This is the first in a series of PRB articles about the S&E workforce in the United States. The second article will examine the interstate migration and stability of the S&E workforce across states.

Much of the research and debate on science and engineering in the United States has focused on a single question: Does the country have enough scientists and engineers to compete in the increasingly high-tech global economy? Many business groups and federal agencies believe that there is a deficit of high-tech workers, while other observers argue that the supply of workers is adequate and shows “no sign of impending shortages.”1

What is often overlooked in this debate is the imbalance of science and engineering (S&E) workers across different parts of the United States. In general, S&E jobs are concentrated in states on the East and West coasts, with fewer opportunities in the Midwest and South.

Louisiana, for example, has difficultly attracting or retaining scientists and engineers because of a lack of job opportunities for individuals with post-graduate degrees or technical skills. Other states such as California are “importers” of high-tech workers: They recruit scientists and engineers from the rest of the United States and increasingly from other countries, creating geographic variation in the technical skills of the U.S. workforce.

The uneven distribution of S&E employment opportunities is important because it points to critical economic and educational differentials between states, rural/urban areas, and racial/ethnic groups. Today, most states strive to enhance the growth of science- and technology-based businesses. Such businesses have been recognized as catalysts for economic growth and for the development of human resources.2 However, to participate in the new knowledge-based economies, these states must have the capacity to educate and retain qualified workers.

An Economic Catch-22 for States With Mostly Unskilled Workers

States gain several advantages from having a highly skilled workforce. Skilled workers earn higher salaries and generate more tax revenue compared with unskilled workers. In 2004, the median annual income for people in S&E occupations was $55,871, compared with $26,412 for people in non-S&E jobs.3

Table 1 shows the median earnings in states with the highest proportions of S&E workers, compared with states with the lowest proportions of S&E workers. Median earnings in Maryland, near the top of the rankings, were more than $15,000 higher than median earnings in South Dakota, which ranks near the bottom. Skilled workers are also less likely to be unemployed and more likely to have health and retirement benefits, which reduces the potential burden on taxpayers and service providers.

Table 1

Median Earnings in High- and Low-Tech States, 2004

Proportion of labor force
in S&E jobs
Median earnings
(all workers)
High-tech states
Low-tech states
South Dakota

Source: Population Reference Bureau analysis of the 2004 American Community Survey (ACS).

The lack of S&E employment opportunities in states like South Dakota stems from a historical focus on farming, resource extraction, and low-wage manufacturing. Today, these states are in an economic Catch-22: Their lack of current S&E jobs makes it difficult for them to compete for prospective ones, because employers are attracted to states and communities with high concentrations of skilled workers.4 High-tech businesses will not relocate to areas without an educated workforce, and educated workers are being siphoned out of these areas by job opportunities elsewhere.

Rural Areas Lag in Proportion of High-Tech Workers

Rural areas in particular have difficulty competing for scarce S&E jobs. In 2000, scientists and engineers were most concentrated in large metropolitan areas—especially in counties located in the Austin, Boston, Denver, New York, San Francisco, Seattle, and Washington, D.C. metro areas (see Table 2). Most of the counties at the top of the rankings have high proportions of scientists or engineers working in government facilities, such as the Los Alamos National Laboratory in New Mexico or the National Institutes of Health in Maryland. Many of these counties are also convenient to nearby universities that offer master’s and/or Ph.D. degrees in S&E fields.5

Scientists and engineers are conspicuously absent in most rural areas— especially parts of Kentucky, Tennessee, and West Virginia; southern “black belt” counties ranging from North Carolina to Louisiana; and a wide swath of counties stretching from the northern border in North Dakota to the southern border in Texas (see map). In 2000, scientists and engineers made up 6 percent of the labor force in metropolitan counties; 3 percent in less populous “micropolitan” areas (areas with populations between 10,000 and 49,999 and that have strong commuting networks with neighboring counties); and only 2 percent in rural areas.6

Table 2

Top 10 Counties, Ranked by Proportion of S&E Workers, 2000

Employed population ages 16+
Employed in science and engineering occupations
% in S&E occupations
Los Alamos County, N.M.
King George County, VA
Santa Clara County, CA
Falls Church City, VA
Howard County, MD
Loudoun County, VA
Fairfax County, VA
Boulder County, CO
Madison County, AL
Montgomery County, MD
Colin County, TX
Arlington County, VA

Source: Population Reference Bureau analysis of the 2000 Census.

Many of these rural counties have struggled for decades with declining employment in resource-based industries—especially agriculture and mining.7 Today, the majority of rural workers are employed in low-wage manufacturing and service jobs.8 To increase S&E employment in remote areas, states need to make significant investments in the education and technical skills of the rural workforce. Many rural areas are cut off from higher education and employment opportunities, but states can reduce the social, economic, and geographic isolation of rural communities by enhancing information technologies and especially high-speed Internet capabilities.

Racial and Ethnic Minorities Face Barriers to S&E Employment

These geographic differences are also accompanied by a racial and ethnic disparity: African Americans and Hispanics are underrepresented in S&E occupations compared with non-Hispanic whites and Asians. In 2004, whites were twice as likely as African Americans or Hispanics to be employed in S&E occupations (see Table 3).

Table 3

Science and Engineering Labor Force, By Race/Ethnicity, 2004

Total labor force
S&E labor force
American Indian/Alaskan Native*
Asian/Pacific Islander*
Some other race*
Two or more races*

Source: Population Reference Bureau analysis of the 2004 American Community Survey (ACS).

Asian Americans were the only minority group with above-average representation in S&E occupations (14 percent). Many Asians have migrated to the United States in order to pursue degrees and careers in science, and today the majority of Asian Americans in S&E occupations are foreign-born.9 In 2005, Asian countries accounted for four of the top-five sending countries for international students studying in the United States.10

The low proportions of blacks and Hispanics in S&E occupations cannot be attributed to state- or county-level deficits in S&E employment opportunities. In fact, counties with high proportions of minorities employ more S&E workers than counties that are mostly white: Minorities are more likely to live in large metro areas where S&E jobs are most concentrated.

However, there is a mismatch in the education levels and technical skills of minority groups and the demands of the knowledge-based economy. In 2004, 30 percent of non-Hispanic whites ages 25 and older had a bachelor’s degree or higher, compared with 17 percent of blacks and 13 percent of Latinos. Among Asians, 48 percent had at least a bachelor’s degree.11

The lack of professionals in many minority communities also means that there are few role models for youth interested in pursuing careers in science. Within metro areas, black and Latino youth are much more likely to be living in distressed neighborhoods with high proportions of people in poverty, high school dropouts, and working-age males who are unemployed to the labor force.12 More research is needed to determine the potential impact of neighborhood characteristics on youth education and career trajectories.

Future Trends in the Concentration of S&E Workers

The current geographic distribution of S&E workers is unlikely to change anytime soon. Many parts of the country have experienced growth in S&E employment during the last several decades—Austin, Boston, Raleigh-Durham, Silicon Valley, and (more recently) Boulder, Minneapolis, Salt Lake City, and Seattle. But most of these cities already have large research universities that have helped fuel economic development in surrounding areas.13

Between 1990 and 2000, the number of people employed in S&E occupations increased from 5.3 million to 7.0 million, but the distribution of these workers across states changed only slightly.14 Recent shifts in the geographic distribution of S&E workers reflect general shifts in the U.S. population to states in the South and West rather than changes specific to S&E.

Spending on research and development (R&D) in the United States is concentrated in a few states, making it difficult for states at the bottom to compete for high-tech jobs.15 The proportion of S&E workers across states is highly correlated with federal spending on R&D, while correlations with private industry and university spending are not as strong.16

Since state patterns in science and engineering employment are heavily influenced by federal funding, it is difficult to predict future trends. Over the next 10 years, some states could win and others could lose depending on unpredictable political priorities related to U.S. energy needs, threats to homeland security, or new health risks. But if the past predicts the future, it is likely that there will be few shifts in the geographic distribution of scientists and engineers in the near term.

Mark Mather is deputy director of domestic programs at the Population Reference Bureau.


  1. National Science Board, An Emerging and Critical Problem of the Science and Engineering Labor Force (Arlington, VA: National Science Board, 2004), accessed online at, on July 6, 2006; and RAND Corporation, Is the Federal Government Facing a Shortage of Scientific and Technical Personnel? (Santa Monica, CA: RAND Corporation, 2004), accessed online at, on July 6, 2006.
  2. Office of Technology Policy, The Dynamics of Technology-Based Economic Development: State Science and Technology Indicators, Fourth Edition, accessed online at, on July 19, 2006.
  3. Population Reference Bureau analysis of the 2004 American Community Survey.
  4. Leslie A. Whitener, “Policy Options for a Changing Rural America,” Amber Waves: The Economics of Food, Farming, Natural Resources, and Rural America (April 2005), accessed online at, on July 14, 2006.
  5. “Scientists and Engineers: Civilian Scientists and Engineers as a Percentage of the Work Force,” 2002 State New Economy Index: Benchmarking Economic Transformation in the States (Washington, DC: Progressive Policy Institute, 2002): 34-35, accessed online at
  6. PRB analysis of the 2000 Census.
  7. David A. McGranahan and Calvin L. Beale, “Understanding Rural Population Loss,” Rural America 17, no. 4 (2002), accessed online at, on July 19, 2006.
  8. LaStar Matthews and and William H. Woodwell, Jr, “A Portrait of Rural America—Challenges and Opportunities,” Research Brief on America’s Cities 3 (2005), accessed online at, on July 19, 2006.
  9. National Science Board, Science and Engineering Indicators 2006 (2006), accessed online at, on July 19, 2006.
  10. Institute of International Education, Open Doors 2005: Report on International Educational Exchange—Leading 25 Places of Origin of International Students, 2004/5, accessed online at, on July 7, 2006.
  11. U.S. Census Bureau, 2004 American Community Survey Population Profiles, accessed online at, on July 19, 2006.
  12. William P. O’Hare and Mark Mather, The Growing Number of Kids in Severely Distressed Neighborhoods—Evidence From the 2000 Census (Baltimore: Annie E. Casey Foundation, 2003).
  13. Office of Technology Policy, The Dynamics of Technology-based Economic Development: State Science and Technology Indicators, 4th ed. (2004), accessed online at, on July 7, 2006.
  14. PRB analysis of the 1990 and 2000 censuses.
  15. Richard J. Bennof, “R&D Spending Is Highly Concentrated in a Small Number of States,” National Science Foundation Data Brief 01-320 (2001), accessed online at, on July 7, 2006.
  16. PRB analysis of the 2000 Census and data from the National Science Foundation. Estimates on people working in science and engineering include both civilian and noncivilian workers.