortage claims, I would say the facts are a lot more complex than are being argued. The claims of general shortages are misleading in my view. What we have is an odd mixture of underproduction in some science and engineering fields at some degree levels. And here I would suggest associate's degrees in technological subjects which are in STEM but not in science and engineering, just to keep those terms distinct. High quality master's degrees in basic sciences are pretty weak in the US, although they're getting better in some fields. And I think employers are often claiming shortages about these categories rather than, they're not saying their shortages of PhDs typically, but it all gets confused in the public discussion. And then we have overproduction in some other fields in some degree levels and the poster child for that, unfortunately, is the biomedical sciences where I don't think anybody would dispute that the number who are graduating with PhDs exceeds the demand in the workforce for PhDs in these fields. There are some counter-arguments. I've heard many of them and I want to list them here so you know that they're legitimate honorable people who don't agree that some people say there can't be overproduction of highly educated scientists and engineers even if they're in excess supply relative to jobs and demand in the workforce. That these people are smart, well-educated, quantitative, problem-solving people and they will find their way in the workforce and they will do well even if they don't do science and engineering and that they will strengthen non-science and engineering occupations by doing so. So that's a counter-argument that you should consider. Well, why is there over and under production? Well, it's a systemic phenomenon I believe. Higher education as it's evolved has rather limited feedback from the labor market. It doesn't have good indicators coming in and adjusting its behavior. Governments do heavily subsidize higher education in a whole range of ways. I list some of them here. But governments have little influence in this country on the allocation of majors or the choice of majors by students. That's atypical in this country and many countries the government really finances the university system directly and it pretty well, it has great influence, shall we say, on how many slots there are by different fields. China's a good example of that. One reason they have such a high percentage is very heavy support from the levels of government in engineering programs. On the faculty side, many science faculty focus on their research and therefore on their PhD students. And research funding from the federal government which is very large and generous despite what people are worried about insufficient funding. I don't think anyone can say that it's small. It supports PhDs almost entirely and postdocs while master's degrees in science are self-financed by the students and associate degrees in science and technology or finance, by state and local governments which have been waning and by student fees. So there's a disconnect there where the bulk of the money from the federal government is going to PhD level and postdoc level. Okay, on the research side, the US is still dominant. It's not, there's no evidence that has fallen behind international competitors in science and engineering. In fact, it's in basic research, it's led the world since World War II. And this is a tribute old, I think, in substantial part to some remarkably wise decisions that the federal government made in the 1940s and 1950s after the war as recommended by Vannevar Bush who was President Roosevelt's science advisor. And if you haven't read his report, Science the Endless Frontier, I recommend it to you. It is a classic example of articulate and careful recommendations coming to the president from his science advisor. The main recommendation that was implemented from that report was that the federal government should begin to support basic research. It hadn't really been doing that. And that its funding should not go into government labs as is done in many other countries but should go to universities through peer reviewed competition among proposals submitted for consideration. And you'll recognize that is the dominant system that evolved and it's a very impressively performing system. It's still globally predominant and it's strengthening. And yet other countries are rising faster. And so they're catching up and they're narrowing the gap and this is primarily European countries and basic research less true in Asia where more applied research is typical and of course development by companies. What about the R&D tax credit? Most corporate leaders have called for doubling the R&D tax credit as necessary if the government wants R&D to not migrate internationally because there are lots of incentives from other countries and lower taxes. It's understandable to me why they would argue that and I have some sympathy for the argument but I think it's fair to say that shifts out of R&D to other countries have many other reasons than the R&D tax credit. We can talk about that later if you want to. The main problem with basic research is that it's structured for instability. You've got positive feedback in the system. Those of you who are engineers and scientists know that positive feedback usually doesn't work well and it usually collapses in some way or goes crazy. It's a bit like a simple model but positive feedback system would be your home thermostat that as the temperature rises it turns the furnace on to make it rise faster, the higher the temperature, the more it calls for more heat. That doesn't work out well in most systems but we have positive feedback like that in the system that's evolved. The more research funding that is allocated or appropriated by the Congress, the more grant claimants are trained and emerge to seek funding. The main reason for this is the way that we've evolved the system for support of PhD students and postdocs. They used to be heavily supported by fellowships and training grants. Overwhelmingly now they're financed by research grants. And I put some numbers up there which surprised me that they were so high. 86% of NSF-financed graduate students are financed under research grants, not their wonderful fellowship program which is very high quality but very small. At NIH it's 78,000, a 78% of graduate students in postdocs. Universities that have federal research funding have not been very good at providing information to prospective graduate students about what the careers look like in these fields. They have the money, they're recruiting them to be graduate students and it's difficult for the students, the prospective students to find out how the recent graduates have done. And they can, under current law, that it was not planned this way, it just evolved this way, they can recruit and finance international students in postdocs if there aren't enough American or permanent legal resident applicants, they can recruit internationally and they can finance them using federal research grant funding. So one of the outcomes of this is a kind of system that is structured for growth in budgets. And this analysis I cite here by David Korn and company estimated that NIH and its biomedical research ecology, if you will, the biomedical research system that it supports so heavily, in order for the system to be stable, it's evolved in such a way that it needs a minimum of a 6% annual budget increase. If it goes below 6%, the system is unstable. And of course it hasn't been anywhere near 6%. Since 2003, but before that it was 14% for five years. So you can see you get all kinds of countervailing forces under that system. But it's a recipe for this kind of oscillation we're seeing for instability and for booms and busts. Here's the NIH budget. The dark line is current dollars in thousands. So you see it went from in 1960, it was, well I can't even read it, it was so low. And it went up above 30 billion after the doubling. Here's the, it went almost vertical. And then it started to taper off and it's come down some in later years. The two lower curves are adjusted for inflation using different indicators of inflation. And they don't look very robust in growth compared to the current dollar graph. But they are very substantial. You can see that quadrupling or quintupling over that period in real dollars is what you see in those graphs. If I hadn't put the nominal dollar graph on there, they would look much larger. But here's what happened in NIH. The green line there is the number of grants submitted, grant proposals submitted that were successful and got funded. And you can see, here's 1998, here's 2003. You can see that during the doubling period, the number, the success rate, if you will, did rise. It had been falling, that was one of the reasons for the decision to double the budget. So it did rise and then it's fallen since. And it's now, the success rate is now below where it was before the doubling started. That's not a good outcome. And it tells you that the system is not stable because it has generated more demand as a result of providing supply. That's very ample doubling period. Here's the NSF funding history in constant dollars. And I think you can see here that it's been a history of sharp rises followed by long plateaus, followed by long rises and then declines and then peaks and then falls. Not a good way to finance long-term basic scientific research. So the system is very productive but it does cause harm and the harms are challenging. For research and graduate education, they're very long-term phenomena and federal appropriations are very short-term phenomena, maybe one year. That's not a good fit. And for students, many of them appear to have been encouraged to go, encouraged by prospects, if you will, to go into career paths that have waned before they were able to graduate because it takes you four or five years to get a bachelor's degree and then maybe five to 10 years additional to get a PhD and or postdoc, depending on the field you're in. By the time you finish, things don't look so good if you have 10 to 15 year alarm boom bus cycles. For research faculty, the consequences have been great when the budgets have been rising rapidly and terrible when they've been flat or declining. You've seen disrupted research projects of high value. You've seen lots of very high quality proposals that can't be funded and you see careers disrupted in mid-career, typically. For universities, the system has provided what I would consider to be perverse incentives for universities, research universities we're talking about now, to leverage up. We saw what happened with leveraging up in the housing market over the last 15 years or so. Universities have incentives to leverage up, not quite the same way, but I think they're equally worrying. First of all, they've been incentivized to maximize soft grant funding to pay their faculty and research staff salaries. So it may surprise some of you to know that in many medical schools, a tenured faculty member is tenured for life but is not guaranteed a salary. The salary has to come from research grant funding and the deans are urging faculty to get it up to 80% of their salaries from research grant funding. Less true in arts and science faculties but I think it's appearing there. They have incentives to increase the number of PhD students and postdocs to be the lab workforce in externally funded research grants and they can, as I said earlier, recruit internationally and finance it with federal research dollars and they've been incentivized oddly enough to borrow funds to build or remodel their laboratories. I don't go into OMB circular A21 but I recommend it to you as bedside reading because it will definitely put you to sleep but it has had major incentive effects, it appears, on some universities. And the result is that if the anticipated or hoped for or planned for budget increases lag, then there's a financial crisis because all the assumptions have not been fulfilled. This is more a problem with NIH and biomedical funding than it is with NSF and other kinds of funding. This is just the cover page of an article I wrote about this in 2008. I re-read it a week ago and I'm sorry to say that was worried about might happen, actually happened. I'm not happy about it but they did happen. This is a biomedical workforce task force that was set up by the director of NIH. I recommend their report to you. It was a very high quality group of people and they came to some of the same conclusions. This was in 2012 and this was just published in the Proceedings of the National Academy of Sciences by four rather stellar biomedical research statesmen and stateswomen if you can see their names or I'll read them if you can in which they're saying we need to rescue the US biomedical research system from its systemic flaws. So I don't think outside of what I don't, I think outside of the political debate about this, I think most leading biomedical researchers are aware of these systemic problems and very concerned about them. Now what about the shortages of science and engineering workforce? Well, first of all, STEM, that magical term that used to be SMET, it was invented by the NSF and they decided to change the order of the letters because SMET didn't sound so great. It was before STEM cells were a controversial issue. So there's some people who say, oh, I'm against STEM because I'm against STEM. Well, they're just, they're not related to each other. It's just an acronym. Anyway, it has no agreed definition. It can, some people say it's 5% of the US workforce, 155 million, by the way. Some say it's 20% of the workforce. The 5% is the estimate of the National Science Board. Their definition is that you have to have at least a bachelor's degree and you have to be occupied in a science or engineering occupation. If you look at people with degrees in science and engineering, again, bachelor's at a minimum, you get 11%. And if you go beyond the bachelor's degree and use other criteria as one of our discussants has done, you'll get to 20% of the workforce. Now, my view is that with such a large range, it's no wonder that confusion in public discussions prevails. If you're talking, there's a fourfold difference in the size of the workforce and people are using the terms interchangeably. The science and engineering part of the workforce, as Daniel Costa said, that's natural sciences, engineering, and social sciences, NSF, National Science Board definition. It's a critical part of the workforce and it's a small part of the workforce, about 5%. And I don't see any credible evidence I've looked. I would have loved to find credible evidence of general shortages in this workforce but I can't find it and nobody else really can find it. If the shortages that are claimed in this workforce were to exist, you would have to see relatively rising real wages for science and engineering occupations relative to other highly educated fields. You don't see that. You'd have to see faster than average employment growth. You see it in some areas and not in others but it's not a clear yes. And you'd have to see relatively low and declining unemployment rates and you don't see those. Again, it depends on the field. So there are no signs of broad, science and engineering shortages but there's clear evidence of large variations within the category, science and engineering. It varies by fields, it varies by time, it varies by geography and I'll give you some quick examples. You can have undersupply and oversupply coexisting across fields and I think we definitely have that. And you can have fields, individual fields that can change dramatically over a period of say 10 to 20 years. And here I think you could see that mechanical engineering used to be a high demand field until the auto industry sort of declined and then almost collapsed. Less so now, petroleum engineering to the contrary was in the doldrums in the 80s and 90s and now is perhaps the most highly paid part of engineering and things have changed. It's a matter of oil prices, a matter of how the industry is doing and the auto industry and so on. So from time, over time you can see within fields even dramatic changes. By geography, and this is important in the political debate, there are these local hot houses. Silicon Valley is the archetype. We have an apartment in San Francisco, I have lots of friends who work in Silicon Valley. It's a hot house right now. It's been a cold house in the past, but it's a hot house right now. And they are just atypical. So they are prone. Anybody in Silicon Valley will tell you they are prone to booms and busts, high frequency booms and busts, high amplitude booms and busts and they have extraordinarily right now, extraordinarily high housing costs which makes internal recruitment from other parts of the country more problematic. Any generalizations from Silicon Valley or similar hot houses are perilous to the rest of the country because they are simply atypical. Some of their concerns may be real, but they can't generalize accurately to the rest of the country. I give some examples here which I'll skip over in the next time. But it's possible then that these claims, the shortages of scientists and engineers that are conventional are simply overgeneralizations by honorable people who are seeing what they see, where they live, when they're living and are saying there must be a national serious shortage because I have that experience. For some though, the shortages they see are actually conceptually in demand rather than in supply. And I found this interesting quote from Kenneth Arrow and a co-author William Capron from 1959. Can you read it? Okay, I'm gonna let you read it. I mean, basically what they're saying is that some people say there's a shortage if there isn't enough activity underway in science and engineering they think there should be more and therefore there's a shortage in demand. And yet these shortage claims prevail and I wonder why so and I think we can give some simple answers to that. First of all, there has been and continues to be a very expensive and highly effective lobbying campaign led by information technology employers. Their emphasis has been on expanding or uncapping entirely the number of temporary visas. The H1B visa, the L1 and so on. I put up here Microsoft and Jack Abramoff. I did not know until I was digging into this book that Microsoft retained Jack Abramoff for quite a few years as one of their active lobbyists to expand the H1B program. It's in his book. I read his book. I think you might wanna read his book. It's very interesting. And more recently we've seen a new organization called fwd.com organized in Silicon Valley with a $50 million lobbying budget. It's Facebook founder who has organized it. Higher education, some support the shortage claims. Their goals are to increase the funding of research, to ensure low cost lab workforce and to sustain their PhD programs. Immigration lawyers mostly support the shortage claims. What are they looking for? I know quite a few immigration lawyers, as you might expect. They're looking for high volume temporary visas. Temporary because that means they have to be, they're not permanent, they have to be renewed. And preferably with the fees paid by employers rather than the migrants. Because you can charge higher fees to employers than to migrants. And some federal agencies mostly pass, I don't think you'll find too many now, have used these claims in seeking larger appropriations or in the case of security related agencies, hires that can be cleared for national security purposes. Where's the opposition to these claims come from? There's some people in the audience I recognize who may not be happy if I say this, but I think it's been poorly organized, poorly funded. You'd think it would come from scientists and engineers. But the associations of scientists and engineers tend to be much more technical organizations. They publish journals, they have conferences, workshops. They do standards. And they're often international organizations, not national organizations. So the, perhaps the leading largest engineering society, IEEE is a global organization, headquartered in New Jersey, but a global organization. And you can imagine they're ambivalent about getting engaged in a debate about US visa policy. They're also balkanized these organizations. There are many, many, many science and engineering organizations. And there are no, essentially no umbrella organizations comparable to the American Medical Association or the American Bar Association. So it's not even a contest. You've got very heavily funded, well organized, professionally staffed lobbying organizations on one side and a kind of balkanized and poorly funded and ambivalent set of organizations on the other. And the top lobbying goal of the lobbying activities I described seems to be to expand the H1B visa program. I assume many of you know what this is, but I'll quickly summarize its characteristics, non-permanent visas, but long term, seven plus year visas, typically. The numbers are capped, but the numbers are large. And the stock, we don't know how many have accumulated from the annual inflow and departure numbers, the net flows, we don't know, but it looks like it's in excess of 500,000. Universities have unlimited access to H1B visas, no cap, and have used them energetically to recruit postdocs from abroad. And here's a puzzle. Most employers are not required to even try to hire domestically before they obtain an H1B visa. It's reported the opposite, typically, in the conventional media. I don't know why that happens, but the press will say, of course the employer is required to show that he or she could not identify a qualified worker for this physician before hiring an H1B. That's wrong. It's true of only a small number of employers, but it is routinely reported that way. These visas have low education minima, bachelor's degree or covalent experience. They have weak wage standards, weak enforcement. There's this indenture argument, which I don't have time to go into, but employers can contract out their H1B workers to other employers. And the largest users are not actually the firm, the U.S. firms that lobby for expansion. They are mostly Indian offshore outsourcing firms. And here's a list from computer world of the biggest new visa approval recipients for 2012 and 2011. So let me close by saying, well, what goals might be aspired to in rectifying these kinds of complicated, they're not simple situations. They're important situations and they need to be addressed, but what might we aspire to? Well, here mine and you'll have others, I'm sure. K to 12, I would say very strong need to improve science and math education for all students, not just focusing on the important, high tier, high performing students. For higher education, it would be nice to de-incentivize some of the incentives for leveraging up. These are incentives that institutions find it difficult to resist, but they're not good for them in the long run and they're suffering now, serious problems. The idea that you wanna expand your core faculty on the basis of soft money research grant funding, not healthy. And I think the universities could also be encouraged to configure their degree composition better for the actual demand in the workforce. Right now it's not very well configured. I think it would be good to clarify the goals of these visa programs. It would be nice if we could restrain the alarms about shortages and thereby modulate the damaging booms and busts and those cycles that I described we've gone through. And to do that it would be necessary to find a way to objectively assess shortage claims. Right now we don't have a way to do that. Short digits are claimed to members of Congress and they have to decide if they believe it or not. We need to have a common definition of STEM occupations. I think one of the discussants will talk about that. And I would say in closing it would be highly desirable to have robust growth in federal science and engineering research funding, but steady growth, not rapid acceleration to 14% a year for five years and then flat or declining, which is what has happened in the biomedical system and is causing all sorts of problems for biomedical researchers, for their institutions and for the overall research enterprise. There's a lot more detail in this book if you're interested and I'd be happy to hear from you. When inevitably you will find something I got wrong, I would like to know about it. And thank you very much. All right, next is Bob Charette and Mike Zaronzo. Thank you, Dan and Michael and Jonathan and you all. Some of you have been at other presentations I've given on my paper that I wrote last year for IEEE Spectrum called the STEM Crisis is a myth, which caused a little bit of controversy to say the least. I'm not here to talk about that paper as much as I am to really talk about Michael's book. I'm biased by this book. All the research I've done, I spent several years independently from Michael looking back over the years at the STEM crisis, actually going all the way back to the early days of electrical engineering back into the 1880s and 1890s and you can find almost verbatim comments from those days about how terrible our scientists and engineering education is. You can find the same issue in terms of the STEM pipeline, although we didn't call it STEM pipelines. The leaky pipeline that we have today actually looks like it's extremely well-patched compared to the 1940s, 1950s and early 1960s were dropout rates approached anywhere from 50 to 60% in engineering schools. When I was a freshman at the University of Massachusetts in electrical and computer engineering after a stint in the military, professors had absolutely no problem flunking 60% of the incoming electrical engineers because you wanted high quality engineers and it was a survival school and that's just how it was. Now, you can increase the education of undergraduates or high school students to try to reduce that and I'm all for that. However, I think as Michael says and I said this in my paper and I say it to my two teenage daughters that you need to spend a lot of time learning science and mathematics just from the way it helps you think and I tell my girls, I have one who's 14 and one who's just about to turn 13. I call her a teenager because if I don't, she gets mad at me, is that they are going to have four to six as a minimum, careers, different careers and they have to learn how to learn and I think this is really an important submessage of Michael's book because what we need to do is focus on getting better science and better mathematics as well as very good English and history and humanities. One of the things if you again go back to the literature the IEEE, before it became the IEEE was an I-R-E and the Institute for Radio Engineers and if you go back to the, there was a number of massive debates that were held by the leading lights of the day about the education of engineers and one of the things that they used to complain about was the teaching of mathematics and the death march, how much high school students hated mathematics and hated how it was taught, as well as the absolute need for a very well-rounded humanities strong engineer because if you didn't have a humanities strong engineer, you had an incomplete engineer, one without common sense. So from my perspective, and I think it echoes what Michael has in his book, is that we need to really look at how do we increase the educational capability of our students and a lot of that does go into improving science and engineering. Now that said, I think one of the things I wish for when I saw it, when I read the book, is I wish there was actually a lot more detail. I know Michael, it's great. It's always great as a fellow author to tell another author that he should have done more work. They always appreciate that. But there is a lot of detail and in fact for those of you who subscribe, say to the New York Times, which has archives that go back to 1840 something, just go and Google or search for engineering education or science and engineering or the state of education and you will find exactly the same things that have been talked about over the last 50 years. My dad was an engineer in the 50s and 60s and 70s and one of the aspects again that I would have liked to have seen more and it sounded more like I'm criticizing Michael's book but it's actually, it's a great book, believe me. But I think that one of the things he talked about a little bit in terms of research dollars, but you can't forget the economics that also come into play. The booms and bust phases of information technology, I've gone through several of those as an electrical and computer engineer. My dad went through it as a physicist. You go through these periods and that's part of the engineering way of life. It's also one of the things I think that everyone who wants to go be an engineer ought to be an engineer but they also know about the issues of having a long-term career because of these booms and busts and also because of just the natural way of technology obsolescence. Because one of the things that again, Michael's book doesn't touch on as much as I would have liked to have seen it but I think it's also a big part of this story is how employers don't spend money on educating scientists and engineers. How the average age of working scientists and engineers has dropped. The difficulty of older engineers to find employment. Now some of this is natural, some of it's been around forever. But I also think that it's much different. My, the field, the career field today is much different than what I entered or when my father entered it in, I think he entered it in 1950 as a physicist. So all these issues come to play into why we think there's a STEM crisis and what are some of the roots of those STEM crises. I wanna just tell you one story which I think highlights kind of the issues that I see and I think that what Michael sees and I'll get off the stage because I'm not supposed to talk for more than 10 minutes and I'm getting close to that time. A few years ago, the current mayor of London, Boris Johnson, who was then the conservative party spokesman on higher education was bemoaning in the Guardian newspaper that Britain was falling behind in educating sufficient numbers of engineers and scientists, especially nuclear physicists. Sir John Rose, who was at the time was the chief executive of Royal's Royce, wrote a letter to Johnson about his article saying, if you don't have a nuclear industry, then anyone who is smart enough to be a nuclear physicist is not going to choose that career. People won't become engineers just for the sake of it. If the industry wants to have a massive amount of engineers and scientists, let them put up the money for it. I'm very tired, to be honest, after 40 years of being an engineer, of hearing companies bemoan that they don't have enough engineers and scientists and are absolutely unwilling to spend the money, either internally in education or at the schools or anywhere else to get them. Let them also stockpile engineers and scientists like they did in the 50s and 60s. If they really are worried about not having them in the future, okay, put 10 or 15% of your scientists and engineers, keep them on the payroll for another 10 years until things turn around. Think about that for a bit. You won't find one company who's willing to do that because the bottom line drives it. So, if that's how you want the field to be, then fine. No more myths, just everybody is like a sports star with a short-term career. But if that's what we want for our professional scientists and engineers, then we ought to talk about it that way. With that, great book, Michael. Now, Jonathan Rothwell from Brookings. Okay, thank you. It's a pleasure to be here today and I want to thank Daniel for reaching out to me and I think I do bring a slightly different perspective here, so hopefully we can get a good discussion here. But let me start. Okay, great, thanks. Let me start by going over the areas of agreement and sort of highlighting what I think are some real strengths in Michael's book and some things that I think add a lot of value to the public conversation about science and workforce policy. So, first of all, I think he's totally right to point out the difficulties of the academic labor market for PhDs. And this is obviously by no means unique to STEM. Last I heard English PhDs were having a hard time finding work as well. I think partly it's a demographic issue with the baby boom generation. I think partly it's the fact that universities can bring in academic students from all over the world. But it does lead to many unattractive features of pursuing a PhD and embarking on a career there. One, if you are married and my wife and I just recently went through this process. I finished my PhD five years ago and every PhD has to think about this. Are you gonna move to Wyoming or North Dakota and you can convince your spouse to go there if that's the only job that happens to be open in your field. So, I think he's right to emphasize problems there. I write to call for more predictable and that's volatile R&D spending. And smarter incentives for the efficient allocation of R&D money. I think that's a great insight. Right to place more responsibility on universities for aligning graduate missions and PhD programs with market realities. And basically correct that the US is still dominant in an aggregate sense on R&D and innovation and these sorts of things. And I do think some of the concerns about the rise of China are overblown. On one point, yeah, I think he's correct but downplays some of the weak per capita educational attainment performance in STEM. He doesn't mention some of that to his credit but I think it's somewhat downplayed. And so to get to some more of my critical remarks, I think the biggest issue I had is an overemphasis on the PhD labor market. And as he intimated in his slideshow, there is a lot more to the STEM shortage debate and the STEM debate overall than the PhD labor market which is a small share of it. And yet most of the book seems focused on that. The other issue is, you know, no surprise I've debated how Sal's been publicly a couple of times now about the shortage issues. I'm on the side that thinks that we could benefit by having more people educated in STEM fields and that they're likely to find lucrative work. So I think the historic discussions about shortage implies that there's no long-term problem. I do think there's a long-term problem. I think the long-term problem is often obscured by the booms and busts in the macro economy. So, you know, during the Great Recession there was not a shortage of engineers in any immediate sense or computer scientists or any other occupation because unemployment rates went up a few percentage points and companies had a much easier time filling open vacancies. But that's a temporary issue and as the economy has started to recover the difficulty of filling vacancies has increased again and data that I've analyzed and will be releasing in a report suggests that it's now as difficult to fill STEM vacancies, especially in the core STEM professions today than it was at the start of the recession. His literature review to introduce another point I think ignores some of the most important scholarship from labor economists. So that would be scholars like Claudia Golden, Lawrence Katz, a few others that would be happy to send references to. The basic argument that comes from this literature and seems to be sort of conventional wisdom among labor economists is that there's been a rising skill premium over the last few decades and this reflects stronger growth and demand than supply. So I would have liked to seen some engagement of that literature. Also, he talks a little bit about the choice of major literature and the STEM dropout rate once you enter college. There's some really innovative work that looks in depth at specific universities and university systems by the researchers I mentioned here. They find much more alarming results than the work cited in Michael's book. Fourthly, I think he's wrong insofar as he believes this that the US education outcomes imply that we have enough elite students to realize the full potential for innovation and entrepreneurship. I'm not sure to what extent Michael subscribed to that view, but he does cite some work that sort of makes that argument and I want to be clear that I think that argument's wrong for a number of important reasons. Even if say sort of the core jobs that Michael's concerned about that are doing research and development is only one or 2% of the US labor market and sort of we have, we can come up with 100,000, 500,000 students that score well in the ACT. The students that are getting abysmal educations that have no education opportunities could contribute just as much if not more to US innovation if given the opportunity and so I think it's really important not to suggest just because we have this inequality in our educational system that it doesn't necessarily affect innovation to me. Inequality is a huge drag on US innovation. And then so the last point, I do think in some ways he downplays some of the weak US per capita performance. Though as I said, US looks very good on an aggregate level. So I'm gonna get into a little bit of data. I'll try to go through that pretty quickly. If you want the slides, I'll be happy to email them to you but let me just say a couple things. I think the book's major claims would be more compelling if they were better grounded in data and quantitative analysis, especially on the existence of the booms and busts even in the PhD level. No surprise, I think the notion of STEM shortage as being untrue in sort of merely a lobbying effort for more H-1Bs is not accurate. And I would have liked to see more guidance on things that I think are critical. I guess in his slides it characterizes the sort of conventional view that we need to improve K-12 STEM education. Subscribe to that view. Although I gotta say, your closing slides caught me off guard because I agreed with just about all of them. So maybe there's less disagreement than I thought from an initial read of the book. But let me just go through a few things real quickly just to kind of highlight some of these points in a little bit more detail. So according, I've got a new definition of STEM that Michael mentioned in his presentation which is more expansive. It's based on surveys of workers that ask them how much knowledge they need in core STEM fields to do their jobs. The basic implication is that blue collar workers and healthcare workers and even a few financial workers and so on actually need to know a lot about in some cases computers or engineering or biology or chemistry. And so these workers if you were to do the analysis rightly should be classified as STEM workers in my opinion. So if you do that, you get a higher share of workers that don't have a bachelor's degree. But in using my definition or using a more core NSF definition that focuses on engineers, computer workers and scientists either way the PhD share of the STEM labor force is like three to six percent. And so it's not a huge part of the STEM labor market. And in many ways they have the smallest wage premiums relative to these other occupations. So if you've got a high school diploma and you're working in a STEM occupation say as an HVAC technician you can expect earn 75% more than the average worker with a high school diploma. Using my STEM definition you can still earn a premium at the PhD level. But if you're working in only in the core STEM professions you have a PhD you're basically earning what other PhDs have. And then just one point, even though I'm conceding that the PhD labor market especially for the academic labor market is not that great, it's still the case that people with PhDs are doing fantastically well relative to the rest of the population. And they're also finding jobs. Not only are they making more money, they're finding jobs. So just to put this in perspective the 2012 unemployment rate for people with PhDs working in the science occupations is 2.5%. So that's probably fallen a few percentage points since 2012. Now let me get into a little bit are there are booms and busts in the labor market for skilled workers. And this is where I wanna talk a little bit about this long-term trend that I alluded to earlier that labor market economists have focused on. This is the return to college education. So back in the 1950s whether you had a graduate degree or a bachelor's degree you were making about 10% more than someone with a high school diploma. That increased pretty steadily. There was a fall from 1970-1980 when there was sort of an increase in supply during that period. But really since 1980 and through the most recent data what you've seen is increasingly large premium for those with a bachelor's degree or graduate education. So now it's above 80% for those with a master's or PhD and it's 60% for those with a bachelor's degree. And the most common way of interpreting this is that demand has outpaced supply which you could define as a shortage. Now among bachelor's degree holders and PhD holders we know that STEM, those who have majored in STEM are in the highest salaries. What I've tried to do is look back historically about how the returns to STEM knowledge, if you can isolate STEM knowledge as embodied in all these occupations and control for changes in education and experience, how has that changed over the last few decades? And my calculation suggests that it's also increased every year since 1980. Now I've also looked at some wage patterns for some PhD holders in some of these kind of core STEM occupations. For computer software developers, wages have real wages adjusted from inflation increased by 90% since 1980, which is obviously spectacular. It's been not as spectacular for some of the core science positions for physicists and astronomers. It was 20%. Even for biological scientists, it's increased, although they're still making their actual level of wages, I think is only something around 60, 70,000, which is quite low for PhD workers generally. So there may be something to Michael's point about biomedical sciences being oversupplied. And this next graph I think really gets it to some of the key issues about whether there's this kind of boom and bust cycle. So I wasn't entirely persuaded by that. This is looking at the millions of R&D dollars adjusted for inflation relative to the number of STEM PhDs graduating every year using NSF data. And what you see is a small dip, so it cost about one million on R&D dollars to get one STEM PhD in 1976, and then that dipped to about 750,000 in 1970. And it's gone up pretty much, I mean, it's gone up dramatically since then. There have been some slight dips here and there, and certainly a peak around the mid 2000s. But to me, if you're saying there's this boom and bust and STEM production and all you need to do to ramp up the supply of STEM PhDs is throw more R&D money at the system, but we've doubled the amount of R&D spending per STEM PhD. So it seems to me that we're not as good at graduating STEM PhDs for the same amount of R&D spending as we used to be, which could be a problem for the workforce. And then I just wanna quickly touch on, if I'm out of time, I can sort of quickly summarize this. Yeah, you're over. Okay, sorry. I'll just say on international exams, even elite US students score pretty low relative to other OECD countries. And countries like Finland and Sweden have come up with some important technology advances like Rovio, the biggest video game production company. They're Spotify, they're Skype. So there are innovations even in some of these small countries. Okay, thank you. So we are starting to run out of time. So we started a little bit late, so hopefully you can bear with us for a few minutes, but I will just quickly let Michael respond to some of the points made by Jonathan and then maybe we'll open it up to the audience. I'd rather hear from the audience. Or is there any key one or two points you might wanna at least respond to, maybe a top one? Well, the long-term, ignoring the long-term problem point. The future is pretty hard. We have some feedback, by the way. There's a positive feedback in the system. The long-term future is pretty hard to forecast. I think nobody's succeeded in doing it very well. I mentioned the embarrassing NSF forecasts that were done in the late 80s, which led to great embarrassment at the NSF and an congressional investigation of why, according to the chairs of the subcommittee oversight and the science committee, as I recall in the house, the question was, well, why were we misled by these forecasts of looming shortfalls? So I wouldn't like, I'm a demographer, demographers do projections, I can do projections. I never believe long-term projections as forecast because we have a lot of experience of failure and it's even harder to do than demographic long-term projections to do labor force, demand and supply, projections that interact with each other. The National Academy did a wonderful workshop report in 2000 in which they basically said all such efforts at the PhD level have failed. We don't have the ability to look in the long-term. So I think we should be cautious about making long-term forecasts. And I do agree that there have been increasing returns to college education, bachelor's level education. It's hard to dispute that. The question is whether that's demand rising more rapidly than supply or a kind of selective recruitment process in which employers have used the bachelor's degree as an indicator, a kind of floor from which they want to hire whether or not there is excess demand. It's, we've seen a big shift out in the, or big rising inequality in income in general in this country. And you can do it by education or you can do it by other variables and it's all the same. We have this big rising gap and a kind of hollowing out of the middle. A lot of it has to do with the decline of manufacturing which used to provide well-paying jobs and careers for people with sub baccalaureate level education and it's declined as we know rapidly as a percentage of the U.S. economy. But I'm gonna stop there because I wanna hear. So questions from the audience. We just ask that you please give us your name and affiliation and please keep the commentary short because we're short on time and try to get to the question. The definitions is- Before you, Daniel Keen, American University and the Urban Institute. And there was talk about bearing definitions of STEM but I was curious if you could give your definition of shortage because I was taken aback by the discussion of the consensus of labor economists because I don't think Goldman-Cats, Cardinal DiNardo, any of those that are looking at the rising skill premium have ever said that that's a shortage and demand moving faster than supplies and generally defined as a shortage. So if you could just talk about how you each define a shortage. Whoever wants to start. Okay, well, yeah, I mean to me I'm not really hung up on whether it's called a shortage as long as you'll agree with me that, and you probably don't, that we should increase the number of people educated in these fields. And I think Goldman-Cats and Oriopolis and Asimoglu, Autor would probably agree that the US and individuals would benefit if they got more education. Well, from my perspective, it's the graduation rates in core STEM areas and whether or not they meet the demand of industry. And if you have a significant number of graduates and the demand is not there, then you don't have a shortage. Let me do it as a negative. I think most neoclassical economists would say that labor markets adjust with a lag and if you see rising wages, this will over time evoke a rising supply of people aspiring to enter those occupations. If you have this kind of rapid fluctuation in demand and supply, particularly demand where things are booming for, well, let's see in the 90s, it was probably what from about 95 to 2001. I don't know how people in the audience would count the high-tech boom of the late 90s, but it was in the range of five to six years and then a bust, the problem of course was it did evoke an increase in the number of students going into majors in fields that were showing rising wages during the boom period, but they graduated in many of them after the system had bust and they emerged sort of bright-eyed, bushy-tailed into a very chilly labor market for the skills they had acquired in response to this demand. So these fluctuations with the lag built into higher education are quite damaging. But in the side, go ahead. James, you talked about how the federal government, how funding drives this, but there's also a part of rooms that quite often have real intrinsic causes, like I was a student in the 60s and of course the post-war boom, which everybody who moved on to the 60s, atomic also was the creation of solid-state physics as a discipline. And given that you point out, it's very hard to tell the future. Can one in fact really tell given what a new field like solid-state physics or molecular biology is growing, how much, what is the optimal choices for funding over a period of 10 or 20 years? Or does one just open the gates and see what happens? Is that, in fact, the best outcome? I think you can't tell what's going to happen. I don't think anybody in the 50s imagined the microprocessor. And I don't think, you were at IBM, you told me earlier, I don't think anybody at IBM imagined that the IBM PC was going to threaten the existence of IBM. They created the IBM PC not to destroy the company and yet it almost did, as I understand the history. So these things are really hard to anticipate, particularly in dynamic fields, such as those we're talking about, where there can be major, there can be nothing happening for 10 years or 20 years and then a sudden innovation or a sudden insight revolutionizes the field. And then if you deal with petroleum engineering, if you know or if I know what the oil price will be in 20 years and what the course of it will be, will it be a steady increase, steady decline or fluctuations up to some level? Then we could probably make a lot of money, speculating in the oil futures market, but it's very hard to know that there are too many variables going on, like political variables and other variables that nobody can anticipate. So I'm not very confident that we can look with foresight and say, well, we know that the leading edge fields in science and technology in 2030 are going to be X or Y. Ross? Ross Eisenbray with EPI. Everybody agrees that we should have better science education, better math education. Who could argue with that? What is in the effect of responding to a short-term need with increased immigration in particular, not even immigration, but temporary migrant workers as a response to industry's needs? Does that have an effect on the educational pipeline and on the interest of U.S. students? Well, that's a dangerous area to go into. I can tell you from past experience and talking to lots of other engineers that the H-1B is viewed by, and again, without native engineers, if I will, without that being pejorative, is as competition, as a way of lowering their wages, ways of asking themselves whether or not they should actually pursue an engineering or science career, especially in the information technology arena where H-1B visas is really taken off. So I think there is a level of discouragement of people saying that why should I pursue this field? Because even if I get my degree, how long is it gonna be before I will have my job outsourced or taken away? So I can tell you from personal experience, I can tell you from talking to people who are in my community that H-1Bs are not viewed with a lot of joy and benefit. Jonathan? Yeah, from what I don't remember exactly I'll tell them ahead. I don't focus on immigration issues, my research too much, but I think something like 5% of STEM workers are on the H-1B program. So it's hard for me to imagine that too many students say that getting a bachelor's degree in computer science are worried about temporary foreign workers taking their jobs. I think they're more focused on competing against other Americans and they may think more globally, like well I have to do well in school and study hard because I have to take into account that my company could relocate to Ireland or there's an increasing trend of US companies moving their R&D facilities to European countries or other countries. I do think they're concerned about competition. I don't necessarily think that they're worried that their classmate who maybe have been born in India or China or Sweden is any more of a threat than the person born in Oklahoma or Nebraska. I suggest you talk to people at the universities about that. Michael? Well there's a section in the book about the H-1B visa and I think it is large enough in the areas in which it's used to be a significant factor if you compare it to the entire STEM or Science and Engineering workforce which is smaller than the broad STEM workforce as you've argued. It may not be a very high percentage but it's heavily concentrated in a few industries and it is, well it's not in computer science. Computer science is not really the right thing to compare it against because computer science is- I use computer science as a very broad term. Okay, okay. Computing, I guess you might compute. Computer science- Both H-1B visas are for computer occupations. Computer occupations, right, but computer science is a relatively small discipline and a relatively small factor in the information technology. Well, so you can certainly major in other fields and get a computer. There are lots of problems with the H-1B visa and if you talk to the people who invented it in 1990 in the 1990 Immigration Act, it worked out in ways that were the opposite of what they had intended. The reason for the cap, which was initially at 30,000 or 35,000 and then was boosted to 65,000 initially in the Senate. The house bill was 30 or 35,000. The purpose of the cap was to signal to the employers that this was a temporary thing. They were going to be getting up above to the cap pretty quickly and they better start investing in education and if they really claimed to have a shortage, they should be investing in education and in service training, if you will, or continuing education among their staff over this temporary period when they could have access to up to 30,000 of these temporary visas. And I think we know what happened. It, they simply went and raised the caps to 65,000 to 115,000 for three years to 195,000. It's now 85,000 plus unlimited numbers for universities. They were added later with no cap. So it didn't work out the way that the sponsors of the legislation had in mind. It was supposed to be a temporary fix and like many temporary fixes in the United States, they become permanent structural elements of the economy. So I think there are some really serious issues with respect to the H-1B visa, but that's not the point of the book really. And the last thing I'd say is that the Jordan Commission, which you referred to earlier, came out unanimously in opposition to temporary worker programs. What the commission said was, if there really is a labor need, the visas should be permanent or they will undercut the working conditions and wages of domestic workers, including recent immigrants. That's not the way the politics have worked. I would just add that one of the reasons it's possible that employers like the H-1B is because of the way the wage rules are set up. And that's too much to get into, but there are also legislative proposals in the house and the Senate to greatly increase the numbers of H-1Bs. So because of when we started, I could take one last very, very quick one. And what is told to high school students is a statement from the Department of Commerce that in 2018, there will be a 17% increase stem-unfilled stem positions. And that's from the Department of Commerce, I don't know what you said that, and so on. That's kind of the broad level statement that is made to these students. I'd like to ask the panel, is that an appropriate statement to make for the future of some other statement you made? Can you do any answer briefly? I mean, I probably agree with Michael. I don't feel very comfortable with the forecasts of labor markets. What I'd rather show is recent his growth in jobs. I think like predicting the weather, yesterday's weather is the best predictor of today's weather. And if you were to look at that, you'd see computer occupations and healthcare occupations since the recession have been to the capacity growing fields. Engineering has been a bit further down the list. But if you look at job vacancies, I've looked at that data in some detail. There is a lot of evidence that engineering job vacancies are going unfilled at much higher rates than other positions. So to me, I would feel confident sharing that information with the caveats that this doesn't necessarily mean 10 years from now, it'll be the same. Anybody else briefly? Well, I would encourage students in your program to pursue science and engineering if they're interested in it. I wouldn't try to convince them by giving them numbers like that. However, will be is never appropriate as a part of a sentence with respect to 2018 or 2028, because we don't know what will be. These are probably based on the Bureau of Labor Statistics occupational projections, which are done every two years. They're 10 year projections. Sounds like that quote was coming from the 2008 projection because it's 10 years 2018. They revised them every two years. And of course, 2008 was they were using data from 2007, 2006, 2005 to make their projections before the financial meltdown and the great recession. So I would look for better data than that, certainly. And as far as job vacancies are concerned, I'm scratching my head about job vacancy data. There seems to have been a change in the way job vacancies are counted. It appears that employers are multiple posting job vacancies and you accumulate by crawling over the web and you get lots of job vacancies. To some degree, with respect to green card visas, there are job vacancies being posted for legal reasons. You have to post job vacancies if you want to petition for a permanent visa. And yet these are not jobs that are actually open. They are being posted for technical legal reasons. So I don't know how to interpret the job vacancy data, but they have been useful in the lobbying campaign. No question about it. They are used routinely to say that they're major shortages, but I think you have to scrutinize them pretty carefully. Is that, thank you everybody. And Christian, the books are $20, is that right? Yes, books are $20. They will be available at the table that you saw when you came in off the elevators. And we accept cash and all credit cards. And Michael will stick around to sign the books if you get them and I urge you to get them. Thank you so much. Please join me in thanking all the panelists. Thank you.