Committee Co-Chair Norman Augustine reviewed the workshop and offered his observations, which he stressed were not necessarily comprehensive but were reflective of what he had found interesting. He started with some broad issues that showed some conflicting underlying beliefs. Charles Vest, he noted, said that we are heading for a train wreck, but the panelists from DOD did not identify such a problem. Two other participants also presented data suggesting that there was not much of a problem. How you see this question of course depends on how you parse the data or on the remit of the person giving the data, but, overall, this presents a confusing picture.
Augustine then observed that the workforce issue has been solved in the industrial sector by moving overseas. Then he asked whether the STEM workforce issue is DOD’s problem or reflects a nationwide issue. Many participants observed that DOD is not going to have much effect on the overall, nationwide problem. Augustine noted, for example, that the United States has 14,000 independent school districts, which by its sheer number presents an enormous challenge to DOD. He suggested that DOD focus on the unique things that it might do—for example, providing summer employment in its laboratories to young people who might be enticed into returning later—a tool often used in industry, and which affords the laboratories an opportunity to identify high-quality candidates for future employment. There is also the DOD school system, he observed: this is a great prototyping tool that could apply some of the things that participants in the workshop have been discussing in an environment that DOD controls.
Augustine then turned to the vexing question of how to predict technical breakthroughs and national security events, the latter of which define DOD’s budgets. DOD, in his view, should come up with a strategy that maximizes its ability to deal with uncertainty. He noted that DOD does a great job of career management for uniformed military personnel but not for their civilian counterparts. He allowed that he could not recall ever having seen a succession plan while employed at DOD; no company would not have a succession plan. Augustine stressed the importance of lifelong learning and suggested that DOD civilians should get “re-capped” in new technologies so that it is not as big a jump to convert them to new areas.
Augustine’s next point concerned the supply of engineers: he noted that, on the one hand, whenever he meets with business leaders they say that they cannot find engineers to hire, but, on the other hand, when he says this during television interviews he gets deluged with e-mails from engineers who cannot find work. He observed, however, that some people let themselves obsolesce with respect to the technologies. Specifically on systems engineers, Augustine noted that they are trained as well as taught—both being essential. It may be necessary for DOD to allow its systems engineers to spend some time in industry as part of the training. Lastly, he noted that
when he graduated from school, DOD or NASA were the great places to work and were considered the leading edge. Today’s DOD might be characterized variously as an integrator or an overseer or a manager of R&D—such work is less exciting. There are nonetheless some opportunities to entice people to work at DOD in these areas, and these might include doing a better job of getting the word out on what it is that DOD does, reducing the time that it takes to hire someone, and reducing the time that it takes to obtain a clearance. The goal is to make DOD an employer of choice.
Augustine then touched on several more issues, starting with the challenge posed by export controls and the offshoring of capabilities, a model that has worked well in other industries but not in defense. He noted that participants had suggested that DOD hire non-U.S. citizens and wondered whether much of the classified work is not being classified at too high a level; he suggested that addressing the latter might allow more noncleared personnel to work on projects and would ease the supply crunch.
Augustine also mentioned that a very important approach for getting more scientists and engineers would be to tap into the enormous pool of women and minorities. He referred to data presented at the workshop showing that these groups are underrepresented in STEM fields, and he asked how we attract more women and minorities into such fields, suggesting that K-12 is the point at which to intervene. He stressed the need for teachers to have a background in the core subjects that they are teaching and suggested that DOD could offer summer jobs to teachers.
Next, Augustine discussed the leakage rate in the pipeline that produces engineers and scientists, owing to students not finding high school courses interesting. He referred to Carl Wieman’s talk, in which Wieman observed that money is being siphoned off from teaching to research, a problem that Augustine suggested could be alleviated by DOD and other funding agencies providing more support for research.
Referring to Panel 1, the session on emerging science and technologies, Augustine added a few future technologies that he considered were missing from the discussion. These technologies included robotics; telepresence; the non-intrusive identification of individuals and tracking of their locations continuously; likewise, the continuous tracking of the location of nuclear weapons; and communicating with computers on human terms—for example, asking the computer in plain English, “Where am I?” or ”Where was I last Thursday at noon?”
Augustine then discussed the problems identified by the panelists speaking on acquisition issues. He noted that there are two tracks: the fast track and the regular track. To interest engineers, one would have to concentrate on the fast track, since no one would want to work on a project in which it would take 20 years for a finished product. Further, over that 20-year time span, there is a likelihood of the project’s being canceled. Augustine suggested that prototyping would be one way for industry to preserve its most valuable skills, those of design and development. He further noted that according to a number of studies, there is an anticipated shortfall of from 1 million to 2.5 million workers by 2020, observing that one could multiply by the DOD fraction of the national workforce and get an estimate by field. He observed that although workshop participants had said that it would be difficult to find such estimates, this was a rough-and-ready way to do so.
Augustine noted that there is an issue regarding financial markets and how to get a long-term perspective. He observed that there are disciplines that are merging, and perhaps DOD should look at its own organization and ask whether it is not too stovepiped to go across disciplinary boundaries.
Then Augustine offered what he considers to be the bottom line: specifically, that we do not know how to predict the state of the world that will drive DOD’s demand for engineering and should instead count on a strategy that deals with uncertainty. This points to a need for a workforce composed of people who are current in their fields and maintain their expertise throughout their careers so that they can change course rapidly; it will not be possible to wait 10 or 20 years to produce a new graduate with a bachelor’s degree in engineering. Augustine further noted that we could enlarge the pool by including foreign applicants, and we could also benefit by reducing the number of items requiring clearances and by making the field of engineering more attractive.
Augustine then opened the discussion to the floor and asked participants for their observations.
• Sharon Levin, a committee member and session moderator for Panel 3, stressed the need to find ways to collaborate globally against the backdrop of new technology development and the spread of talent worldwide—not everyone is going to be in the United States. She agreed with David Chu about focusing on the demand side, given the limitations on funds, and on export controls and non-citizen workers.
• Anita Jones, a committee member and session moderator for Panel 2, suggested considering what was not heard: Panel 2 did not hear much on the shortfalls in the technical workforce at specific times. She further agreed with Norman Augustine that adaptation will be the key, suggesting that there are many ways for DOD to implement such a strategy. For example, if more engineers are needed in a certain field, one can simply raise the salary by $20,000 per year as in the case of petroleum engineers. Jones also observed that what did not show up in the data is the ability of engineers to move to a different field rapidly. As a corollary, one might conclude that moving within STEM is not so hard, and DOD can implement polices to enable this.
• Frances Ligler, a committee member and session moderator for Panel 1, continued with the theme of what was not said during the workshop and offered two additional ideas: first, instead of fellowships of the type that would compete with those of the National Science Foundation, DOD could offer fellowships aimed at bringing people to its laboratories and immersing them in DOD problems. She described such efforts at the Naval Research Laboratory, which are enthusiastically received. She hastened to add, however, that the money for this comes from non-DOD sources.
• Committee member Leif Peterson noted the tremendous pressure on defense budgets and underscored the need to define requirements well in order to compete effectively for resources. He asked if the workshop has found a good definition of the STEM requirements, and, projecting into the future, what these requirements will be.
• A participant offered a number of reactions to Augustine’s wrap-up comments. He considered the question about whether we have a national problem or a DOD problem to be very perceptive. DOD’s attempts to understand the longer-term supply of STEM employees have been affected adversely by inaccurate metaphors—for example, describing the workforce supply in terms of leakage. Although there is a gradual winnowing that occurs, it would nonetheless be useful to think in terms of behavioral choices of individuals, which can be influenced through inspiring students or by financial and research support in school. The participant noted that another point not heard during the workshop was that ASD(R&E) does not have strong internal evaluation capability on supply, career development, and the retention of its workforce but has only recently focused on internal evaluation. What is lacking is a more rigorous analysis, going beyond “counting noses,” of the benefits of the program relative to the counterfactual case of no such programs. Referring to David Chu’s comment equating scholarships to rents, the participant described an Army program, GrADSO (i.e., Graduate School for Active Duty Service Obligation), that offered to send people to graduate school for 3 more years beyond the 2 years for graduate school that was very popular. A further point, the participant noted, is that there is the profit motive and the assumption that firms will always be short term in their perspective, but we have heard that Lockheed Martin, for example, has some collaborations that are longer term. Lastly, the participant noted that, in response to the question of how to position ASD(R&E) for the uncertain future, all panelists had said to hire more flexible STEM workers. Thus there are two components: the flow of workers, but also their flexibility.
• Committee member Mary Good, continuing in the vein of limitations to DOD’s meeting its STEM needs, suggested that DOD needs to change travel policies and that, for example, the Institute of Electrical and Electronics Engineers (IEEE), whose meetings technically coincide with DOD’s interests, holds 90 percent of its international conferences outside the United States.
PANEL SESSION MODERATOR SUMMARIES
The moderators of the panel sessions next offered their summaries.
Panel 1, on Emerging Science and Technology in the Next 15 Years
Frances Ligler, session moderator, synthesized the list of technology areas discussed in Panel 1 that will be important in the future. They include the following: complexed, trustworthy computers with simple interfaces, including capabilities for design, modeling, communications, and data mining; systems engineering, including
social behavior, human-machine interface, and data-to-decision capabilities; and autonomous systems, including multifunctional materials, robust chemistries, and self-sustaining power. The DOD requires STEM-trained personnel who can work across disciplinary boundaries and interface with sophisticated computational systems. STEM-competent staff are also required with the ability to understand other cultures and to participate in the global S&T community for technology awareness, collaborative exploration of cutting-edge science and engineering, and anticipation of technological surprise.
Panel 2, on Estimating STEM Workforce Needs Under Future Scenarios
Anita Jones, session moderator, explained that there is no dearth of data addressing the STEM workforce needs; there have been numerous surveys made over a long period of time—this is not a new topic. There was, however, an absence of discussion of whether the data are consistent. An interesting point is that temporary-visa holders, who are earning a large percentage of higher degrees, are not leaving in droves but are in fact more likely to stay than not. DOD leadership is thinking about future STEM needs. Jones noted that there is, however, little data on the effectiveness of different kinds of intervention from K-12 or even K-20, and it would be very helpful to have those data; the intervention that you want to make is very different at different stages of an individual’s development. Anecdotally, it would appear that the STEM workforce in DOD and the laboratories and those working in direct support of acquisition are affected by exogenous factors, Jones observed. For example, the BRAC has had an effect on the STEM workforce. Also, a depressed economy delays retirements and causes the seniority system to squeeze the availability of junior STEM appointments, particularly in the laboratories. There is also the looming possibility of reduced federal budgets and a new era of austerity flowing from the debt crisis. There do not, however, appear to be good data on shortfalls in supply, but it will nonetheless be important to pay attention to what DOD and industry have done to be adaptive and to retool.
Panel 3, on Limitations to Meeting Workforce Needs of DOD and the Industrial Base
Sharon Levin, session moderator, noted that the speakers and panelists presented and discussed data on proficiency in mathematics, science, and reading and showed that this has not declined in the United States. There is, however, attrition in the numbers who are in the STEM pipeline as they progress through high school and college and make career choices. Levin mentioned the discussion on the important question of how one defines “STEM workforce.” The definition can depend on the question being asked and the context. The Bureau of Labor Statistics supplies data that fall into the Standard Occupational Classifications, which can be selected and combined in whatever way makes sense. There are also data from the National Science Foundation, which have been structured to be used for some purposes. Nonetheless, there does not appear to be a consensus on what constitutes a STEM worker, nor on how one can go about measuring STEM.
Harold Salzman of Rutgers University thought that there was not a STEM crisis requiring government intervention but that markets will respond to demand signals, although markets are becoming segmented. A lot of lower-level needs with respect to computer science and information technology are being met offshore. This means that if less of the lower-level work is being performed domestically, we can redirect the focus within the United States to higher-level skills.
Committee member Burt Barnow described the difficulty in ascertaining whether there is a labor supply shortage. It is unlikely that we will have any great shortages, but there can nonetheless be mismatches between supply and demand for a particular skill. As the market is changing, he noted, there will be a time needed for transition to a labor force with a different set of skills addressing new technology. Retraining and other such opportunities to update workers’ skills will be important to make the best use of the workforce that we already have.
Dixie Sommers, Bureau of Labor Statistics, presented BLS data and occupational forecasts out to 2018. There will be increasing demand in the future for areas that are of importance to DOD, implying recruiting competition for computer sciences and life sciences and less for engineers, although it should be noted that the latter included both engineers and those working in sales. Here again, the definition of STEM could matter.
Michael Finn, Oak Ridge Institute for Science and Education, presented information on one element of the limitation regarding the non-citizens who received doctorates. Finn observed that the stay rates did not go down. China, India, and other countries in particular have a high percentage of PhD graduates staying in the United States as of 2007, notwithstanding the very high growth rates in their economies. There should be an ample supply, given these stay rates, and DOD has the opportunity to tap into this supply, provided the restrictions on employing non-citizens can be lifted.
Rick Stephens, with the Boeing Company, talked about STEM from the standpoint of practical applications, stressing that it takes almost twice as long as it used to take to train people to perform on Boeing’s assembly lines, for example. He suggests that there needs to be consistency and stability in funding on STEM. In addition, an important attribute of the STEM workforce is the ability to adapt and be flexible.
Levin observed that the data presented during the workshop were by and large sourced from outside DOD. The data provided by DOD were of coarse granularity and relate to the total number of scientists and engineers. There is work underway at DOD and by its contractors to develop a better quantitative understanding of its STEM workforce, but the committee does not have access to this information as yet. DOD will not be able to plan effectively without this information in one comprehensive database.
Daniel Oliver, session moderator, identified some key points and noted some reinforcing linkages to other sessions. Referring to the remarks by Katrina McFarland, Defense Acquisition University, Oliver suggested that, whether we have a crisis or not, having heard the president of the Defense Acquisition University calling her workforce of 147,000 mediocre, we can say that we have an issue. Wesley Harris gave a case study on corrosion and the shortage of 1 million workers, reinforcing a point made in the session of Panel 1 by Lyle Schwartz and underscoring the importance of not overlooking traditional basic disciplines. Referring to an earlier talk in the session on Panel 1, Oliver noted that the best value we get out of 6.1 (i.e., DOD-funded basic research) is the human capital—this is a supply-side approach. However, David Chu, in the Panel 5 session, advised us to deemphasize the supply side. Oliver suggested that DOD shift the emphasis to 6.2 funding, with its more interdisciplinary character. Regarding Carl Wieman and his emphasis on metrics, Vallen Emery asked how we measure learning outcomes. Most of the metrics used in accreditation are not direct but indirect—for example, salaries used as a surrogate for productivity of graduates. Oliver noted that, referring to the tools that Assistant Secretary of Defense Lemnios has in his tool kit, James Gates suggested that Lemnios has a pulpit and access to forums. A final point made during the Panel 4 session was the linkage between STEM and accreditation and accrediting organizations, which can provide a lever for change.
Panel 5, on Ensuring an Adequate Workforce Capability in an Uncertain Future
Robert Hermann, session moderator, described that each panelist in the Panel 5 session had talked about actions that DOD could take. All of the panelists fundamentally talked about organic things to do within DOD. Vallen Emery, Army Research Laboratory, believed that these actions were constrained by the fact that he could not find a larger strategy. This point was re-emphasized by David S.C. Chu, of IDA. Jennifer Byrne, Lockheed Martin Corporation, described what it meant to her to be an engineer in her career. She emphasized the need for collaboration and international engagement. She conveyed a sense of participation and activity on the part of Lockheed Martin, which appears to be addressing STEM. This echoed what was said earlier in the workshop by Rick Stephens of Boeing and Edward Swallow of Northrop Grumman, who expressed the view that they could address the STEM issue with their firm’s organic capabilities. Katherine McGrady, of CNA, focused on priorities that DOD might want to establish in this area and was trying to take the edge off of what she perceived to be an overemphasis on STEM for all of the functions within the government. She made concrete suggestions on DOD’s prioritization of what it can do internally with the resources that it has. Chu proposed that it is demand, not
supply, on which DOD should focus. He also discussed the requirement for citizenship and the pressures that this places on the workforce supply; the globalization of science and technology; and what options there are to mitigate the tension between the two. Hermann also culled points made in the question-and-answer period, including the importance of the technician workforce and the constraints imposed by conflict-of-interest rules on the workforce moving between DOD and defense contractors—a point made by Norman Augustine from his own experience moving from the Department of the Army.
