Scientists and researchers are the backbone of knowledge-based economies. Without them we would not have the worldwide web or many other advances we take for granted in our lives, from healthcare to food safety, even engineering feats such as power stations or the Channel Tunnel. Now ageing, but also the fact that science seems to have become an increasingly unpopular subject with students, have raised fears that we could run of scientists.
Yet, as business and governments invest in research and development (R&D), demand for researchers in areas such as information technology (IT) and biotechnology continues to expand. The number of researchers in OECD countries rose from 2.4 million in 1990 to 3.4 million in 2000, a 42% increase, and demand is still expanding – the EU estimates it will need 700,000 new researchers to meet its commitment to increase investment in R&D to 3% of GDP by 2010. The US National Science Foundation projects that some 2.2 million new jobs in science and engineering will be created over 2000-2010, especially in computer-related occupations. In Japan the University Council predicted in 1998 that demand for masters students would exceed supply by 2010.
But where will all these new scientists and researchers come from? A recent UK government report bemoaned a 16% drop in enrolment in chemistry and 7% drop in enrolment in physics and engineering between 1995 and 2000. Meanwhile, countries such as Australia and Italy worry about replacing the large numbers of “baby boomer” faculty staff when they retire; some 70% of full professors and 35% of all science staff in Italian universities are over 50. Indeed, meeting the demand for scientific talent is so high on government agendas that it will be a central issue for discussion by science and research ministers from OECD countries as well from China, Israel, Russia and South Africa when they meet in Paris on 29-30 January 2004.
Such concerns are not exactly new. Back in 1945, Vannevar Bush, the Director of the US Office of Scientific Research and Development, warned that “with mounting demands for scientists both for teaching and for research, [America] will enter the post-war period with a serious deficit in our trained scientific personnel”. The launching of Sputnik in 1957 and the space race between the Soviet Union and the United States amplified his concerns and led to the expansion in the supply of scientists and engineers, not only in these two countries, but also in western Europe and Japan.
But forecasting demand for scientists or other specialised workers is inherently risky, not least because most projections are based on past trends. It is almost impossible to determine with any accuracy how industries and technologies will emerge or develop in the future, and thus what their scientific needs will be. The US National Science Foundation warned in 1989 that by 2006 America’s production of scientific skills at university level would fall short by hundreds of thousands. The fact that such a dire prediction has turned out to be wrong is a salutary message for caution by policymakers, not least because there is a risk of overreacting and producing a glut of scientists, and so fuelling a “brain drain” as graduates, unable to find work at home, go abroad.
In fact, most labour economists debunk claims of future scientist shortages, pointing out that the market will resolve them, as an increase in demand will result in a rise in wages for fewer scientists and this, in turn, will increase incentives for more students to pursue scientific subjects and so raise supply again. However, this logic may apply less to academic fields, where public universities or governments are the main employers and where public R&D is the main determinant of demand. Here, salaries may not adjust easily to a drop in supply or to competition from other sectors such as IT.
Shortages should also lead to a drop in the unemployment rate for scientists. In fact, unemployment rates for recent university graduates have historically been low across OECD countries. They ranged between 2% and 5% in OECD countries in 2000. In the US, the unemployment rate of PhDs in science and engineering during the recent economic boom was even lower, at 1.2% for graduates who had been out of school for three years, according to the US National Science Foundation.
But these overall figures do not show where particular types of scientist are working. In fact, the NSF found that some 4.2% of science and engineering PhDs were working outside their field of training, chiefly for financial reasons, a change in professional interests or lack of opportunities in their field.
In other words, while few scientists are out of work, a significant proportion of them are not finding jobs in occupations that are closely related to their studies. This would weaken the claim of a widespread shortage of science and engineering graduates, but may signal another problem: “mismatches” between what the market (industry or academia) needs and is willing to pay in terms of research, and the skill sets, interests and salary aspirations that graduates have.
One of the main trends cited in reports claiming future shortages is a drop in the supply of new graduates. However, OECD data show that young people have never been as highly educated as today. More than a quarter (26%) of the OECD population aged 25-64 had completed tertiary level education in 2001 and about one fifth of all university graduates obtain degrees in science and engineering.
The supply of PhDs is of particular interest because most research positions require PhD level training. National data for both France and the US show a downward trend in science and engineering PhD graduates since the late 1990s. But historically, data on enrollment and graduates show that supply follows changes in demand, albeit with a lag, which has been the case with enrolments in the biological sciences in the US which have reversed their decline since the emergence of the biotechnology industry in the 1980s. There has been a slight rise recently in US students pursuing science and engineering programmes at graduate level (4% in 2001). If these increases are sustained, it should lead to a reversal in the overall decline in science PhD graduates and reduce the risk of general shortages or supply shortfalls.
Many people predicting a shortage of scientists cite the growing share of foreigners among science and engineering graduates. It is true that foreign students make up an important share of the S&E population in several OECD countries, particularly in the US where NSF data show that more than a third of all PhD degrees in science and engineering, and almost half (47%) of all doctorates in mathematics and computer science are awarded to foreign students. The number of US citizen doctorate graduates in science and engineering increased or remained stable from the late 1980s through to the late 1990s, so shortfalls were occurring probably because demand was driving the increase, not because national supply was shrinking.
Foreign supply may well fill gaps, but it is not certain this will always be the case. Recent data show a drop in foreign enrolment and graduates in the US, as students from India and China, which produces a fifth of the world’s supply of PhD graduates in science and engineering, increasingly find educational opportunities in other OECD countries, such as Australia and the UK. They may even be staying at home. In addition, data on holders of temporary visas for high tech workers and of which universities were among the top recruiters, show a drop in petitions since the downturn in the US economy in 2001 and greater scrutiny in the aftermath of 11 September 2001.
Perhaps the central issue behind concerns about shortages of scientists in OECD countries is the realisation that the growth of OECD economies depends on investments in knowledge, including an ample supply of scientists and engineers, and a perception that young people are either less interested in science than before and/or less academically equipped to pursue research careers. Such perceptions are reinforced by news such as a 2001 NSF public opinion poll in which two thirds of respondents agreed that the “quality of science and mathematics education is inadequate”. International benchmarking studies such as the OECD PISA survey or the TIMSS (Third International Math and Science Study), which measure literacy and competence, reinforce concerns in countries where youth ranks low in maths and science.
The quality of science teaching may also be an issue. The UK report that lamented the drop in physics and chemistry graduates pointed to the poor pay of teachers as a cause of reduced teaching quality and hence student performance. Consequently, policymakers have called for increasing teachers’ pay as well as improving the quality of maths and science teachers.
Not rocket science
What can governments do? Over the past few years, OECD countries have modernised their science curricula to make them more responsive to student needs and demands from industry and academia. Some have created interdisciplinary programmes, linking biologists with computer scientists, for example, in order to meet demand for skills in bioinformatics. Others have shortened degree programmes in a bid to reduce drop-out rates.
Universities in the EU have also moved to harmonise degrees up to doctorate level to improve recognition of diplomas and foster more mobility between member states. Efforts have been made to increase the participation of national minorities. Several OECD countries have also acted to improve access and participation of women among science and engineering graduates. And it is not just a question of encouraging women scientists back into the job market; firms could also do more to keep them in research.
Meanwhile, industry involvement in higher education, including PhDs, is increasing in many countries as a way to better prepare graduates for work in industry. Indeed, this is a key challenge for EU countries where currently only 50% of science and engineering graduates work in industry, compared to 80% in the US.
Barriers to staff mobility can also aggravate mismatches or shortages. Making employment more flexible in higher education and public research institutions is one way of helping to renew staff and keep younger people in research. Several OECD countries have reformed employment law in the public sector to allow researchers to work in industry for a limited period. The EU has made it easier for researchers to move across borders, for instance, encouraging unemployed Spanish biologists to seek work in the Netherlands where there are vacancies, but barriers like language remain, as well as issues related to moving a family or finding work for spouses.
Of course, better data on demand for researchers and on the career paths of science graduates would also bring some objectivity to the debate. But there is no guarantee that this will put an end to predictions of a looming shortage of scientists.
Bush, V. (1945), Science the Endless Frontier. Report to President Roosevelt by Vannevar Bush, Director of the Office of Scientific Research and Development, July. US Government Printing Office.
Butz, W. et. al (2003), “Is there a shortage of scientists and engineers? How would we know?”, Issues Paper, Rand Corporation.
Mervis, J. (2003), “Scientific Workforce: Down for the Count” in Science Magazine, 16 May.
OECD (2003), Science, Technology and Industry Scoreboard, Paris.
OECD (2003), Governance of Public Research– Towards Better Practices, Paris.
©OECD Observer No 240/241, December 2003