LITHUANIAN SCIENCE IN TRANSITION: STATISTICAL ANALYSIS VYGINTAS GONTIS Institute of Theoretical Physics and Astronomy, Vilnius. decade starting from 1990 essential changes in Baltic T HE LASTand it is reasonable to call it wasa markedofbytransition. Social economical countries as period changes are so deep and quick that even quantitative estimation of these changes be- comes a real challenge for scholars and practitioners. Everyone has acknowledged that statistical information collected describes the period, when law, social and economical environment, legal status of institution, methodology of statistics and many other things were changing rapidly. The transformation of academic community and scientific activity in the region and Lithuania especially was very directly influenced by the global geopo- litical changes as well. Having in mind the complexity of the processes involved and possible high level of uncertainty we are taking attempt to present statistical analysis of transformations in Lithuanian R&D sector. This is stimulated by the real need to increase correspondence of science policy-making decisions with actual dynamics of processes in scientific community. Countries of Central and Eastern Europe (CEE), experiencing very rapid social and economical transformation faces difficulties to secure appropriate public funding for science and higher education. Simple comparison of available quantitative indicators shows trends to considerable collapse of science potential in CEE. At the same time countries of the region in their EU accession process and rapid restructuring of sci- entific systems lose their own ties in the region. The need to share the experience making reforms of science and higher education, adjusting the scientific infrastructure to the na- tional and regional needs provides additional arguments for us in this study. We present various Lithuanian Science Indicators Key Figures available in the period from 1990 to 2000. The Lithuanian Department of Statistics is the main information source for us . Information from other sources and definitions involved were adopted when needed [2, 3]. Lithuania, as well as any country in transition suffered dramatic changes during the period of conversion from the planning to the market economy. As it is obvious from the dynamics of prices in their levelling with world prices , Gross National Product (GNP) grew up to 10 times in hard currency (USD), but considerably declined measured in stable, for example year 2000 prices, see Fig. 1 (Data: Statistics Lithuania and The Bank of Lithuania [2, 3]). Significant structural changes in the country’s economy and drastic shrinking of the markets in the East were prevailing in the transition of 1991- 1997. Despite the decline in GNP (2000 year prices) the quality of Lithuanian products and services, exports to the West and general international competitiveness of National economy increased considerably in this period. How should we estimate changes in R&D sector? Let us start our analysis of science transformation in this period from the number of R&D Staff, which provides reasonable indication of the scientific potential in the country. In Fig. 2 we present change of number of scientists (persons engaged in research REVUE BALTIQUE 25 Fig. 1. GNP per capita in 1991-2000, USD C u rre n t P ric e s 2 0 0 0 Ye a r P ric e s 4500 4000 3500 3000 2500 2000 1500 1000 500 0 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Fig. 1. GNP per capita in 1991-2000, USD. Fig. 2. R&D staff R e s e a rc h e rs S c ie n tis ts 25000 20000 15000 10000 5000 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Fig. 2. R&D staff. activity having scientific degree or academic titles) and researchers (scientists + techni- cians and equivalent staff involved in research) in Lithuania from 1990 to 2000. Lithua- nian S&T suffered an abrupt decline in 1990-1994, when major social economical changes took place. This decrease has occurred mainly due to the elimination of the so- called branch institutes of the former times, as well as the withdrawal of the assisting personnel from the science field. The decline of number of scientists and researchers in this period is in close correlation with decline in number of R&D institutions (Fig. 3) as 26 LITHUANIAN SCIENCE IN TRANSITION Fig. 3. Number of R&D institutions 2000 147 104 1998 115 120 1996 121 86 1994 78 67 1992 75 102 1990 102 Fig. 3. Number of R&D institutions. Fig. 4. R&D staff – full- time equivalent R e s e a rc h e rs S c ie n tis ts 16000 14000 12000 10000 8000 6000 4000 2000 0 1990 1992 1994 1996 1998 2000 Fig. 4. R&D staff – full-time equivalent. well as some recovery of researchers later, when new type institutions were established. The number of researchers is much more flexible than number of scientists and more sensitive to the changes in the labour market. Stable decline of scientists is alarming for S&T policy makers. In the Fig. 4 we provide R&D staff numbers calculated to the full- time equivalent (FTE = head count – 2/3 research and education staff in higher education sector), as only 1/3 of financing and work time are devoted to the research in higher edu- cation sector. The Governmental and business sectors experience major collapse in com- parison with higher education sector – Fig. 5. Scarcity of the budget financing and REVUE BALTIQUE 27 Fig. 5. Researchers by sectors Hig h E d u c . G o v e rn . B u s in e s s 8000 7000 6000 5000 4000 3000 2000 1000 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Fig.full-time) by ten Fig. 6. Researchers (full-time) persectors.thousand labour ( 5. Researchers force in selected countries EU 50 S we e d e n 70 F in la n d 65 D e n m a rk 48 E s to n ia 35 P o la n d 29 Hu n g a ry 26 L ith u a n ia 30 L a tv ia 19 0 10 20 30 40 50 60 70 80 Fig. 6. Researchers (full-time) per ten thousand labour force in selected countries. insufficient attention to the science education are among main reasons of decline in gov- ernmental sector. Very small business sector, only 3.3 % of total number of researchers in 2000 clearly reflects the situation of very low science commercialization level. On the other hand, there are obvious problems with data collection from business sector and partially accounted experimental development. To restricted bases of R&D statistics makes Lithuanian data from business sector hardly comparable with international indica- tors . Acceptance of international methodology has to improve the situation in the nearest future. Considerable increase of R&D institutions in the last period Fig 3, proba- bly reflects already improving Lithuanian statistics. We can point out that the research 28 LITHUANIAN SCIENCE IN TRANSITION Fig. 7. Number of scientists by age and sex in 1992 1600 1400 1200 1000 800 600 400 200 0 <3 0 3 0 -3 9 4 0 -4 9 5 0 -5 9 >6 0 ye a rs ye a rs ye a rs ye a rs ye a rs M a le s 50 697 1007 1409 858 F e m a le s 93 340 598 615 226 Fig. 7. Number of scientists by age and sex in 1992. Fig. 8. Number of scientists by Age and sex in 2000 1200 1000 800 600 400 200 0 <3 0 3 0 -3 9 4 0 -4 9 5 0 -5 9 >6 0 ye a rs ye a rs ye a rs ye a rs ye a rs M a le s 61 451 906 1055 1038 F e m a le s 32 341 606 576 267 Fig. 8. Number of scientists by age and sex in 2000. and higher education system in the jurisdiction of the Ministry of Education and Science consists of the 75 following bodies: 15 state and 4 private university type higher educa- tion schools with 100 thousand students (2000-2001); 4 state and 3 non state colleges; 29 state research institutes and more then 20 smaller state research establishments. Depart- ment of Statistics accounted other 72 R&D institutions in 2000. The total number of re- searchers in Lithuania is comparable with other countries of Central and Eastern Europe see Fig. 6, where the full-time equivalent numbers of researchers in selected countries is provided, Statistics Yearbooks of Estonia, Latvia, Lithuania, OECD in figures . Fig. 11. R&D financing in 2000 year prices, Mill. Lt, REVUE BALTIQUE 29 scientific degrees conferred Fig. 9. Number of1USD=4Lt. Budge T o ta l Dr. Hab il. t Dr. 53 0 0 00 450 250 400 350 200 300 50 21 5 0 200 11 0 0 50 100 50 50 00 1996 2 7 1 9 9 1 1 9 91 9 91 9 9 3 1 9 9 8 1994 19 1 000 1 9 9 5 9 9 9 9 6 21 9 9 7 1 92 0 0 11 9 9 9 98 2000 Fig. personnel by field of science in Fig. 10. R&D 9. Number of scientific degrees confered. 2000, % Fig. 11. R&D financing in 2000 year prices, Mili. Lt, 1 USD = 4 Lt. Medical Sci. 10 % Hum anities Natural Sci. 22 % 8% Agricultural Sci. 6% Physical Sci. Social Sci. 15 % 19 % Technological Sci. 20 % Fig. 10. R&D personnel by field of science in 2000, %. The dynamics of age-distribution among scientists is much more threatening; see Fig. 7 and Fig 8, where we present the distribution of scientists by age and sex in 1992 and 2000 correspondingly. There isn't any hope for positive changes in the nearest future as only 1.7 % of scientists are under 30 years old and only 16 % are under 40 years old. The aging problem is the real challenge for Lithuanian science policy makers. The aging isn't so threatening for females, though total number of females is rather low – 34 % of all scientists. Unfortunately, the information on internal and external brain drain is hard- ly available in our country. Accurate forecast of age-distribution would be problematic. In any case, it is obvious, that Lithuania must increase the number of young scientists. In order to satisfy the minimal regeneration needs it is necessary that 300-400 young scien- 30 Fig. 12. Science financing and research quality in Central LITHUANIAN SCIENCE IN TRANSITION Science”,v.283,1999 and Eastern Europe, “Science”,v.283,1999 L ith u a n ia L a tv ia B u lg a ria E s to n ia R u m a n ia P o la n d C ita tio n H u n g a ry G E R D /G NP S lo v a k ia C ze c h R e p . S lo v e n ia E U a v e ra g e US 0 1 2 3 4 5 6 Fig. 12. Science financing and research quality in Central and Eastern Europe, “Science”, V. 283, 1999. tists become researchers for R&D. The number of scientific degrees conferred is slightly growing see Fig. 9, and has to be supported financially in the nearest future. Some changes in the distribution of R&D personnel by field of science have also taken place during the last 10 years. Fig. 10 provides information on the situation as of 2000. In the Fig. 11 we present information about Lithuanian Gross National Expenditure on R&D (GERD) in 1991-2000. In this period Lithuania experienced very rapid econom- ical changes, which make financial estimates rather complicated. Major changes are: hyperinflation in 1991-1992, the national currency Litas (Lt) introduced in 1993, Litas pegged to US dollar at ration 4 Lt = 1 USD in 1995. The most damaging influence to the Lithuanian science structure is related with the crash of Soviet Union in 1989-1991 and cut off of all previous contracts with the Eastern partners. Unfortunately, it is very com- plicated to recover the financial impact of these changes. We do expect that the total fi- nancing of Lithuanian science decreased nearly twice from 1989 to 1991. Fig. 13. Number of international publications (reviewed by Science Citation Index) in Baltic REVUE BALTIQUE 31 Countries 600 500 400 300 200 100 0 1993 1994 1995 1996 1997 1998 1999 Estonia Latv ia Lithuania Fig. 13. Number of international publications (reviewed by Science Citation Index) in Baltic Countries. To eliminate the inflation in Fig. 11 we presented GERD in 2000 year prices. Monthly consumer price index was used for recalculation, source: Statistics Lithuania. Lithuania has one of the lowest ratio GERD/GNP in comparison with European countries see Fig. 12 . Low investments of firms in R&D and insufficient public sci- ence funding are the main reason of this. According to 2000 statistics, public R&D ex- penditure was 0.37 % of GNP and other only 0.26 %. Firms should invest much more to the product development and production technology development and related technology transfer. Even Lithuanian Government share in GERD is rather low in comparison with other countries. EU experts recommend increasing governmental R&D financing first, expecting synergy with business sector contribution . Despite many negative trends in the development of Lithuanian science in transition the research output and quality has experienced signs of improvement. Let us chose the number of publications registered in Science Citation Index (SCI) system as criteria for comparative measure of science efficiency. In Fig. 13 we present the number of SCI pub- lications in Baltic Countries (Source: Science Citation Index  ) from 1993 to 1999. Estonia and Lithuania experienced significant growth of publications from 250 to 500. Numbers like 400 publications may seem very low compared to the output of Western Countries. However, everyone has to have in mind that for very long period there were big restrictions in Baltic Countries to publish in international journals and there still exist the language barrier. On the other hand, everyone has to acknowledge that there are over 100 local scientific journals in Lithuanian, where researchers publish over 2500 articles per year. If we tried to calculate the actual cost of one publication we would get very low estimate in comparison with other countries. The average citation index of Lithuanian international publications, see Fig. 12 should be estimated as very high in comparison with other countries, especially having in mind very low science financing. Finally we 32 LITHUANIAN SCIENCE IN TRANSITION can conclude that since 1990 Lithuanian scientific community has been reduced consid- erably. One might have expected the output of the scientific activity to shrink as a result of transformation process, low science funding and rapid attrition of the scientific com- munity. However, the scientific community is till productive with growing desire of more intensive international co-operation. 1. Research Activity, Statistics Lithuania, Statistical Bulletin, Vilnius, 1991-2001 2. State Higher Education and Research Institutions in Lithuania, Lithuanian Infor- mation Institute, Vilnius, 1994. 3. Monthly Bulletin, The Bank of Lithuania, Vilnius, 1993-2001. 4. V. Gontis, The Baltic Paradox, “Lithuania in the World”, 1996, Nr. 3, p. 26-27. 5. Main Definitions and Conventions for the Measurement of Research and Experi- mental Development (RD), Summary of the Frascati Manual, 1993. OECD, Paris, 1994. 6. I. Dagytė, J. Kristapsons, H. Martinson, Baltic RD System in Transition, Södertöns högskola, 2000. 7. Science Financing and Research Quality in Central and Eastern Europe, “Sci- ence”, V. 283, 1999. 8. STE Hannu Hernesniemi, Financing Perspective for Lithuanian Science and Technology policy, Etlatiedo Ltd. 9.8.2000.