The limitations faced by the U.S. Department of Defense (DOD) and its industrial base in meeting their science, technology, engineering, and mathematics (STEM) workforce needs in both the near and long term are discussed in this chapter. While there is no evidence that a shortage of workers with the STEM skills necessary to meet the workforce needs of DOD and the industrial base currently exists, except in selected areas such as cybersecurity and selected intelligence fields, meeting the workforce needs associated with emerging technologies in the light of existing workforce trends and DOD policies could be problematic. First this chapter examines some of the supply and demand issues shaping the limitations likely to be faced by DOD and the industrial base in the coming years, and it then recommends some approaches that DOD might take to mitigate these limitations.1
One overarching issue is whether high-performing students enter the STEM pipeline in sufficient numbers to meet the growing demand for STEM-educated workers, as discussed in Chapter 3. Data from the Programme for International Student Assessment (PISA) (OCED, 2011)2 on the science and mathematics literacy of 15-year-olds worldwide suggest that while the proficiency in math, science, and reading of U.S. students lies in the middle rank of member countries of the Organisation for Economic Cooperation and Development (OECD), the percentage of top-performing students achieving level 5 or 6 (the two highest) is nonetheless high compared to that of other member countries; moreover, the United States produces twice as many high-performing students in absolute terms as does the next largest producer—Japan, with 40 percent of the U.S. population (Salzman, 2012; Salzman and Lowell, 2008). Data on the supply side of the pipeline show the following:
• Over the period 1993-2009, interest in pursuing a STEM degree in college remained relatively stable, with the percent of college freshmen intending to major in a STEM field ranging from 21 to 26 percent. 2010 saw a
1 In this chapter, unless otherwise noted, STEM includes the physical sciences, biological/agricultural sciences, mathematics/statistics, computer sciences, and engineering.
new high of 27 percent, largely due to increases in those intending to study the biological/agricultural sciences and engineering.3
• With the exception of computer science, interest in specific STEM bachelor’s degrees has increased consistently over time, with biological/agricultural sciences and engineering the most popular. In computer science, the number of bachelor’s degrees awarded increased dramatically from 1998 to 2004 but fell sharply through 2008 and remained flat in 2009.4
• The percent of all bachelor’s degrees awarded in a STEM field has been stable, ranging from 16 percent to 17 percent over the period 2000-2009.5
• Comparing STEM degrees awarded, on the one hand, to freshmen intentions, on the other, suggests that many students who enter college intending to get a STEM degree do not ultimately graduate with one. For example, approximately 22 percent of freshmen who entered a 4-year college or university in 2006 reported the intention to major in STEM;6 in 2009, only about 16 percent of degrees were in a STEM field.7 This phenomenon is most notable in engineering and, to a lesser extent, the physical sciences.8
• More than 50 percent of the doctorates awarded in the years 2006-2009 were in a STEM field, about a 9 percent increase from the beginning of the decade.9
• Among employed people in 2006 who had graduated in academic years 2003-2005 with a bachelor’s degree in a STEM field, about 63 percent were in a STEM or STEM-related occupation; the comparable number was roughly 81 percent for those graduating with a STEM master’s degree10 and even higher for those at the doctoral level.
The U.S. economy is becoming more dependent on STEM workers. Indeed, as noted in Chapter 3, STEM occupations are projected to grow slightly faster than other occupations. Underrepresented groups such as women and non-Asian minorities are potential target groups for increases in the STEM workforce; STEM occupations pay above the average for these groups, and adding them would increase diversity.
Indeed, recent data indicate the following:
• Women accounted for approximately 57 percent of all bachelor’s degrees earned over the years 2000-2009. The percentage of bachelor’s degrees women earned in STEM fields during this period ranged betwen 10 and 11 percent; for men, however, the percentage of degrees earned in STEM fields ranged between 23 and 25 percent, more than twice the rate for women.11
• Although the share of African-Americans and Latinos in the overall pool of college students has been growing over the past 3 decades to about 26 percent of all undergraduates (including those seeking a 2-year degree), they still account for less than their 33 percent share of the college-age population would imply. Moreover, minorities (other than Asians) are even more underrepresented in STEM fields. While the overall percentage of 24-year-olds
3 The other broad categories under consideration are physical sciences; mathematics/statistics; computer sciences; and engineering. Not included are social/behavioral sciences. See Appendix Table 2-12 in National Science Board (2012).
4 Computer science, narrowly defined, is a relatively small field compared to other degree fields leading to employment in computer-related occupations such as electrical engineering. Recent data (see IPEDS) indicates that bachelor’s degrees in computer science are once again rising. See Appendix Table 2-18 in National Science Board (2012).
5 See Appendix Table 2-19 in National Science Board (2012).
6 See Appendix Table 2-12 in National Science Board (2012).
7 See Appendix Table 2-18 in National Science Board (2012).
8 Note, however, that an examination by Xie and Killewald (2012) of three cohorts of high school seniors (1972, 1982, and 1992) found that “there is little evidence that science suffers from a ‘leaky pipeline’ during the college years that disproportionately steers students away from scientific fields.” Moreover, according to Xie and Killewald, “teenagers’ expectations of their future educational outcomes are full of noise” and “many students shift into and out of science, especially around the time of entering college.” Further information is available in Xie and Killewald (2012).
9 See Appendix Table 2-27 in National Science Board (2012).
10 See Tables 35 and 36 in National Science Foundation (2010).
11 In addition, from 2000 to 2009, the share of all bachelor’s degrees awarded to women declined in computer sciences (by 10 percentage points), mathematics (by 5 percentage points), and engineering (by 2 percentage points). See Appendix Table 2-18 in National Science Board (2012).
in the United States holding a STEM degree is 6 percent, it is only 2.7 percent among African-Americans and 2.2 percent for Latinos (Mervis, 2010).
Not all indicators on the flow of talent into the STEM pipeline are promising:
• While there has been only a slight decline since 1977 in the percent of high school graduates who go on to complete or enroll in a STEM field in college, the percentage of “talented” students (defined as the top quintile on the ACT or SAT) doing so peaked for the 1992/1997 cohort and fell by almost 50 percent for the 2000/2005 cohort, suggesting that these “talented” students are being attracted to degrees and careers other than STEM (Lowell et al., 2009).12
12 STEM includes the life and physical sciences, engineering, mathematics and information technology, and science and engineering technicians (and excludes the social sciences).
• While the retention of STEM graduates in STEM occupations 10 years after high school graduation rose, on average, from 34.8 percent for the 1977/1987 cohort to 43.7 percent for the 1993/2003 cohort, retention in STEM occupations for the most-talented group seemed to decline, although the decline was not statistically significant.13
These data raise questions about the perceived attractiveness of STEM occupations relative to others available to talented individuals. One issue is relative salaries. Evidence suggests that especially for males and U.S. citizens, relative salaries do have a bearing on who does science (Stephan, 2012, pp. 5, 153-156). As data from the Bureau of Labor Statistics (BLS) in Figure 4- 1 (on the preceding page) show, with the exception of some IT-related occupations, jobs in management, finance, the medical professions (primarily, medical doctors and dentists), and law typically pay more on average than STEM occupations in either industry or academe. Moreover, postsecondary faculty positions in STEM fields often require many more years of education and training than for these other occupations, with the exception of some medical specialties.
Furthermore, the earnings profile over a career varies by occupation. For example, according to Stephan (2012), early-career PhD engineers (i.e., those who have had their doctorate for less than 10 years) earn about 1.6 times more than those with a bachelor’s degree in any field who are aged 25-34. Early career PhD physical scientists earn about 1.4 times more, whereas early career PhD life scientists earn less than 1.3 times more (Stephan, 2012, p. 154). The picture is no better when one examines the relative earnings of late career scientists, that is, those 10 to 29 years into their career.
Taking into account the cost of obtaining a PhD in terms of earnings forgone during the years pursuing the degree and subsequent years of training as a postdoctoral fellow, Stephan (2012; p. 157) estimates that the present value of an MBA degree (which typically takes no more than 2 years to complete) is, on average, about $3.2 million dollars, while the present value of the PhD (which often takes 7 or more years to complete) is much lower at about $2 million dollars (Stephan, 2012). Furthermore, Stephan finds that those with MBAs from the best programs can expect to earn (over their lifetime) four to five times more than the average MBA,14 while PhDs hired at top research universities can expect to earn (over their lifetime) only about three times more than the average PhD (Stephan, 2012).
Using data on the composition of the STEM workforce based on the 2000 decennial census (National Survey of College Graduates 2003), Table 4-1 adapted from Levin and Barker (2010) shows the initial entry15 visa of STEM16-educated (by highest degree earned) migrants in the United States as of 2003 by birth region and entry cohort. Several trends are evident:
• The entry visa types have changed over time, with temporary visas now outnumbering permanent visas (i.e., green cards). See Figure 4-2.
• The country of origin of these migrants has changed dramatically over the decades, with migrants from Asia, especially from China and India, growing much faster than migrants from Europe. See Figure 4-3.
• Among the temporary visas types, temporary work visas (primarily Hl-B) have grown the fastest and are now nearly as plentiful as temporary study visas. See Figure 4-2. Nonetheless, commercial firms continue to cite the lack of H-1B visas as a significant problem in hiring needed talent.
13 The change in retention from 44.8 percent to 43.2 percent.
14 At least temporarily, the financial crisis has dampened the expected returns to careers in finance.
15 For a period of at least 6 months.
16 Excludes those educated in the social sciences.
|Region/Country||All Visas #||Green Card %||Temporary Work %||Temporary Study %||Temporary Depend. %||Temporary Other %|
Note: Numbers are subject to rounding errors.
SOURCE: Adapted from Levin and Barker (2010).
FIGURE 4-2 STEM-educated migrants in the United States in 2003 by initial entry visa type and cohort. NOTE: Numbers are subject to rounding errors.
SOURCE: Levin and Barker (2010).
FIGURE 4-3 STEM-educated migrants in the United States in 2003 by birth region (country) and cohort. NOTE: Numbers of visas for Asia includes those for China and India.
SOURCE: Levin and Barker (2010).
Temporary Work Visas
H-1B visas are an important vehicle by which migrants enter the STEM workforce. Such visas likely account for the largest number of highly skilled workers who are entering the country with temporary work visas.17 These visas are typically issued for 3 years and can be renewed for an additional 3-year period. The visa was started in 1990 with a cap of 65,000 per year; in 2001 the cap was tripled to 195,000 per year for 3 years but has now returned to its original level. Universities and non-profit research institutions are exempted from the numerical caps entirely (Wasem, 2012). Starting in 2005, an additional 20,000 visas were granted to students who had received master’s degrees or doctorates from U.S. schools and were thus exempt from the cap of 65,000 (P.L. 108-447). In 2010, the United States issued more than 118,000 H-1B visas. This was down almost 25 percent from the nearly 155,000 issued in 2007, but this is likely only a temporary downturn due to the poor economy in the United States. The available data suggest that most H1-B visa recipients work in science and engineering (S&E) and S&E-related occupations. In 2009, 35 percent of new H-1B visa recipients were employed in the category of computer-related occupations.18
Considering educational attainment, in FY 2009, 58 percent of new H-1B visa recipients had an advanced degree, including 40 percent with master’s degrees, 6 percent with professional degrees, and 13 percent with doctorates. The distribution by degree-level varies by occupation, with 83 percent of mathematical and physical scientists holding advanced degrees (44 percent with doctorates). Among life scientists, 87 percent hold advanced degrees (61 percent with doctorates).19 It is likely that a substantial number of those with PhD degrees are in relatively low-paid postdoctoral positions at U.S. universities.
Overall, these data demonstrate both the use of the H-1B visa as a way for foreign graduates of U.S. schools to undertake postdoctoral training or otherwise pursue careers in the United States, at least temporarily, as well as the importance of the H-1B visa in bringing foreign-educated individuals to the United States, especially in STEM occupations.
The use of H1-B visas to meet STEM workforce needs in the United States is, however, a continuing source of controversy. Industry argues that these workers are meeting shortages of workers with particular skills; others argue that the inflow of these workers may be discouraging U.S. citizens from pursuing education and jobs in these skill areas (Levin et al., 2004). Likely, the answer lies somewhere in between these two positions. It is doubtful that
17 Other categories of temporary work visas include the J-1 Exchange Visa, which is often given to lower-skilled workers and summer visitors and the L-1 Visa issued for intracompany transfers. The latter category has been growing very rapidly and from 2006 to 2010 averaged about 76,000 annually. See Figure 3-36 in National Science Board (2012). Here the NSF definition of S&E is utilized.
18 See Appendix Table 3-19 in National Science Board (2012).
DOD and its industrial base. In the short run, further constraints on H1-B visa entrants may make it more difficult for DOD to recruit citizens if these constraints increase competition for them in the private sector. In the longer run, however, if market forces cause wages to increase following a tightening of H-1B visa policy, more citizens may eventually pursue careers in STEM occupations.
Sandia National Laboratories has a hiring pathway by which a foreign national can become a member of its technical staff. The first stage for the prospective staff member is to become established as a staff member (e.g., in a postdoctoral position or as a limited-term employee). Next they are converted to Foreign National Interim Technical Staff, which includes a requirement that they concurrently pursue a path to U.S. citizenship. Due to the classified nature of Sandia’s work, the prospective staff member must obtain the necessary security clearances and successfully pass a comprehensive counterintelligence investigation. Upon completion of the latter, or receipt of citizenship, the individual becomes a member of the technical staff.
Temporary Study Visas
Non-citizens (primarily with temporary study visas) also play an important role in the production of S&E degrees in the United States, primarily at the master’s and doctoral level.20 In 2009, foreign students earned 38 percent of U.S. S&E master’s degrees. In computer sciences and engineering, however, they earned 46 percent and 43 percent, respectively, of all such degrees.21 And within engineering, they earned more than half of the master’s degrees in electrical and chemical engineering. At the doctoral level, temporary residents earned 35 percent of all S&E degrees awarded in 2009. But they accounted for 57 percent of doctoral degrees awarded in engineering, 44 percent in physical sciences, and 54 percent in computer sciences, although only 29 percent in the biological sciences and 8 percent in medical/other life sciences.22
A large number of these temporary residents, especially at the doctoral level, stay in the United States for at least 5 years after graduation, although the numbers vary by source country. Analysis of data from the Social Security Administration (Finn, 2012) shows an average 5-year stay rate of 62 percent in 2009 for temporary residents receiving a science or engineering doctorate in 2004, with China and India having the highest percentages at 89 percent and 79 percent, respectively. For the 1995, 1997, and 1999 cohorts of foreign national science and engineering doctorate recipients, the stay rates tend to fall slightly from 6 to 10 years after graduation, although stay rates for these cohorts are considerably higher than the 10-year stay rates of earlier cohorts (1991, 1993). While there are tremendous differences in stay rates by source country, these have remained stable over time. Moreover, despite media reports of a “brain drain” of foreign scientists and engineers out of the United States,23 Finn (2012) states that “stay rates are more likely to increase in coming years than to decline” because (1) the share of foreign science and engineering doctoral degrees recipients coming from countries with the highest stay rates has been increasing and (2) those intending to stay in the United States after graduation as reported in the Survey of Earned Doctorates have increased since 2004 (p. 14). Finally, in an earlier report, Finn (2010) posits that the performance of the U.S. economy may affect stay rates, although stay rates declined only modestly during the recession of the early 2000s.
DOD and its associated contractors have special and legitimate needs to hire STEM personnel who can obtain security clearances. Under current practices this generally requires U.S. citizenship, and special problems therefore can arise in hiring in STEM fields in which large proportions of students at U.S. universities are foreign nationals.
20 At the bachelor’s level, non-citizens account for less than 5 percent of the degrees awarded, although in 2009 they accounted for about 9 percent of the degrees in electrical and industrial engineering. See Appendix Table 2-19 in National Science Board (2012).
21 See Appendix Table 2-26 in National Science Board (2012).
22 See Appendix Table 2-28 in National Science Board (2012).
23 See, for example: Herbst (2009); Lee (2011); Wadhwa (2011).
Security clearances are typically classified at one of three levels: confidential, secret, or top secret. Gaining access to sensitive compartmented information or special access programs can also necessitate a top secret clearance. DOD-issued clearances constitute the vast majority of initial clearances (Government Accountability Office, 2010). In the context of the pool of STEM workers available to DOD, the need to obtain a security clearance is a two-fold source of constraint on supply. First, the time required to obtain a security clearance for citizens represents an impediment to success in DOD’s hiring process. Second, this requirement reduces the pool of the potential STEM workforce for DOD in fields in which non-citizens represent substantial fractions. While this is not much of a problem at the BS level in engineering, where international students represent only a small percentage,24 it is a significant problem for positions requiring graduate engineering degrees, where the percentage of temporary residents, as noted earlier, is much higher.
A recent study of personnel security clearances found that progress has been made in reducing the time to adjudicate applications (Government Accountability Office, 2010). Specifically, the report noted that DOD was able to meet the goal of adjudicating 90 percent of its applications within 60 days. The process also allows for the DOD to give interim clearances at the secret level once an investigation on an individual has been opened and no initial problems have been identified (Department of Defense, 1999; Secretary of the Navy, 2006). (In contrast, temporary access at a top secret level can be granted only if the applicant already has a secret or a confidential level clearance.) Otherwise, a secret level interim clearance can be given to those requesting top secret level access and only if a local review of the personnel security questionnaire (PSQ) is found to reveal no eligibility issues. Those DOD commands that do not impose restrictions for facility access according to security clearances could potentially hire STEM researchers, with the caveat that their continued employment requires a security clearance at the appropriate level. Despite recent improvements, it is possible that some top STEM talent, who can be hired on the spot by private-sector recruiters, may be deterred from pursuing DOD careers by delays in their appointments because of clearance issues.
The system of personnel security clearances is far from being the only set of controls placed on those performing defense work or on the goods and services they produce, including their export or so-called deemed export. A technology on the U.S. Munitions List, administered by the U.S. Department of State, is subject to export controls (Congressional Research Service, 2009). Another set of controls applies to so-called dual-use technologies on the Commerce Control List (CCL), administered by the U.S. Department of Commerce. Activities conducted within the United States, such as sharing knowledge of a technology with a foreign national residing domestically, may constitute a deemed export, requiring a license or exemption under the International Traffic in Arms Regulations (National Research Council, 2009a, p. 34). The system of controls and their implementation is formidable, and interested readers are urged to consult the substantial secondary literature on this topic (Center for Security and International Studies, 2005).
Unfortunately, U.S. government policies regarding the funding of higher education (and particularly graduate education) in STEM fields lack coordination with policies regarding temporary visas for education and work as well as for visas for permanent residence. For example, large amounts of financial support from federal agenagencies—via research grants—are used by U.S. universities to finance the graduate education of international students in STEM fields. Immigration policies allow universities essentially unlimited access to such students, as well as to international “postdocs,” i.e., those who have earned PhDs in countries other than the United States and then come to work as postdocs in U.S. university labs with financial support from federal research grants. Surprisingly there are almost no credible data on this apparently large and growing population of international postdocs, though it appears that the largest country of origin is China. Meanwhile, as noted earlier, more than 100,000 temporary skilled workers each year are admitted with H1-B visas, mostly in STEM and related IT fields, for temporary but multi-year work. Yet the number of permanent visas available based on these same skills is much smaller, resulting in large backlogs of temporary visa holders seeking permanent visas.25
24 In 2009, only 5.8 percent of the undergraduate degrees in engineering were earned by temporary non-resident students. See, Appendix Table 2-19 in National Science Board (2012).
25 It now can take as long as 6 to 10 years to obtain the coveted “green” card that grants permanent residency to skilled-immigrant workers from China and India, for example (Wadhwa et al., 2007). Moreover, Hira (2010) argues that “most of the top users of both the H-1B and L-1 visa programs sponsor very few, if any, of their workers for permanent residence.”
The DOD must compete with the civilian sector for job opportunities available to STEM-trained individuals. Data provided by the Bureau of Labor statistics (BLS) can help gauge the strength of competing demands from the civilian sector. BLS provides data on employment and wages, including employment projections by occupation (see Chapter 3 of this report). It should be noted, however, that in addition to the array of occupations normally included in STEM employment numbers—engineers, math and computer scientists, and life and physical scientists—BLS included STEM technicians, architects, postsecondary teachers in STEM fields, STEM managers, and those in STEM-related sales jobs in a recent study of the STEM workforce (Cover et al., 2011, pp. 3-15).
• As reported in Chapter 3, of the 7.3 million employed in STEM in the civilian sector in 2010 (accounting for 5.1 percent of the overall workforce), the greatest number is employed in computer-related occupations.26
• For the period 2010 to 2020, STEM employment is projected to grow by 16.9 percent, which is slightly higher than the projected growth rate of 14.3 percent for the workforce as a whole.
Although these are projections premised on assumptions regarding future GDP growth which are themselves subject to considerable uncertainty, they do at least suggest relatively strong growth in the civilian sector in those occupational categories likely to be most sought after in the coming years by DOD and the industrial base. Moreover, Sauermann and Roach (2012) found in a survey of PhD students that “a faculty research career is the career path most often considered ‘extremely attractive’ and ranks among the most desirable careers for over 50% of life scientists and physicists,” suggesting that DOD will continue to face competition from academic institutions for PhD-level scientists. There are several other important issues that DOD and its industrial base must confront in order to meet its STEM workforce needs.
The DOD’s ability to pay uniformed, civilian, and, indirectly, contractor STEM workers competitive salaries will be a further issue to consider in developing DOD’s STEM talent. Data from the Congressional Budget Office and the Project on Government Oversight suggests that STEM workers above the bachelor’s level are paid less in the civilian federal workforce than in the private sector (POGO, 2011). For example, CBO finds that individuals in the federal workforce with a professional or doctoral degree earn (in wages and benefits) about 18 percent less than their counterparts in the private sector (Congressional Budget Office, 2012). Other federal agencies such as NIH, NSF, and EPA have Title 42 authority “to appoint highly qualified consultants, scientists and engineers at a pay scale [up to $250,000 per year] outside civil service laws described under Title 5” (National Research Council, 2010). Salaries may impact the STEM pipeline by providing a signal to prospective STEM majors.27 For example, the number of persons graduating with a degree in computer science increased, with a lag of a few years, as wages for computer programmers increased in the early 1990s. Furthermore, the number of persons graduating with a degree in computer science declined as wages stagnated in the early 2000s (Figure 4-4). Similarly, Ryoo and Rosen (2004), in their examination of the engineering labor market, found that engineering enrollment decisions appeared to be sensitive to engineering career prospects (as measured by the present discounted value of earnings in engineering relative to alternative professions). Lastly, a study of science and engineering PhD students by Roach and Sauermann (2010) found that students “concerned with salary, access to resources, and the desire to conduct downstream research and development” are more likely to prefer a career in industry over a career in academia. These results suggest that pay is an important aspect of the value proposition the DOD can offer to a prospective employee.
That said, extended discussion with a senior representative from DOD’s Office of Personnel and Readiness-suggested that there is no shortage of qualified applicants for the positions advertised on the DOD website and in
26 Postsecondary STEM teachers are excluded from the estimate of STEM employment since the information available from the Bureau of Labor Statistics’ Employment Projections Program does not distinguish teaching field.
27 Freeman (1976) established that “the supply of new entrants to engineering is highly responsive to economic conditions.”
FIGURE 4-4 Computer science bachelor’s degree awards and computer programmer real mean salaries, 1992-2008.
SOURCE: Kuehn and Salzman (2013).
trade papers.28 This is particularly true in the current job climate where available positions are scarce. The committee was advised that applicants come because they find attractive the opportunities for greater responsibility in the government labs during the first 5 years of work than there would be in private labs. DOD entry level compensation was declared to be sufficiently attractive, particularly when one includes bonuses. Another DOD source noted that at the top end of the salary scale, somewhere around $160,000, the private sector enjoys a distinct advantage since the government is not competitive. In these circumstances, one has to be concerned with the appropriateness of the skills of the persons being hired and the potential deleterious impact on activities.
To attract top talent, the work DOD and its industrial base offers must offer sufficient challenge and importance to excite the most creative and highly skilled workers, and to motivate them to achieve peak performance. “Pay for performance” personnel policies can be implemented, but if the work is not sufficiently exciting, pay alone will not be enough.
In the past, major DOD procurement programs have been a sufficient source of widely visible program development challenges, attracting and motivating the talented workforce DOD desired. However, major DOD procurement programs are decreasing. For instance, there are currently only two major aircraft programs in development, following a steep decline in numbers of new starts since the Second World War (Figure 4-5). There are, however, exciting smaller-scale programs in DOD in a number of areas that may be less visible to the general public. For instance, the Defense Advanced Research Projects Agency (DARPA) continues to support advanced concept technology demonstrations across a spectrum of disciplines.29 The Special Operations Command sponsors cutting-edge field experimentation in the academic environment of the Naval Postgraduate School. These types of programs tend to precede the competitive phase of the acquisition process, which leaves them relatively free of
28 Pasquale “Pat” Tamburrino, deputy assistant secretary for civilian personnel policy, personal communication.
29 Note, however, that some have argued that DARPA has become too risk averse. See, for example, Ignatius (2007).
FIGURE 4-5 Number of new fighter and bomber starts per decade.
SOURCE: Carlson and Chambal (2008).
bureaucracy. Innovation and creativity are encouraged. High-risk projects are allowed to “fail,” and researchers, practitioners, and students are encouraged to push the envelope.
Another such example is the Rapid Reaction Technology Office (RRTO), later folded into the Rapid Fielding Office within ASD(R&E), which was charged with developing counterterrorism technologies and employed rapid prototyping. DOD established the RRTO in 2006 in response to the constantly evolving threat of asymmetric warfare, including, for example, the use of improvised explosive devices (IEDs) in the Iraq and Afghanistan theaters of operation. Established under the director, defense research and engineering, it focused on developing technologies that can mature in 6 to 18 months for the purpose of countering insurgency and irregular warfare. The RRTO provides a diverse set of quick-response capabilities for counter-terrorism while attempting to stimulate interagency coordination and cooperation. The office operates without a formal charter or governing document, and the director has much flexibility for carrying out the mission. Approximately 50 percent of the office’s projects have resulted in fielded technologies, altered concepts of operation (CONOPS), or other concrete changes in larger systems. Such projects included the Persistent Threat Detection System for persistent ground surveillance through a tethered aerostat with an embedded camera; a Biometric Automated Toolset for screening personnel in mobile applications; and the SKOPE intelligence cell, a joint analytic cell with the National Geospatial Intelligence Agency, U.S. Special Operations Command, and U.S. Strategic Command. Strategic investment by DOD in programs of this nature appears to be an important cornerstone of creating an increasingly attractive workforce environment (National Research Council, 2009b).
In a similar vein, the Lockheed Martin Skunk Works® has, over its nearly 70-year history, created breakthrough technologies and landmark aircraft that continually redefine flight. Guided by the mantra “quick, quiet, and quality,” the Skunk Works® requires a flexible workforce capable of quickly forming and disbanding interdisciplinary project teams. To meet this need, the Skunk Works® uses a matrix organization that minimizes paperwork and delays in moving people between teams. Core engineering groups maintain skill sets and tools to support their disciplines. Program managers draw their teams from these talent pools. Likewise, NASA developed its Engineering and Safety Center (NESC) in 2003 to provide an independent test, analysis, and assessment capability to NASA programs and projects. The NESC operates independently of mission directorates and reports to the Office of the Chief Engineer. The NESC operates through technical discipline teams (TDTs), each led by an agency-recognized NASA Tech Fellow, who is an outstanding senior-level engineer or scientist with distinguished and sustained records of technical achievement. The fellows provide leadership and act as role models for NASA discipline engineering communities beyond the TDTs which are drawn not only from NASA but also from other federal agencies, industry, and universities, making them diverse teams that can provide robust, creative solutions to complex problems. Another
government agency supporting high-risk ventures is the Advanced Research Projects Agency-Energy (ARPA-E), which funds specific high-risk, potentially high-payoff, energy research and development projects. ARPA-E has been set up to be a lean and agile organization with special hiring authority to bring on program directors and other program leadership with the ability to offer limited-term rotational assignments. Thus, individuals from all sectors are able to assume temporary positions lasting roughly 3 years. The agency empowers them to make technical and programmatic decisions for the projects they oversee (PCAST, 2010; Yehle, 2011).
Until 1998, the DOD budget included a category “6.3B” for systems advanced development that supported rapid prototyping programs (National Research Council, 2001).
For any organization seeking to maximize the productivity of its professionals in science and engineering, high-quality, up-to-date facilities and equipment are essential. In addition, the availability of such facilities and equipment enhances the recruitment of talented scientists and engineers.
The committee is aware of a series of reports that describe limitations experienced by DOD research labs in this regard.30 However, the most recent such report is already several years old, and it appears that there was little action in response to the recommendations during subsequent periods of budgetary stringency. The committee agrees with the perspectives expressed in these reports. For example, a recommendation from a 2001 report noted that DOD “should continue to pursue world-class status for the Service laboratories” and emphasized that this should be done “not only to obtain the highest-quality results from its research, but also to attract superior scientific and engineering personnel who want to work where the best research is done” (National Research Council, 2001). DOD’s need for outstanding science and engineering in support of its increasingly technical missions does require that serious attention be paid to ensure that facilities and equipment available to DOD scientists and engineers are of the highest quality.
In order to attract the highest-quality workers, the DOD should consider personnel policies as they relate to the ability of DOD to attract, retain, and develop the STEM workforce it needs. In a July 2010 study entitled Defense Acquisition Workforce Modernization from the Center for Public Policy and Private Enterprise at the University of Maryland, the authors posit that “to effectively develop the required human capital for the modern acquisition environment, we believe that DOD should enhance its recruitment processes; improve the hiring process; strive for quality not quantity; provide compelling wages; incentivize employees for improved performance; and, incentivize employees for additional training and education” (Gansler et al., 2010). These imperatives can be generalized to the STEM workforce as a whole. For example, sabbatical and expanded internship programs, as well as online and anytime/anyplace programs, address not only recruiting and retention issues but also the increasingly interdisciplinary competencies required by the workforce.
DOD will be have to compete for scientific talent in the changing environment of globalization. Witness the growth in higher education and the development of technological infrastructures in S&E in China and India, two large suppliers of U.S. STEM graduates. In these countries, there is going to be a demand for these graduates that did not exist previously. Even Russia, whose scientific enterprise has suffered from the migration of scientific talent since the breakup of the Soviet Union, has witnessed the appointment of a U.S. engineer as president of the new graduate research university—the Skolkovo Institute of Science and Technology—a collaborative effort with MIT.31
30 For more information see, National Research Council (1990, 2001, 2005). See also JASON (2009), p. 26.
31 According to press releases from MIT, “This institution aims to break new ground in bringing together Russian, US and global research and technology—and in integrating teaching, research, innovation and entrepreneurship” (MIT News, 2011).
Furthermore, some countries are trying to attract back home that segment of their native-born scientific talent that has been educated in the United States.32 Small countries such as Singapore are luring scientific talent from the best universities and labs in the world in order to build a cutting-edge science enterprise that is intended to transform Singapore “into a knowledge-based, innovation-driven economy.”33 China and Taiwan, among others, are also actively seeking foreign talent and are now rapidly developing their own technological infrastructures, making STEM careers in such graduates’ home countries increasingly attractive (Wadhwa et al., 2009). In this more competitive environment, one would expect that the quality of the research environment as well as the remuneration paid to scientists and engineers will become increasingly important, particularly for foreign students who have come to the United States seeking economic success.
To counter these competitive pressures, DOD should offer highly competitive career opportunities for outstanding scientists and engineers. DOD advertisements for STEM applicants may be neither visible enough nor attractive enough in conveying the exciting research underway under DOD auspices. Because these jobs are about national security, they should be seen and advertised as critical to the defense of the nation, thus appealing to patriotic instincts. Awareness efforts could be informed by advertising campaigns such as those that have been developed by the Marines. These campaigns appear to be effective in creating a sense of purpose (the defense of the nation, referencing “the Marines”), exclusivity (referencing “the Few”), and a profound sense of superiority (referencing “the Proud”). DOD scientists and engineers are not uniformed combatants, but their work is an essential part of our national security mission and the U.S. role in promoting global peace. A concerted effort on the part of DOD to bring awareness to the vast contributions of its highly diversified STEM workforce could go a long way toward moving DOD into the vanguard of crafting a “heroic” image for the agency scientists and engineers whose work is vital to U.S. national security. This was in fact a major factor in attracting talent during the Cold War era.
The committee knows of no DOD recruiting effort for civilian scientists and engineers that is comparable to those for the uniformed services, and yet there could be substantial commonalities with the military system. DOD scientists and engineers play central roles in creating the tools with which the military service members operate on a daily basis. The development of a sophisticated civilian recruiting effort that identifies DOD scientists and engineers as working closely with military personnel in ways that are critical to national security could be highly effective. Moreover, many technically oriented students are attracted to intriguing and unique applications of science and technology, some of which are being led by DOD. These include globally controlled, unmanned aerial systems (UASs); “smart” weapons; sophisticated night vision; and the integration of complex communication and data that can be deployed in real time in battlefield conditions. Outreach and recruiting efforts could be amplified by offering highly qualified young science and engineering students internship opportunities in R&D in appropriate DOD labs, thereby exposing them to the exciting science and engineering challenges faced by the DOD.
There is no question that STEM disciplines will continue to grow in importance as defense capability becomes more technology-driven. To respond to this, one possibility would be to create a specialized recruiting function within the DOD that would be responsible for STEM recruitment and hiring. A second task of this office would be the identification of a list of higher education institutions that produce the students who best fit the demands of the workforce in the disciplines of the greatest interest to DOD (e.g., civil engineering, electrical engineering, petroleum engineering, etc.). A place to start would be with institutions that currently offer courses at DOD facilities and others with which cooperative structures exist, for instance, the Community College of the Air Force, eArmyU, and others.
32 See for example, Lim (2011); Sharma (2011).
33 The Agency for Science, Technology and Research (A*STAR) is the lead agency for fostering world-class scientific research and talent for Singapore. A*STAR oversees 14 biomedical sciences and physical sciences and engineering research institutes, and six consortia and centers, located in Biopolis and Fusionopolis as well as their immediate vicinity. It also supports educational programs in S&E at all levels of instruction. For more information, see http://www.a-star.edu.sg/.
In addition to the issues noted above, there are several exogenous factors that may have an impact on DOD’s ability to hire and manage an effective STEM workforce, including the following:
• The failure of the congressional “Super Committee” to reach agreement on budget cuts, which will likely result in another massive reduction in the DOD budget in the coming years. While the White House and Congress agreed last summer on $487 billion in cuts to defense spending over the next 10 years, even deeper cuts are threatened if Congress fails to pass a new plan for deficit reduction. In that case, the Pentagon budget will be cut by a total of roughly $1 trillion over a decade, beginning in January 2013.34
• The history of large swings in DOD funding.35 Defense spending increased sharply to over 9 percent of GDP in the mid-1960s as U.S. involvement in Vietnam expanded. After large-scale withdrawal from Vietnam began in1969, defense spending as a share of GDP fell to less than 5 percent of GDP by the end of the next decade. The Soviet invasion of Afghanistan prompted an increase in defense spending to about 6 percent of GDP during the early 1980s. After the Berlin Wall was opened in November 1989 and communist governments in central and Eastern Europe collapsed, defense spending as a share of GDP dropped to the historically low level of about 3 percent. Defense spending increased again to nearly 5 percent of GDP after the attacks on September 11, 2001, and the wars in Afghanistan and Iraq began. In the committee’s experience, DOD has dealt with tightened budgets by reducing, often disproportionately, funding for workforce training and development. In addition, reductions in the STEM workforce seem to have been carried out in a manner having more to do with numbers and less with justification premised on impact to military capabilities or quality of the workforce.
• The need for DOD to manage a potentially large increase in retirements when the recession ends and housing and securities markets rebound. As reported in Chapter 3, about one-third of the DOD civilian STEM workforce is eligible to retire (see Figures 3-19 and 3-20). Moreover, this eligibility rate is more than double the estimated retirement eligibility rate of the defense industrial base workforce. While the actual rate of retirement is low for both workforces, DOD is likely at greater risk from future retirements. The need to recruit, develop, and retain highly skilled employees across both traditional and emerging STEM disciplines such as translational computing, autonomous systems, systems biology, innovative materials, and efficient manufacturing should be a DOD priority.
Finding 4.1. Stable funding for the recruitment and development of STEM human resources is essential to their effective management.
Recommendation 4.1. The DOD should fund STEM recruitment and development in a manner that facilitates stability, such as multi-year programming, “one color” of money for STEM related costs,36 or funding based on a percent of total obligational authority. This would facilitate stability for long-term STEM investments and greater consistency across and within the services. In addition, DOD should require all services to justify, as part of the approval process, STEM-related manpower reductions in terms of impact on technology-based capabilities and, where appropriate, whether there has been sufficient return on investment from those who have recently completed postsecondary education paid for by the government.
34 See for example, Barnes and Entous (2012).
35 See for example, Austin and Levit (2010).
36 Congress provides funds to the DOD in different appropriation accounts (“colors of money,” a term of art used in day-to-day discussions within the DOD). DOD military personnel are paid from one account—the “MILPERS” account. DOD civilians, especially STEM-related civilians, can be paid from more than one account, such as the Operation & Maintenance (O&M) account or the Research, Development, Test, & Evaluation (RDT&E) account. These accounts are managed with different sets of rules including programming procedures, approval levels of reprogramming, and duration of funds. Therefore, civilians paid with different “colors of money” are funded differently in terms of both procedures and funding levels. This can cause significant disruptions and disparities across the services when it comes to employment programming, hiring, training, RIFs, awards, and so on.
Finding 4.2. The U.S. STEM workforce is heavily dependent on non-citizens. DOD will need to reassess its requirement for security clearances for many STEM positions along with the processes by which many of its systems are developed and procured.
Recommendation 4.2. The DOD should find creative ways to hire STEM-qualified non-U.S. citizen personnel to support and advance designated S&T activities. Consideration should be given to those aspects of programs that are not classified and those that could accommodate lower-level clearances. The process should be codified and repeatable to ensure a sufficient number of candidates under appropriate circumstances. It is understood that this could require both policy and legislative changes, including but not limited to adapting the H1-B program, and the issuance of exemptions under ITAR and other applicable laws and regulations.
Finding 4.3. The United States, including DOD and its industrial contractors, is competing in an ever-growing world market for top scientific and engineering talent. For the DOD to recruit top STEM talent in competition with commercial firms, universities, and others, it must commit to improving the STEM workforce environment. The DOD must become, and be perceived as, an attractive career destination for the most capable scientists, engineers, and technicians, who are in great demand in the global talent marketplace.
Recommendation 4.3. The DOD should strengthen its ability to recruit, educate, and retain top STEM talent by offering competitive salaries and a constructive work environment, providing challenging and interesting problems in the workplace, enabling existing talent to keep up with the newly emerging scientific trends, and providing opportunities for the retraining of its STEM workforce to meet changing scientific and technological needs.
Finding 4.4. Because of the increasing acquisition costs of major systems and continuing pressures on DOD budgets, the number and variety of major weapons being developed and fielded have shrunk significantly in recent decades. This dynamic has a dampening effect on recruiting for the DOD STEM workforce.
Recommendation 4.4. The DOD should support, wherever possible, experimental and rapid-prototyping programs that push the cutting edge of science and engineering, in order to both maximize new technology applications and to attract the best and brightest STEM workers.
Finding 4.5. The DOD has centers of excellence across its own institutions, but the quality and the modernity of both facilities and equipment vary widely, marginalizing DOD’s ability to compete broadly for top STEM talent.
Recommendations 4.5. The DOD should establish high standards of quality for both facilities and equipment and fund them appropriately.
Finding 4.6. DOD’s personnel policies with regard to recruiting, hiring, paying, retaining, and incentivizing additional training and education are not currently optimized for maintaining the best STEM workforce.
Recommendation 4.6. The DOD should consider changes in personnel policy that would enable it to move more nimbly to make competitive hiring offers in DOD-critical scientific and engineering fields. Some of these changes can be made internally within DOD. Where this is not currently possible, DOD should seek legislative and/or regulatory relief. The following changes warrant consideration by DOD:
• More active outreach and recruitment efforts, aimed at civilian hires, of needed scientists and engineers that emphasize the many exciting technologies that are being developed by DOD and their potential contribution to the nation;
• New measures to expedite recruitment offers for occupations in which DOD determines that it must compete with more nimble corporate recruiters;
• Additional authority to expedite security clearances needed for such positions, including permission for temporary hiring into non-sensitive roles pending confirmation of security clearance;
• Actions to protect or “ring-fence” science and engineering positions determined by DOD to be critical capabilities, thereby protecting the loss of such capabilities due to future RIFs and hiring freezes; and
• Further provisions to incentivize DOD scientists and engineers to seek additional continuing education and training in rapidly developing areas of science and technology.
Austin, D.A., and M.R. Levit. 2010. Trends in Discretionary Spending. Washington, D.C.: Congressional Research Service.
Barnes, J.E., and A. Entous. 2012. Pentagon to lay out next year’s budget cuts. Wall Street Journal, January 25.
Carlson, B., and S. Chambal. 2008. Senior leader perspective. Developmental planning: The key to future war-fighter capabilities. Air and Space Power Journal 22(1):3.
Center for Security and International Studies. 2005. Security Controls on the Access of Foreign Scientists and Engineers to the United States. Washington, D.C.: Center for Security and International Studies.
Congressional Budget Office. 2012. Comparing the Compensation of Federal and Private-Sector Employees. Washington, D.C.: Congress of the United States.
Congressional Research Service. 2009. The U.S. Export Control System and the President’s Reform Initiative. Washington, D.C.: Government Printing Office.
Cover, B., J.I. Jones, and A. Watson. 2011. Science, technology, engineering, and mathematics (STEM) occupations: A visual essay. Monthly Labor Review 134(May):3-15.
Department of Defense. 1999. Personnel Security Program Regulation. DOD 5200.2. DOD, April 9. Available at www.dtic.mil/whs/directive/corres/pdf/52002p.pdf.
Finn, M.G. 2010. Stay Rates of Foreign Doctorate Recipients from U.S. Universities, 2007. Oak Ridge, Tenn: Oak Ridge Institute for Science and Education.
Finn, M.G. 2012. Stay Rates of Foreign Doctorate Recipients from U.S. Universities, 2009. Oak Ridge, Tenn: Oak Ridge Institute for Science and Education.
Freeman, R.B. 1976. A cobweb model of the supply and starting salary of new engineers. Industrial and Labor Relations Review 29(2):236-248.
Gansler, J.S., W. Lucyshyn, and M. Arendt. 2010. Defense Acquisition Workforce Modernization. University of Maryland: Center for Public Policy and Private Enterprise.
Government Accountability Office. 2010. Personnel Security Clearances: Progress Has Been Made to Improve Timeliness but Continued Oversight Is Needed to Sustain Momentum. Washington, D.C.: Government Accountability Office.
Herbst, M. 2009. Why the U.S. Is Losing Foreign Graduates. Available at http://www.businessweek.com/technology/content/mar2009/tc20090318_162454.htm (accessed October 3, 2012).
Hira, R. 2010. Bridge to Immigration or Cheap Temporary Labor? The H-1B & L-1 Visa Programs Are a Source of Both. EPI Briefing Paper. Document 257. Economic Policy Institute, February 17. Available at www.epi.org/publication/bp257/.
Ignatius, D. 2007. The ideas engine needs a tuneup. Washington Post, June 3.
JASON. 2009. S&T for the National Security. McLean, Va.: MITRE Corporation.
Kuehn, Daniel, and Harold Salzman. 2013. The labor market for new engineers. U.S. Engineers in the Global Economy. Richard Freeman and Harold Salzman (eds.). National Bureau of Economic Research, forthcoming.
Lee, B. 2011. Reverse Brain Drain in the U.S. Available at http://www.pbs.org/wnet/need-to-know/the-daily-need/reverse-brain-drain-in-the-u-s/11027/ (accessed October 3, 2011).
Levin, S.G., and A. Barker. 2010. Studying Foreign Talent in the Science and Engineering Workforce Final Report to the Alfred P. Sloan Foundation. Grant number 2008-5-29 SEW.
Levin, S.G., G.C. Black, A.E. Winkler, and P.E. Stephan. 2004. Differential employment P patterns for citizens and non-citizens in science and engineering in the United States: Minting and competitive effects. Growth and Change 35(4):19.
Lim, L. 2011. China Aims to Renew Status as Scientific Superpower. Available at http://www.npr.org/2011/08/01/138837512/china-aims-to-renew-status-as-scientific-superpower (accessed October 3, 2012).
Lowell, B.L., H. Salzman, H. Bernstein, and E. Henderson. 2009. Steady as She Goes? Three Generations of Students through the Science and Engineering Pipeline. Paper presented at Annual Meetings of the Association for Public Policy Analysis and Management, Washington, D.C., October 9.
Mervis, J. 2010. New Answers for Increasing Minorities in Science. Available at http://news.sciencemag.org/scienceinsider/2010/09/new-answers-for-increasing-minorities.html (accessed October 3, 2012).
MIT News. 2011. Skolkovo Foundation and MIT to Collaborate on Developing the Skolkovo Institute of Science and Technology. Available at http://web.mit.edu/newsoffice/2011/skolkovo-agreement-1026.html (accessed March 28, 2012).
National Research Council. 1990. Recruitment, Retention, and Utilization of Federal Scientists and Engineers. Washington, D.C.: National Academy Press.
National Research Council. 2001. Review of the U.S. Department of Defense Air, Space, and Supporting Information Systems Science and Technology Program. Washington, D.C.: National Academy Press.
National Research Council. 2005. Assessment of Department of Defense Basic Research. Washington, D.C.: The National Academies Press.
National Research Council. 2009a. Beyond “Fortress America”: The National Security Controls on Science and Technology in a Globalized World. Washington, D.C.: The National Academies Press.
National Research Council. 2009b. Experimentation and Rapid Prototyping in Support of Counterterrorism. Washington, D.C.: The National Academies Press.
National Research Council. 2010. The Use of Title 42 Authority at the U.S. Environmental Protection Agency. Washington, D.C.: The National Academies Press.
National Science Board. 2012. Science and Engineering Indicators 2012. Arlington Va.: National Science Foundation. National Science Foundation. 2010. Characteristics of Recent Science and Engineering Graduates: 2006. Available at http://www.nsf.gov/statistics/nsf10318/pdf/nsf10318.pdf (accessed March 27, 2012).
OCED (Organisation for Economic Cooperation and Development). 2011. Lessons from PISA for the United States, Strong Performers and Successful Reformers in Education. Paris: OECD Publishing.
PCAST. 2010. Report to the President on Accelerating the Pace of Change in Energy Technologies Through an Integrated Federal Energy Policy. Washington, D.C.
POGO (Project on Government Oversight). 2011. Bad Business: Billions of Taxpayer Dollars Wasted on Hiring Contractors. Available at http://www.pogo.org/pogo-files/reports/contract-oversight/bad-business/co-gp-20110913.html (accessed September 13, 2011).
Roach, M., and H. Sauermann. 2010. A taste for science? PhD scientist’s academic orientation and self-selection into research careers in industry. Research Policy 39(3):12. Ryoo, J., and S. Rosen. 2004. The engineering labor market. Journal of Political Economy, 112(S1):40.
Salzman, H. 2012 (unpublished). The New STEM Labor Market Segmentation: Implications for Meeting Workforce Needs of DoD and the Industrial Base. Presentation to the Workshop on STEM Workforce Needs for the U.S. Department of Defense and the U.S. Defense Industrial Base.
Salzman, H., and B.L. Lowell. 2008. Making the grade. Nature 453:2.
Sauermann, H., and M. Roach. 2012. Science PhD career preferences: Levels, changes, and advisor encouragement. PLoS ONE 7(5):e36307.
Secretary of the Navy. 2006. Department of the Navy Information Security Program.
Sharma, Y. 2011. ASIA: “Brain reclaim” as talent returns from West. University World News. Available at http://www.universityworldnews.com/article.php?story=20110415201701401 (accessed October 3, 2012).
Stephan, P. 2012. How Economics Shapes Science. Cambridge, Mass.: Harvard University Press.
Wadhwa, V. 2011. We need to stop America’s brain drain. Washington Post, October 4, 2011.
Wadhwa, V., G. Jasso, B. Rissing, G. Gereffi, and R.B. Freeman. 2007. Intellectual Property, the Immigration Backlog, and a Reverse Brain-Drain: America’s New Immigrant Entrepreneurs, Part III. Kansas City, Mo.: Ewing Marion Kauffman Foundation.
Wadhwa, V., A. Saxenian, R. Freeman, G. Gereffi, and A. Salkever. 2009. America’s Loss Is the World’s Gain: America’s New Immigrant Entrepreneurs. Kansas City, Mo.: Ewing Marion Kauffman Foundation. Wasem, R.E. 2012. Immigration of Foreign Nationals with Science, Technology, Engineering, and Mathematics (STEM) Degrees. Washington, D.C.: Library of Congress.
Xie, Y., and A. Killewald. 2012. Is American Science in Decline? Cambridge, Mass.: Harvard University Press.
Yehle, E. 2011. No home run yet for ARPA-E, but chief says “motivated” team’s on track. Greenwire, April 7.
This page intentionally left blank.