ࡱ> #`bjbj5G5G8W-W-q%"""8#DL#\L?#Z$"|$|$|$W%*&\&0>>>>>>>$@hjB>E7W%W%77>|$|$? < < <7\|$|$> <7> < < <|$# pO%"9 <w>$?0L? <C;C <C <l 'p}, <_0{3 ' ' '>>;j ' ' 'L?7777D,"," R&D ACTIVITY AND PATENTS IN CEE COUNTRIES Jurica `imurina, Ph.D (corresponding author) e-mail:  HYPERLINK "mailto:jsimurina@efzg.hr" jsimurina@efzg.hr Tomislav Gelo, M.Sc e-mail:  HYPERLINK "mailto:tgelo@efzg.hr" tgelo@efzg.hr `ime Smoli e-mail:  HYPERLINK "mailto:ssmolic@efzg.hr" ssmolic@efzg.hr University of Zagreb, Faculty of Economics and Business Trg J. F. Kennedy 6, 10000 Zagreb, Croatia Tel: +385/1/238333 R&D ACTIVITY AND PATENTS IN CEE COUNTRIES ABSTRACT The R&D activity is acknowledged to be an important component for growth and development of countries and subsequent growth differentials among countries. On one side we have creation of knowledge through R&D process and on the other side we have the actual output of the R&D process in terms of patents. Even though Central and East European countries are not at the technology frontier they too produce knowledge which is not only important in terms of royalties, but also as a prerequisite for successful technology transfer, assimilation and diffusion from technology frontier countries. In this paper we analyze the impact of R&D process and output (patents) on growth and development on selected Central and East European countries. The selection of countries is based on data availability, which is in many cases problematic, so we use pooled data and panel data in order to perform the analysis. Key words: R&D, patents, technology diffusion, growth INTRODUCTION The importance of technology, technology transfer, technology diffusion and technology creation has occupied a significant portion of studies on growth differentials of countries. Historically, technology has played a significant role in development of what we today call modern or developed economies. Great differences among countries in terms of per capita income and overall development started to emerge in 18th century with full realization in 20th century. By the end of the 20th century gap in terms of per capita income between the richest and the poorest country was roughly 1:400. Such development coincides with the First and Second Industrial Revolutions. If technology was so important in making the development gap so large, it seems this issue needs a closer look in terms of possibilities and capabilities of catching-up with developed countries. Here we analyze selected countries of Central and east Europe in order to look into possibilities of closing the technology gap and thus subsequent development gap. HYSTORICAL BACKGROUND The technology, technology change and technological progress have played and important role in development of human kind. In this research we will not discuss overall history of technological progress, but rather, we will take a stand from an important historical event that happened in 18th century Britain. This event was the First Industrial Revolution, so we build our story from this point in time onward. The First Industrial Revolution was a point when growth of per capita income and simultaneous increase of populations was possible. Before the Industrial revolution economies would be stuck in the Malthusian trap, and any progress in per capita income would be eaten by growing population. In terms of Economics as a science, Adam Smith is the starting point of the time. In 1776 Adam Smith published his Wealth of Nations. It shifted the classical tradition from France to England and led to the further development of Economics. As suggested by many, the Wealth of Nations was first and foremost an attack against the principles and practices of mercantilism. The economy of Smiths time was still primarily agricultural and commercial, rather than industrial. Although the spinning jenny and the water frame were already invented, and James Watt patented the steam engine in 1769, but the diffusion of these technologies will take still take some time. Nef (1943) distinguished three technical inventions that were most important for an extraordinary growth in British industrial output. First, the puddeling process which made possible widespread use of coal in the manufacture of bar iron; second, the adoption of the steam engine, and finally, the use of power-driven machinery for spinning. Furthermore, the steam engine and spinning machine sponsored by Arkwright started to influence economy on a greater scale after the patents by Watt and Arkwright were disputed in court in 1785. After this ruling, the new method of making iron becomes extensively diffused. It can be seen here how the legal system had important influence on diffusion of innovations and inventions in eighteenth century. Weather the ruling was just or not is here beside the point, and after all, the judge made the final decision in a court of law. This proves the importance of the legal system and the overall rule of law. There are several possible factors that may have influenced low TFP growth (and R&D) in the early nineteenth century Britain. Smallness of markets, weakness of science and formal education, inadequacies of the patent system, the continued high rewards to rent seeking, and the difficulties of securing compliant behaviour on the part of workers may have contributed to the slowdown to a certain level. As far as the government is concerned, its policy did not play an active role in correcting failures nor in any other way did it intervene to correct market failures, compared to the successful government role in the Asian success stories like Korea, and Taiwan. The policy had quite the opposite role. This was viewed in the crowding out effect of public spending during the Napoleonic Wars. These financial pressures pushed for more protectionism during the eighteenth century and to rise in taxes during the industrial revolution (Crafts, 1996). By the beginning of the twentieth century the US took over as the industrial leader over Britain. The technological lead of the US was very real and the gap became even more substantial during and after the World War II. As Nelson and Wright (1992) argue, on the microeconomic level, the US firms were significantly ahead in application and development of the leading edge technologies. US made up the largest portion of the world trade, and overseas branches were often dominant in their host countries. Today, that is no longer the case. US technological lead has been eroded in many industries, and in some, the US is even lagging behind. There are two distinctive slices of the US dominance in the post war world. One is the dominance in the mass production, derived from favourable historical access to natural resources and single largest domestic market. The other part of the story is the lead in the high technology industries induced by massive private and public investment in R&D and scientific and technical education that the US made after World War II. Even though these investments stem from earlier institutional foundations, the leadership in this area is much the product of the post war era. However, it is sometimes argued that the strength that American companies possess is less based on technology per se as in the organisational efficiencies stemming from mass production and mass distribution. One of the most spectacular success stories in the US in the inter-war years was automobile industry. It was a blend of mass production methods, cheap materials and fuels. The technological leadership itself was more lasting in the industries where there was connection of mass production and organised science-based research, e.g. electrical industries and chemical engineering. Abramovitz (1993) distinguishes different ways in which technology has influenced economy in nineteenth and twentieth century. The first, but not the crucial, difference is the pace of technological progress; however, the character of technological progress seems to be more crucial in this division of centuries. This may be the reason why the conventional capital accumulation has played such an important role in growth accounting for the nineteenth century and a much smaller role in the twentieth century. In the nineteenth century technological progress was heavily biased in a physical capital using direction, only to shift toward intangible (human knowledge) capital using direction in the twentieth century. This bias produced substantial contribution of education and of other intangible capital accumulation. The technological change of twentieth century tended to positively influence the relative marginal productivity of capital in terms of education and training of the labour force at all levels, from deliberately acquired knowledge through R&D investment, and in other forms of intangible capital (e.g. support for corporate and managerial structures and cultures, development of product markets subject to the infrastructure of the economies of scale and scope). The bias shift of the twentieth century encompasses the change in employment patterns. The shift occurred from agriculture (low education levels) to manufacturing, mining and construction (intermediate education levels) to services (relatively high education levels). There are several factors that contributed to this shift. First, there was an increase in income level per capita and associated Engle effect on the structure of the final demand. Second, growths of the service industries, due to requirements of exploitation of scale intensive technological progress (e.g. trade, communications, and finance, legal, accounting and engineering professions). Finally, there was a technology bias toward agriculture and industry, where the productivity of labour was raised more than in services. The process of industrial development, with the increase of complexity of technologies and research activities, from the end of nineteenth to the beginning of the twenty-first century has been marked by the formal organisations (R&D laboratories of big firms, government and university laboratories, etc.). This situation is different from the situation until the end of nineteenth century, where individuals, inventor-entrepreneurs, dominated developments in technological breakthroughs. The turning point of introducing major institutional innovation of the in-house industrial R&D was in Germany in 1870. However, product and process innovations by firms took place some hundred years before the said time, but it was the German dyestuffs industry which first realised that there is some profit in research of new products and development of new chemicals processes on a more regular, systematic and professional basis. Even though it is a fact that in past centuries or millennia before 1870 there were many inventions, professional R&D lab seemed like a giant leap forward. This was especially reinforced during the Second World War. Science was already very important in the First World War, but the Manhattan Project and the outcome at Hiroshima was what impressed on people throughout the world. The power of science was evident, especially the Big Science. Many other inventions and innovations from both sides (e.g. radar, computers, rockets, explosives) resulted from large R&D projects which included government and industrial and academic engineers and scientists (Freeman, Soete, 1997: 299-300). In terms of organizational capacities to perform R&D, we have seen that by the outbreak of the Second World War there was extensive research network with organized research laboratories along with the related institutions in government, university and industry. The researchers in said institutions were employed on a full time basis. As any other industry, R&D industry can be a subject to an economic analysis, with recognition of some unique characteristics. The output of a research process is a flow of new knowledge, both of general character (basic research) or specific application (applied research). The output may be incorporated as flow of models, sketches, designs, manuals and prototypes for new products, or of pilot plants and experimental rigs for new processes (experimental development) (Freeman, Soete, 1997: 6). TECHNOLOGY, R&D AND PATENTS Some authors consider technology to be a very pulp term that is hardly definable. Radoaevi (1999) is of the view that technology as a concept has no clear boundaries, and where generation and diffusion process is deeply embedded in the institutional fabric of economy and society. The forms of technology may vary according to the level of disembodiment from patents and licences to those embodied into machines or persons, i.e. tacit knowledge. As stipulated by Jones (1971), it is important to distinguish between science and technology. Technology is know-how while science is know-why. On one side we have science producing knowledge, while on the other side we have technology, which helps to produce wealth. We usually have a choice of importing technology and conducting local R&D. These two ways of acquiring technology should not be in conflict but rather they should complement each other. According to Sachs (2000), it is evident that most of the new technology innovations come from developed countries, which accounts for some 15 per cent of total population. A second part, containing some half of worlds population is able to adopt the new technology generated by the developed countries in consumption and production, while the remaining part, containing around a third of the worlds population, is actually technologically disconnected, neither innovating at home not adopting foreign technologies. Process of production and experience in use yields increases in productivity of a new technology. Because of this fact, the new technology is allowed to substitute the old technology, which as a consequence increases profits to who ever hold the patent for the technology. With time there will be new inventions as experience accumulates and the new technology will substitute now mature technology. The lifecycle of technology and the discounted profitability of new technologies will be determined by the rate of innovation and the rate at which production experience accumulates (Young, 1993: 447). As stipulated by Romer (2001), there are common characteristics to all types of knowledge. First, they are non-rival. This is to say that the use of some knowledge, no matter whether it is the Pythagorean Theorem or a soft-drink recipe in one application makes the use by someone else no more difficult. However, private economic goods are in contrast rival. This is to say that use of a particular pair of shoes by an individual precludes its simultaneous use by someone else. The fundamental properties of knowledge stipulate that competitive market forces cannot completely govern the production and allocation of knowledge. Once a particular knowledge has been discovered, its marginal cost for an additional user is zero. This would suggest the rental price of knowledge in a competitive market to be zero. However, in this case creation of knowledge would not be motivated by private economic gains. The conclusion here is that either knowledge is sold at above its marginal cost or market forces do not motivate its development. Therefore, competitive model cannot be fully applied here. However, there is another possibility. Although knowledge is non-rival in essence, it can be excludable. This is the case when we are able to prevent others from using it. The excludability in the case of knowledge will depend both on the nature of the knowledge itself and on law, regulations and institutions governing property rights. The good examples are the patent laws, where the inventor is given the right over the use of their designs and discoveries. So, if some knowledge is excludable, the producers of new knowledge can take out a license on the rights to use their knowledge at a positive price, and earn positive retunes on their R&D efforts. Stephan (1996) argues that science should be in a focus of economist for three reasons. First, science is a source of growth. The lags between basic research and subsequent economic consequences may be considerable, but economic impact is indisputable. Second, labour market for scientists, and incorporated human capital, are a fertile ground for study. Third, the reward structure has evolved in science that goes towards solving appropriability problem associated with the production of public good. It is further argued that science makes technological innovation possible, but on the other hand science itself is influenced by technology. Economic theory based on competitive markets underpins the poor incentives that market provides for production of public good. The providers of public good cannot appropriate benefits derived from use. However, the appropriation relates to rewards that are market based. Sociologists and economists have demonstrated that scientific work is not necessarily based on market, but rather on non-market reward system. Here, a particular argument is the fact that there are rewards to being first in recognition by the scientific community. The priority in awarding by recognition of scientific community depends on importance that the same community attaches to the discovery. One of the most important is the practice of attaching the name of the scientist to the discovery (e.g. Haleys comet, Phillips curve, Says law). Another way is in the form of prizes. Among others, the Nobel Prize is the best know, with the largest award. However, many other prizes exist with smaller awards. Further, many countries have societies where scientists are elected (e.g. National Academies of Science, Engineering, and Medicine in the US, the Royal Society in England, the Acadmie des Sciences in France, Croatian Academy of Science and Arts). In the extra institutional variety of rewards, a successful scientist may be rewarded for speaking or through consulting fees. A scientist with successful patents may generate future income flows, which has been a standard practice, especially in the life sciences (e.g. as scientific advisors and director of new companies). The bottom line reward for a scientist may be in the satisfaction derived from solving the puzzle. On the account of more direct measures of technologies there are several approaches to this problem. As stipulated by Keller (2004), technology is an intangible that is difficult to measure directly; however, there are three widely used approaches. These are inputs (R&D), outputs (patents), and the effect of technology (higher productivity). Regarding the data on R&D, not too many countries report substantial amounts of R&D. Because this measure capture primarily resources spent towards innovation, with omitted resources spent on imitation and adoption of technology, the data for R&D cannot typically be analysed for middle and poor countries. However, the data on R&D becomes more available as countries incomes rise. There is further information because surveys include R&D conducted by affiliates of multinational companies located abroad. One important aspect of R&D is that the returns on publicly funded R&D are lower than the returns to privately funded R&D. In comparison to R&D, patents have the advantage that the patent data has been collected for longer time. However, there are some issues in using patent data. Small number of patents may account for most of the value of all patents, thus the mere count of patents may not capture technology output well. The other problem lies with the choice of patenting or not patenting, which eventually lies with respective firms. Thus, some innovations never get patented. However, there is a further issue, which involves codifiability of technology. In the case when technology is partly non-codifiable, patent data will omit this part of information, thus technology in this part will not be captured well. Research and development Basic research (as opposed to applied research) is sometimes seen as research without any specific goal or apparent purpose. It is due to the character of basic research, which appropriation of invested funds may be far into the future. This is the reason why companies do not fund and engage in many basic research projects on a larger scale. They need appropriation of their funds as soon as possible, while basic research sometimes cannot warrant any results even within a lifetime. Thus majority of funds and efforts concerning the basic research comes from government and universities. This is not to say that basic research is less important, but the mission of governments and universities allows more funds to be dispensed into this area as generators of social gains rather than just short or medium term financial gains of companies. In terms of costs of acquisition and adapting to local conditions of new knowledge, Chavas, Aliber and Cox (1997) emphasise that the process of adoption can be slow. There is a lag between R&D investment and the subsequent payoff, which can vary with each technology or industry. These lags, as a rule, are longer if R&D is basic research, and shorter if the research is more applied. Furthermore, funding of research is of concern, namely if it is private or public. If knowledge being generated is of public character it is appropriate to fund it from public fund, however, if property rights can be privately assigned, e.g. through patents, it private institutions have incentives to invest into such R&D. Due to a fact that basic research has long period of payoff, incentives for private involvement are weak. As a result, public funds are assigned for basic research in most case, and applied research with shorter payoff period is supported by private funds. However, the internal rates of returns from both public and private R&D are fairly high in the agricultural sector in the US. In the structure of funds, pure or basic research has the character of relative publicness, and it is mainly financed through public funds, universities and other non-profit institutions. On the other hand, industry meets a greater portion of applied research and development, while some funds are still devoted to basic research. Along with the formalised R&D, a significant amount of improvements is originated through learning by doing and learning by using (e.g. Arrow 1962, Rosenberg 1982, Hollander 1965). In terms of costs of acquisition and adapting to local conditions of new knowledge, Chavas, Aliber and Cox (1997) emphasise that the process of adoption can be slow. There is a lag between R&D investment and the subsequent payoff, which can vary with each technology or industry. These lags, as a rule, are longer if R&D is basic research, and shorter if the research is more applied. Furthermore, funding of research is of concern, namely if it is private or public. If knowledge being generated is of public character it is appropriate to fund it from public fund, however, if property rights can be privately assigned, e.g. through patents, it private institutions have incentives to invest into such R&D. Due to a fact that basic research has long period of payoff, incentives for private involvement are weak. As a result, public funds are assigned for basic research in most case, and applied research with shorter payoff period is supported by private funds. However, the internal rates of returns from both public and private R&D are fairly high in the agricultural sector in the US. In conventional thinking of economists, R&D is seen as generating one product: new information. Cohen and Levinthal (1989) suggest additional aspect, where R&D is not only generation of new information, but also enhancement of the firms ability to assimilate and exploit existing information. It is further argued that R&D obviously generates innovations, however, at the same time it also develops the firms ability to identify, assimilate, and exploit knowledge from the environment. This is also called a firms learning or absorptive capacity. On these lines, a firms ability to imitate fits into the scope of absorptive capacity, but it also includes the firms ability to exploit outside knowledge of a more intermediate sort. This includes basic research findings that provide the basis for subsequent applied research and development. Basant and Fikkert (1996) argue that the available studies from developed countries do not take into consideration factors that are at least as important as R&D for the developing countries. Since these countries mostly operate within the technology frontier, the expenditures of firms in developing countries on disembodied technology through licensing is of importance. The results from the research of Indian manufacturing firms indicate when all industries are taken together there are high private returns to expenditures on both technology purchases and R&D. Furthermore, the rate of return to technology purchases outweighs returns on R&D. Regarding the R&D activity and growth, Allen (2001) argues that the increase in R&D and acceleration in growth of the capital labour ratio coincide with increased wage gaps by schooling within industries, and that increases in R&D are associated with wider gaps by experience. In terms of technological change that varies across different industries, it should not be surprised to see correlation between technological change and wage growth for this group that would persist if the pace of change were accelerating. Caselli and Coleman II (2001) stress the importance of well-known fact that an increase in technical efficiency plays a critical role in log term growth. This fact has led research to concentrate on R&D process and data. While this data is commonly used for analysing developed countries, it is argued that it may be unsuitable for developing countries. This is due to an assumption that developing countries operate inside the technological frontier, so the efficiency gains may be acquired through technology already available in developed and technologically advanced countries. Thus Caselli and Coleman II use computers diffusion in order to encompass both developed and developing countries. The reason lies in the fact that computers have been introduced recently, and computers are clear case of embodied technology. In order to diffuse computers they have to be physically installed, which makes it easier to measure technology adoption. On the other hand, it is very hard to measure diffusion of disembodied technologies. In this respect there is strong evidence that computer adoption is associated with high levels of human capital. Furthermore, good property-rights protection, high rates of investment per worker, and a small share of agriculture in GDP enhance computer adoption. There is also evidence of a negative role of the size of the government and a positive of share of manufacturing in GDP. As suggested by Helpman and Coe (1995), in the light of openness to international trade in goods and services, FDI, and international exchange of information and dissemination of knowledge, a countrys productivity depends both on its own R&D and on the R&D by its trade partners. A countrys own R&D efforts produce traded and nontraded goods and services that induce more effective use of existing resources and thus raise productivity levels. On the other hand, own R&D makes acceptance benefits from foreign technical advances possible. The more a country is able to take advantage of these benefits the more productivity it becomes. Furthermore, the benefits of foreign R&D can be direct and indirect. Direct benefits are those which relate to learning about new technologies and materials, production processes, or organisational methods, while indirect benefits stem from imports of goods and services that have been developed by trade partners. In both cases the foreign R&D affect a countrys productivity. Patents Furthermore, in our analysis we will take account and analyse patents. Patents are considered to be an indicator for output of a research and development process. This indicator may be considered as very handy in valuing output of research activities. However, there are some shortcomings to this indicator. For example, a company or an individual sometimes do not want to disclose results of their research in order to preserve exclusive or monopoly rights to their invention or innovation. Therefore, some research result never see patent offices, thus are not included into the patent statistics. Other shortcoming of this indicator is that patents do not have same weights regarding their impact on an economy. There are merely summed up. However, trends in patenting are clear indicator of research activity, thus cannot be neglected. The final big issue regarding patents is patenting process itself. It may take some time before an invention or innovation is actually patented. This may depend on the number of examiners, which may change in time. This may cause delays of patenting process in some years, thus the data may show fluctuations that are not caused by the lack or bust of R&D activity but merely administrative constraints to the patenting process. However, patents are generally considered to be a good proxy for output of R&D activity with all the shortcomings that come with it. Historically, establishment of patenting system has lead to a boost in readiness of individuals and companies to reveal results of their research to the rest of the economy. The patenting process is a guarantee that a company or an individual will preserve exclusivity to invention or innovation for some time. Furthermore, this has led to greater influence to economies in general, as with the patenting innovations and inventions were available to general public much earlier that what would be the case without protection of patents. R&D AND PATENTS IN CEE While the problem of knowledge is a problem in some Asian and African countries, the same cannot be said for the transition countries. However, both groups of countries share the disproportion in creation of new knowledge. Furthermore, in Central and Eastern Europe, the accumulated stock of knowledge is significant, but it is not utilised. Thus, creation of knowledge by itself does not automatically mean a country will benefit from it. Much of the research and development (R&D) in ex-communist countries went on military research. While the military R&D may have spin-off effect in other areas, it seems that these countries did not benefit from it to a greater extent. Beside the data on GDP, employment and domestic investment already described in section 6.1.1, R&D, residential and non-residential patents, as well as data on scientist and technicians will be used for this analysis. Research and development expenditure (% of GDP): Expenditures for research and development are current and capital expenditures (both public and private) on creative, systematic activity that increases the stock of knowledge. Included are fundamental and applied research and experimental development work leading to new devices, products, or processes. Patent applications, non-residents and residents: patent applications are applications filed with a national patent office for exclusive rights for an invention--a product or process that provides a new way of doing something or offers a new technical solution to a problem. A patent provides protection for the invention to the owner of the patent for a limited period, generally 20 years. Researchers in R&D (per million people): Researchers in R&D are people trained to work in any field of science who are engaged in professional R&D activity. Most such jobs require completion of tertiary education. Technicians in R&D (per million people): Technicians in R&D are people engaged in professional R&D activity who have received vocational or technical training in any branch of knowledge or technology. Most such jobs require three years beyond the first stage of secondary education. Figure 1: Log R&D (% of GDP), 1990-2004  Figure 2: Log R&D (% GDP)/log GDP, 1990-2002  From the figure 1 we can observe tremendous changes in R&D expenditures as percentage of R&D (Bulgaria, Croatia, Czech Republic, Hungary and Slovak Republic), while other countries experienced almost negligible fluctuations. Among these countries Bulgaria has diminished R&D activity beyond recognition. Croatia, on the other hand, managed to return its R&D expenditures to levels before the drop. Czech Republic seemed to have reached a new lower steady level of expenditures. Other countries experienced either a permanent drop or fluctuate around the same trend. From the scatter plot in figure 2 we can see the relationship between R&D and GDP in the observed period. However, only few countries have data prior to 1995, and even that data has missing values. We can see that Poland, being the largest economy in the sample, does not exhibit higher levels of R&D expenditures, while Slovenia contributes a lot to R&D. The extreme value for Bulgaria can be considered as an outlier, since all other values for Bulgaria lie more to the left. While Czech Republic and Croatia exhibit modest levels of R&D expenditures given the GDP values, other countries (except Slovenia) show more modest figures given the size of the economy. Figure 3: Log R&D/log Scientists and technicians in R&D per million people  From figure 3, we can see that Slovenia is leading the show again. Although Bulgaria has few values close to Slovenia, these may be considered as an outlier given the overall performance of Bulgaria. We can observe Croatia, Czech Republic, Hungary, Romania and Slovenia to cluster above the linear fit, while other countries lie beneath the linear fit. It is obvious that smaller countries in general have lower R&D expenditures and fewer scientists. However, Slovenia may not be considered to be among bigger nations in the sample, but it does have highest R&D expenditures, along with the highest number of scientists per million and the highest GDP per capita in the sample. Figure  SEQ Figure \* ARABIC 43: Log R&D/log Scientists and technicians in R&D per million people, no Slovenia, 1990-2002  When we exclude Slovenia, again picture (see figure above) does not change very much. There are few Bulgarian outliers, but main conclusions stay the same. Figure  SEQ Figure \* ARABIC 44: Residential patent applications (logs), 1994-2002  From the data on residential patents we may observe output of R&D activity across selected countries (see figure above). While Bulgaria, Croatia, Czech Republic, Hungary, Poland, Slovak Republic and Slovenia exhibit relatively stabile or increasing trends in patent applications we cannot say the same for the remaining countries. Estonia exhibits clearly an increasing trend, while Latvia exhibit quite clear and dramatic drop. Lithuania and Romania exhibit volatility, which is not so dramatic. Figure  SEQ Figure \* ARABIC 45: Non-residential patent applications (logs), 1994-2002  From the figure above we can observe non-residential patent applications. Among all countries Croatia exhibits the most dramatic increase. Other countries do exhibit increase also, but it is far less dramatic than in Croatia. Even though this indicator has increasing trend for all countries, it will not be considered in the formal regression analysis due to its unit root nature, while other variables do not exhibit this property, thus cannot be combined in the same regression. Figure  SEQ Figure \* ARABIC 46: Residential patents and R&D, 1994-2002 (logs)  From the scatter plot above Slovenia to be on top, however, this has not yielded the top position in patent applications. It seems that Romania and Poland have the highest number of inventors; however, their population sizes are large given the sample too. Among all countries it seems that only Baltic countries are lagging considerably behind, given both expenditures for R&D and number of patents. To some extent this is not surprising given the population size. Figure  SEQ Figure \* ARABIC 47: Patents and scientists, 1994-2002 (logs)  We can see from the above figure that the linear fit between scientist and patents is negatively sloped. This is to say that residential patent applications do not follow the increase in the number of scientist but quite the opposite. Countries with fewer scientists per million people seem to have higher number of resident patent applications. This may be due to the fact that many inventors are not counted as scientists, especially in the area of mechanical and electrical engineering where lab size is not the dominant determinant like in chemistry. The other issue is that a large number of scientists work in public institutions that are not very much interested in patenting their achievements but rather getting published, which diminishes actual number of patents given the discoveries. Figure  SEQ Figure \* ARABIC 48: Patents and GDP per capita (logs), 1994-2002  From the figure above we can observe the relationship between GDP per capita and patents. The linear RT   4 6 r t   ' ( ) 8 9 ; s ԳŦԔŦԂŦwog\ThoKmH sH hmchmcmHsHhmcmHsHhxmHsHhmchxmHsH#jhmchxUmH sH #jhmchxUmH sH hmchx0JmHsH#jhmchxUmH sH jhmchxUmH sH hmchxmHsHhmchxmH sH hmc5mH sH hmchx5mH sH  TV8 : b : ; s op $da$gdoKdgdoK $da$gdoK $da$gdmcdgdmc $da$gdmc CDRT\xzPQkʿxxpg[PHPHh \mH sH h \h \mH sH hDhD5mH sH hD5mH sH hDmH sH hbe^hbe^H*mH sH hbe^mH sH hoKhoKH*mH sH h_dmH sH hoKmH sH hoKhoK5mHsHhmchmcmHsHhmchmcmH sH hoKhmc5mH sH hoKhoK5mH sH hoKmH sH hoK5mH sH hmchoK5mH sH QR*+&&. $da$gd+ $da$gd $da$gd[ $dha$gd \ $da$gdD $da$gdoKdgdoK*+$$..05588881;2;<<<??^A_A÷ïÜÔwod\owoQhmch[mH sH h0AmH sH hmch0AmH sH h~EmH sH hmch~EmH sH h[5mH sH h~E5mH sH h[mH sH hAmH sH hmch[OJQJh )OJQJhmch[6mH sH hmch[mH sH hmH sH h[mH sH hmch mH sH hmch+[6mH sH hmch+[mH sH ..00558881;2;<<??^A_A:H;H $da$gdM\ $da$gd[ $da$gdm $da$gd0A $da$gd~E $da$gdA $da$gd ) $da$gdoK d`gdoK_A:H;H8Q9QWWWW X([)[m_n_rasaee i ikkmmźynf[SHhmchwmH sH huymH sH hmchuymH sH hVbmH sH hmchVbmH sH h#27mH sH hmch#27mH sH h2 mH sH hmch2 mH sH h[mH sH hmch[mH sH hKcmH sH hmchKcmH sH hM\hM\5mH sH hM\5mH sH hmH sH hmchmH sH hM\mH sH hmchM\mH sH ;H8Q9QWWW([)[m_n_rasaee i ikkm $da$gdw $da$gduy $da$gdVb $da$gd#27 $da$gd2 $da$gdKc $da$gdm $da$gd[ $da$gdmm!s"swww$w|y}ymznz||~~~~vw $da$gd5$d`a$gd[ $da$gdoK $da$gd# $da$gd0A $da$gd[m!s"swww#w$w~~~~~aeuvwR=e<0ƺzrgYgMgMgMgh@qh@q5mH sH h@qh@q5\mH sH h@qh@qmH sH hXP mH sH h%6mH sH hmch5mH sH h5mH sH h[5mH sH h%65mH sH hAmH sH hmch[mH sH h2 h[5mH sH h2 5mH sH hmchwmH sH h#mH sH hmch#mH sH h0AmH sH hmch0AmH sH wQR<=/01234567_abt $dha$gd6{gd6{gd@q $dha$gd@q$dh`a$gd@q067=?_`belm_fg}~ۺɫ푂skZEZE)jh@qh@qOJQJUaJmH sH  h@qh@qOJQJaJmH sH hYdmH sH jr=h@qhYdUmH sH hYd6OJQJaJmH sH #hYdh@q6OJQJaJmH sH h6{mH sH j h@qh6{UmH sH jIh@qh6{UmH sH #h6{h6{6OJQJaJmH sH #h6{h@q6OJQJaJmH sH h@qh@qmH sH h@qmH sH tɉʉ./de{|Z[\]^_܏ޏߏ{|} $dha$gdYdgd@q$dh`a$gd@q $dha$gd@q $dha$gd6{~܏ݏސߐܒ67$+,BCEFvwV]^tuwxƚ͚ΚŶŜō~ojh@qh@qUmH sH jvh@qh@qUmH sH jϑh@qh@qUmH sH jsth@qh@qUmH sH h@qh@qmH sH jYh@qh@qUmH sH  h@qh@qOJQJaJmH sH )jh@qh@qOJQJUaJmH sH 'h@qh@qOJQJaJmHnHsH u+}~ސҒӒԒՒ֒גْؒڒےܒ6$dh`a$gd@qgd@q $dha$gd@q689 !"#$vxy{|MNOPQRSTgd@q$dh`a$gd@q $dha$gd@qTUVҙәŚƚ $da$gdoK$dh`a$gd@qgd@q $dha$gd@qΚ ﷬th0JmHnHuhYd hYd0JjhYd0JUh?jh?Uh@qh5mH sH Uh@qh@qmH sH jh@qh@qUmH sH 'h@qh@qOJQJaJmHnHsH u)jh@qh@qOJQJUaJmH sH  h@qh@qOJQJaJmH sH  fit is a bit ambiguous because it marginally slopes downward. In essence, there is no clear relationship. We can conclude here that number of patents have ambiguous relationship to increases in GDP per capita. We can observe that higher GDP per capita does not make people more inventive. It is more likely that populations of respective countries play a role. We may also conclude that some countries have larger innovative potential than others, which has nothing to do with per capita incomes.     PAGE  PAGE 1   $da$gdoKh]hgd5M &`#$gd4^6&P1h:p&Y%/ =!"#$% DyK jsimurina@efzg.hryK 2mailto:jsimurina@efzg.hrDyK tgelo@efzg.hryK *mailto:tgelo@efzg.hrDyK ssmolic@efzg.hryK .mailto:ssmolic@efzg.hrDd x$$0  # A2ؖxM?@\`!ؖxM?@\4N9x]tTՙ$ \@ Qb0P@Ƅ?J@$!$ D{=Glz-[ٺ,ڊj1liYX+jon{37x{gn}{wcX߇?<ܛ ˺gެ rxY=` ]ˏ/]t2!>vF?9oix/ckpzq @ɿ-`#?kz_]}{F8mؓ#g8?2%Y΃%zl}-\ɟ߀I"<ձ U>،]h,|B#\ '7*^ ڑO^+Ѝl/~rwbF`=wFhg˜>amG\&ONO;'7<\œ0e Yvsp\aǜ7%C:c ;2-cJ|Mx}vxp5s$<F6hKEGz/i|/I _o_U69=WgjHf+7.F^5p,'pۗN|Y= =v|ϻ_/|1MSxX5yBL]^L30$41=v !> q]//0dndү٣_sWӅ]>Fw#ϻ}ލaƟϱW`N &`TmR! !tb:j\bi!YiV|UPLb:1ڝ֮ Ӯ]-ܵ+v\aaavEM DMMvE L͈A]A|kWK`ծ@GPkW_jW_GԮb,v߹ϵ:Vل1cs&2\ 9191\׹1BLG #tLs1ALstӑ $t$b:2\u]z5M׹|cs |u.rLG)`:Q(]"t 1br׹ NUg\_(`u./Hu:]_G?]"yc|B7}] \c'rb,rb]czbz bz]c+t$0s0Ĕu|=ObkW䘎St!qT:f׮&u'>+10n#us&VV fd8\#?U׹m6Lo#6 ֹΦ\ST:s~'׹@ չΦ\_s\d>oOƣ\(pȉ]"vLo'vv]"ǴPBBL B]"t1NH:Suޟ1S׹| |u]"ǴH"BL"]"t 1N >ZQӽRG ~}Z1~ ]"y >+ֵ9&)pb!'&!'&90Ldt]c:E)NAL |^+\I?UOEڞ[P=t>NS4B>NC>Nu.rL+`:t]"t31[r 1 R +W0""ӯճgYk,;O׹|L7D7S׹WY\䘖(`ZBi bZ\Vt6!z9s0CtN+؟B׮| |w"Ե+rLR.BLBLҵ+rL*`:ӹܠ%׮ NvUgkW_qxLV>dۇG?]"y|ZYv)q!!90]Nrt]cB4 Z AړG//0=LG\Dq\IV[\ܫ{ 9q/r^]">L#>>]"t1]u.rL+0 Ĵ1H}h:/|\Մ|\|\\䘮I`0ukQL ktӵ %t-b6hɵ+(H+G//0=LG?]"y >UϫԵ9)pb!'!'90]Ozt]cZi!Ui]cA n@L7R*Z\sݍ'^a5&ݍN m=u݈ݨ\nRt!MEiՄV#AsKsARr׹@//0=LG?]"y5 >ՠϫuDj8QKȉZDscU4Ji1:90LftscE-nALtg s#r1?NIwqȀIYy{ԿI2O{y{ߧ;4|=޴?tӾF+?/_~;/? k>|~B`.rcR??E' }߁}]r)I!7[x&J3?Z:<=n{li鶧Ŧ۞VI}r?Wn{[am鶷iފLlοn1'鶿]nao;ԚnhG~֙nפ{M}Y~3SagIk?/Ynru_Xng-7cXn~?+t_/+?|ݢ7[,w_=!=^ُJ~@Ǐ=l>>}#ҧ\KXcxYܐGn.A1lȅK M&[22auKb2jm#9O ]ȬȬ\KhPa2k׬]lw`gl RSaӝ;~yrA3 i@4ܑFL 1mDLZ,rL+`vAy_ؗـ)->_=rMwnTr&[lv,xdFLwnG˖Y@&l>Ɂ[=9?|yyd3ל2kȄ:ء3v3v5r烟:w@ܯ 9s?r~cbb1ݩNBLw";/_O%=>_zܹQ!]@Qc`#3js 22algbb5עC9ձ|0\-sBK烝|.B }Ʈ N=u뤑3*pABGb|kdZZ@yw|xm lqO[:$C>Ȱ[: ku?!?xCP7yrfgrf/rf1ݧ>BL!tG~Lb1Gw㛡Ֆ3 .B\/lu/Pf06yj<:xtiϊס&Im:ڱ 6~b{~Wo3+paB#>>7rLUQBLEL#Зscΰ,->g[GFKu hoytxϣá:go,Зu( :6~PAAc>>7rL+`zÈa?߼ד3Ɯarz|ΰע>G#2uuCM/c)<:ab[zRlq<_^+uI6kyr,# >Nϱt{>ψ9u9ᶞ}l%ϸo Z/'ga[X=c"?OLs1j}q~29MO>bؾs֛ʏaV{ʸ^oĵ^+EDsEȞP߃9 _fbk̙#)`%\?W`z\fV窊Bإ2<C>{؄ۛvщ3\۝̟ƌNboE3`0Z͓ ?IԘCv7օ?=5y$j!zJg${݋S ?ESSLSp?͏BpIJ?ۻl>u>se|D ĨWsO,v4/S{=P>,#ꐬ//A(4!}iڱ9ưr_7bQkPyGͅ=d9mگZ _;V%e=LK籆}m_^nsnCxn9 1\bczrs+Mb-go͔{n@{n^E=;^^U$>'x8+m w\ck}e͕>%y~KT3ruFdT?FXFQ曌!7BFo2lc!XtWGϾ"%_Cqi}:N)NqrW<._2Jd2 d8㤎#-FO;;c b<3ohqԮɱTaݑN"yfrjr1\.wFw:w;̱l;Mqa;C m;l;/ ƾ(bsF>#Fiv.苬-ϝTyiNZOێsT N`l`l+gK-3zAnlY1Y1Ya;`Τ'?wEA1d칂W+1ϵ#_MVSX|sj8 /jx.6bz]~CEE:/ԥk2QJ~[pkUn޶zÚ>E3u܎ZLΜo[룵|NA[ 59=YJ[G9+am;~|Z]0c}PsoW7lp;u7[= U*8@3S`jir1Й ڵm^+uc3 }5uyߞp7ž߹د.6Tl.&m^2pefC+ ԻbMf**@t_4|z#t,װ3XV^YQ׃tef!arb:"c i]}@>Hn|`=Ût&U6aQ~LuLTxr]Wi K^|m `-j#ȝ(څaŹ ÈGߏ'\ _qZf&ǸHHbm^ƱhUHIpl<6n@2QDilsiRibn;y@f&1O39O37Y iV44XA^<̱?>u1U*'UYY*_TF|UU9*7f}~6p7,e(JY.kPsX@wUu6ºߠ|~.(Ͽ-Vck k޿^cADx*[?X{b@Q{ S\XL+[2~{Vek3ԞѽÅ|6Acz:2.C7fU~NebʬxX U-saMu#!1[떲cߠ{ 4X{;,֍6ޡY̓ccA%qlpݐ߅#?%(oXcm=1ot~$FQƺw%2FO{j?ͅKkO"\+d{-:NҞJ}ȸVek{= + N]~]u u7.ta=kk*2VcG F֞kȲ:kZ"Ve}#Oq9>+i ZGUZa;Z86gD9k$@ڈfUX[AȞ qfxbXF/z{H*[?XS;܅n4PH>kaU~RU>pcks1⽯нװ}bm8ua:0(46maU~RԾݫ]Xog-5'ːltVe+Ս˭|Zc퍌W㽵!9Yt#]_kLEӕX׭=_c]mi|"׺}>BP \X⳦Cz !h |f4ks`-= u=LA/]b_ x[D/93=ڔo]XNk=C+YX2 nu4ː 2>M)u|bz ;ޅ >kRZ/;ªl`Yzgf2`gG]/sb=X NzcaU~^5rЅƣ+Ag4*[?X@ ^nMgYPc;u!9Ve+ٯ?"eZ_.JHunX&̷C'me=ֿ]Xi}tƺY:iaU~Xnk=X{Y{[[}bM(jWG]X+~JX_,5BVva݈gj% 2m 6DgU~F9x݈.4.5UUJjvևhFcBh=& u"YwgRXwrm@tVe4־\X Xi}&mxul@? 9Ϊl`mlq?pa5EVMq6-V^!`V:/Ec G^-V3[Ye/~9[#WVK-wǹ!8Y Wc9~n p=t뺰g5,ڳ 3VeWVt-)t5!]S|b_yR}[m­B֗.du)^ y x}55Pd\[Ve^^'=΃18kKilc!vX7O7 v^aVOgY$ZDNK_%C÷ z܈m>fa@iQc'jBkku76>IL}{vҞtu?DWªl`d^X`wi? ^@8XǢ(^: Xm{c f @,(g݇,Z:Ef >kyrU@7joо-M=7ž7:Yb.a:'V[CwnXiq!6׭~>r-h?!"+`[(9}Xi J K-־}cнޅu(tΝښ,:+@[ꢱ5ZߡvVX_wb͹DwѭjYSTX"kGU+P[+:.!Cui_GC-~T!k=Cj''-Is4VgmCXדĺgCh`([?X}Sȵt+բ3KymϺ0'VZuzZ_Ov`AօMܗ:/?}([?X[c۝ =s|t 0=RmmY|AݧYvWvlАggkg3rm~Oǒ6 >:A~k+U ت>Ҙ cl#zL}@W1aOOCi:lcCW~5Ӑچ5ň k^>~M"֦63pWݦVJ)61kKy=ktzatcI}>|VQ@t-HX}wʼ8?zޭiqWa}'yJad'+NzVI Omfeȳg}謀U8|op?'Ib7Rq^QA=|"K Säzf0f$[/q.r6#Rw%fTKbQیjl3=78FyR=<qi 0ZV8 è# Km0wG៾Rh6K?yӵJ>iyJ|sπH /9L ܠ5_ eC!kOhn}YCɡT>Z=j( PX˻{PDrHk um$d]8pXn;VAQwCp-KUn֘ nâcNw'">ҍ#)6D/5Sq, Em#b Z9j'՝SNٕƍxsA0O $^>&' WYKQ$d#%>Pax Q(1(-<3$qdĨG~Ѭ'7{~miP3DD;bOE6WAc4Q*l!%v8MlSljUm'0joBkV#RK=l`U{ F>j;`) OހY|^c7qbUF>l=XնG\-Z;AOe}/j -hS ]X|l lq`dmJm+k`OAէv]{p>m2ޡ̮5Z.J-۫^65Еm¦Y|jjkaWljFV#טJ٦]ޱ= JOljK`7kL'010HZ^#JX,sv@^n` |`S tX6k5̬YøfyQJ`MzoR|^cZ9v82v%- /Y|վB}lLFaͺޡf 2k Y^FMm jv~6Tڙb<.}rmdZn-og YmǦVSc5)P6Յ'_9[lba>XrPV#w٨6Am6scwFҪ6R|^٪B hgΐ.RťFr}gWmlRLP) j@>BCpMg Dw7gFߙ3xfDj8%'~ mlj=+wu칰T.ϫAljV#_Dqվ!){@v90v$cCLw1ݙ@L?d' #uF>jbKCKlɷqKNjkQm5\<8 ~ ]:wĪ [C&[}f#5jcVjg8j 5\8!N5sT/1T{ @{W}+U8jBljV#טuC]aօ\́<0qQڹc$1f\љ6X1fQfP3#yU%8&0&e{j81~;ۚcm>*93tҭ[ <-6سZ7(g8ya.ڜˬL]%XU8[0-WjF5/'hm_GZf#-u$/lf-sYK͚kEzΡEzlI[:̳%­D6!9[m#4/^ Uyf{mC4me2t`[n-Om rt{LHUA,bamm5j+QY׬BլX{tj7$j ԉBWqyή~oNO5-J+#gR!ᴗlwv~bK=)O])e״N?9ݹdwl6ǺVVֿm1iKtS]yL*'l>YGrxݴ{=rʞ^7|!G=DFM#\qiH(EBD~Wg(d "ܵs=NOw8RNĝӨ/X/FX[,c11F4n% ^*ʓȝ!.ȯ%l 5uczwBͿÑY TRDd x  0  # A2]L B;Z=`!]L B;Z4N9px] |U?s[ B)&KEHXՏZ UD*YP~XUJ"|f澻w-.3ޓgΙ3g\eʈV%֒:|HKpvU&zm\:oe#:+~epi1Ko*'s2 /91?b$ET{Q3Lnx>3($G} %ɛ?o/~M!ly7J^}4CFՆW{[Yڪ˾}9z!ژxeeS9HSS[!|.}1N+~/|u:x%Wf}3iK䛨J:yxuZgPJ}5I񳝊y|8K}qƆ oHߏ$~s(/}ʐF!)COr8|ᲉX#cpim6QA(Q6Qr4GN6BC3ӡJC:yxu::AS:v$/N7B3Jge;Wcшھʳ[ճ}4ѾF(q@ƾvX=aV }ӡC=#=t 6fDPא%;Uxy|e{H|<9Vkw`߁Hzj: h7sދ O'v)›Exvhz=:cJ7dZ6 ;Qj`Fهxv}ASdZ6 Y3gSqb p"\l\{t>õ?aZ \l\ 8ßkHÔ\ύj0د!܇x4%8%W- H4nҞ5ːς_Fzn, Fծh׭vuGn ];t&{𚐏: ~`Y-bҟٹjY;Wx gUZ8WsS\?q\ٕq9\l\?(Zgٸ~{uHospղap..Bχ#4ס-G:Uˆu" =K1/xpc_sղap\9)kÿR}_צp}حFוxez68y\W+ǷO2Fl-19@GeqLcxϖ~bȹYVy$9ۣrncIMEa"]Õc[l\{npi _5e<ŵ)Yu8 Azc-AˆY6\k}gzYT']hW-Zq32:~+ўː~hW-Wsxy)c./_iW C*^"^鈵Fx8ꎵWH[h@֝׳>DZ7=ek$Ecu) ])Fr u>sZ;W-{{|oӷdz}6pe?‡5o\[GHt-:^S4>\_uBH_@:Uˆy~2XhzC#W- muW{ڸmpkZ6 <{#4dlV] Uˆ/ 9^+82&eZ5>\Ǩ_~@ˆŕ뙫z=o_QzMu u~U}Bu=gx{fּBX xϱVe{6/ruf]y"& =! ]]8u!uG.]{)c>\ m\ '3Z6 SYÕSmUˆ^ۇc$Ü0al\yoOGz|<h 8j0R}߷bFrY߈ +J>q)qxq=϶|V M ܇>Ҳ鹎% 8nx-c鿖Qb&eLRkCb -))^:zE\+ \f9len tWzTgd<:W> -\l\aƛs@grY%δswlz1j\QChlϑeܐpJY젚}cn&l4JoXѲ鸖krهH<ɋ+s'[~?"jxaZ6 9XMzYĽe_XOqN㱞yMq^^\rCs}ʘ^昛i0dLMpr-k Z6 IhvQ͵BѰph0~k<{s׃<4dd'W-W>{'{-=rqL̽ǨeqK4ܶa;.v7S5zq/<۾oqܻ:&vZ6 My~f&kc1y>-SMGfXns[cOb1VˆTLAmGLxwTƄ|0&l .fH޻{|ȹweC)?;Q#_0~m ߩإ.E#v˒;vy"`PC2uc6@ 1ol#m\e Љ,bGD7yA\l\Wt5#&ժ7'\H+e->HpVٸ"ߙel:L1ieծNѮw:jKv3vN)-&u21GKM5׶'en}dz6%U#}Z)\΄,}p޻aOR[yǘxAiK..gzf߻P޻PzFtdu:Rtwlf\Q(Q|;2<]1>78b%kF7J8}P}c(jD?.w~Tw:*jxݖ qj3dmMwsո==l_B;HN˜CۚkzH<+\|N?|V0э㣝} i3u_gPyݱm+Wu"bb|q]j{?111&{fuZePeJe޹qmNfPcNdž>}'MD-oG#>d5f8#N:H.e~f2Ee{Tl8+j/3&ˏl::|9\&yj,lw,΅)=\>yXsYg u>[y_$Fq㔿IaFd|u2^d|~R'@g[W,EeLwW&֣''4FZ(FVDo 9+櫹C>̀LSq?4=OǿH۫s%fT}*meY^f[{`Sv˲m4\|t/F}—}~{F%~Ջ`;oȳzєa::0/YwQQ?\py0Q0Kta>{bJ$f_lJRs?*f섖g$ךG?Ot=g[d;:5Eejf|^y}4qbx(4$Ó% {?3/Q/DτvTCASk&}" }@i~.Ӄz[ϐbfńT}B"Gt7J04E1=5%לpN$r{$Z*mČZxY3@g-1fA$`;-DF[1ʪS>J(Oy|x59Z} #gimҢr72aQ2^#]kV2VFUFA8,$c)z[Ơ"x6loQY6q;%qmR:(_6ӧy"yC,Nhg7=BA炢URB]8b@[DNhwh 5j!SeQ8=j/ޢ6.7B\p-ZNnViD,AB~8נhyb1wЎ:Mm1mmQ`n󡁫W'ڏx1ZIYyZn֛a,m-WE -w%hh7ѯ:ݮQ9O6@ӳ. hB\PnWVf sAm& m Kq.('m!Oh_48ʹ.B\Pzf:Tڨ8:.A sA*i'M?.m sA&П`o/Ne#$炢݆1&kATA,dm9@t]Uq.(؂9+>h#%z|*Nx\hmBƹhh1;vk{Uۭ6߃ۍuU.Έ%炢`V]B:Jh +`Wb ##Z$ .ē sC +,-e֬t7>mMR'NѹУs{ -]MB4 0!ꁚQnVM_d U:Qq.(MYGH{ܑz+rX"Yhy̅qa5.L5"{#)'==eq{BRmZuںh#NHAj`wǽkHǽklhq+\h1jK4E lfKm:Mn.z}t_ ̭t >B*Mh!G#Ye!Tj]-&ZDxD˭ay`n]12ctus炢O˭܆(C6rWM USIip:](:S #Whh2T,@:O PeAJm BdD_ 8L[wvi[Ohn# ̭QX'0Z!4Ћ ]hgD㜽[gR MWhKj?O^&Zh-O-܎_Sј}FA2 r8;8սW(|3Q!68y$mX w5[R-l2n+L&LIHN?GQWTWO?{BU7Z铲,5AقDd x0  # A2+I;9?&3Z`!I;9?&3ʜ4N9x] |U?s[K V@v(PE(  ARBDp?UTjVJV4 b\X¾4@YT@;ͼ»ewy?sΜ9 d=Z$_mfػn /{5 kcǶ0x?޻IߌY3Y:xU䴋7&>9ɕDŽ9">h7l)Ie0%Mn%䏉+e(#3 t,0Q5yr~%.ū0]NREv1C /Z;y),QQgˁ)_V}y۞"8(E'CցR$܇)$eI#]QDt]QDi0t02}l)kE(Km2)"|ja0 +3~Z$^StD݂pt H=|h*,Sh;rMW;ۡ [rݪֽTh>dtvZ7P]TL~a St~%/7hrh |~NJY%,(ס"%R Y } f ++=ۅeX="A餴 Y~կhDŃNWN^wg櫨BOG::]N$PkNLJ>fWI񳑈i|xuqL}o ^MԢ瓉͏>Lnh~76c )eq2SNO9Q63vg&akhkM=bSDtQNNǡ z'>Mt!nHN7n81~KGَ':`_Oq`ck=}}KE6RE(kc"ux ՘.5P/g/cz 6jh?X뵘f\M &ezY"?s9\9 {ŵ b7׋YµK L_t+Ds~p' ӆ̞'$ޏq~ލi5-*,Eޅ?`A2?`zU5 AslLk\Z%׽lW*eA!v6vg+\.8ӌ`4W)#s9s h\We+W)׏툽Ӄ\g:2LKM\\K Ĵnu>Wr`+LB\\v]ҮM:\ v]c6F[|VsjSgmy'yMUʪ\ (h ի'jYSzUAs7ơ 7㆔pgܾs4Q \o&ۖ~p]x;Ю+횃D OL*ez1k8p}9} Gw0`ɤ\!vW,:pm2vm'1gjW)OiOAg qAާ6`zkkEߦMz g-q>){$b_=\gCվ{KURdw"vL;pZ)\oAE1q~p1ƴ.,pJ3q~pAϿrϒ/!:sL\\Gz-4¿У ח{LK4q-g#y(;ӝ&R'ĞX7^Tb]ū+-Lŵ; Qq4G`+Wz*_]zŵ nfmkz1/|GbZ+ٔj|fgSH$ S:ٔԮv}ʥ]u۵MN]? F OҎh|v z_3c*eJkwϣ\> W\+>5}}K|橀uF5]u:k>pMEhLJY?}N}+Egu X+/ L7;KRgD~^iϒހ#Zu2ם2zkCu-Cu:h7԰h ƙl}aa |m7bv})\/C_)֖~p;sOgjNAw>Y;R֏zQJc;xvk+(\;tB W*eZ]ss\fLs#bn;Y=sOrķEkԷhMܷ\9-֢.:h[~p j|b\k ;\+tnuk qsx sZWky:_3ϙJY?c|9qIZEc\v:IY?‚2va?xLY#i ŵ/r-ƀ=p!Jކ|]\izH[Ϻ[~~ XC\L?ڬ'gUҼ/O{-sbMDr5!>iIY?tYK8 k 0kͺ ep^Yk4_}4>qLvzD\c\]טJY?Hqvi΢{g\1JX\FYGr>حuhWAu9n' Mck|^tK:ݒ@n:ݒw.8mLcMxN{bm. Ge Ye,9CDiiC%^n܉62}Uq=ؿl|8O/x5k#9}zi^:y3nk~?[[8eUۭ췲wY"G1d';=;6lN&>|>l MnKMn6̻qt{u]tӟCgYS\;?L~>qRb"6Ň$[ǎ:⤆ZtXY>2 2Iu':gΫB@Z>zs̯eg#y[e:gԹM(|;['Y/uXy~~[(O(<"sqdguSdg SaSv)8WHX_zt}.@!c߱("rta!p;Jt$<yNGdyVZ'Yܯ.j*#g6xXIvKTʨ2M1S`qN GkI@O8gDGVN{8j<X~N ? D[zû;N?K]~x'!J&?L{c|l3mt:}2M9~ |G=PN~}|DXÅ_ď7=gkw#,2T~:ӱ`Lm$ن{X J>^Opzq) $&7i~W{Y_UbG< $k򎆥Qږ~zDsL_'8r@aQw˙2uSrCF{k΄9ݡdKTt{:ן]XS͵qdyvj 1DZoܑj%ΏhE} MNF|ad/ZG-.2DX X ~qQƝ(I8,$C.jc/Ѿ_#DsTc 'NxITqA pUrQ1r _!>b- P*(m0x ڍ0AQN7W8 `F{GWn&nCF9h6A= zh (us6$<5elY~ 4Uj> 3jl@Swn336ܦܦ{v7j`<Ռ6-Pz`@HjVVf),قfT dL-VϊvD @C_JrH# Z'(}:YAF98.H y7vŠ,7np|hAK弢O`8,F9hofF9hCa_Xr^mmao+= vZЮhDs+Z{qtK_/#ns pla6cK[Ѷ8 R弢-Flʛބ4F9hؓaE[s4yE[ {a|cAh:eꖚQQ+66gl9G< }- 4Ubx^js 63ړ < /Z9r^k: /#r4ʩ"e|TF95}f@͘1&*dhK`6z9$Z8ڇF9s8f`(8Fk3RcTʋ ˠhqthʍe.O?1Sk@HwCk1 Z*&({;WsqrbAۃZ{|6т֔-(-݃" Z+,Q+ZF AF @ 97,h r޵:jaZ2qVfVט^7Y hʡ!LZVFˡ ZKgkzMM[5mlVmhm ԈY=yc&=ychCίFxM#8*Wmƅbc\(]קW\-Dnop{ƙ-A6h Şі`&F*1b#y"]ָd[Md7ςr^ѾB |jӏ ~4VmB27X|}agns0r [f¿o-YeqHCmv[oneFv#pؒ9d95"5Ǒ9GL39l@WL6o?t1PkڊJjcsGn欿@ך5]8׶e$G|oO@5,hS@Dޓʆ۬Vb 97ҭ=.=nZ#GtHG+ef-NyZ$c&}̧gbk>X k5<Ȝln+be??f(AHܿIyC3;.$faI\Dd x0  # A2Bq@ rt`!Bq@ r£4N9zx] pTUN:% O?`D~Ÿ( A"QAYKڙUvf&[;3g]Ygfqfb게；nwo8Tv~{^n[c~Xؗ?:5ܳh^O15 t/[>bz~?}37ݹ1c[.G⯇T?Ǝagpx71ONc1\`]S!޾ұ[ؠqoo.|f{?yNwڛI3<:C# 6~򃽬׸S@ΕD l Y5S`p ~e'-Y N2>?|k ~%n|oKXsUk \ n7raO<wlomA\?Xɦ c !\g2+YI<#tJ\ ~ǽƅ?FA_\L),r#{B\x|7}b==3c FcKք+ cƅ9,\q`0y0.b1w5bzg14L4% 0-ϡXB~˟s0}_1}_`צ0Ѿ<׈) cJ){"'R5k|O,'hpb!' '$S$t !ӁW#g{, L1BLLi`:A頞T7dnonbWQ}}|L|lzCt!C!7DP Lb:17 Vo{C`_zC`_b_[S3npoZOX'[#dƟl?٦u!'3}$rLi`:a0G"t 17}$rLs40!41I>RNL)A|qq!G G>9b>9#50IHtdHy"H}FGGjt}$-`)>YLSFƔQSFPy#Dr"1hBLG#Mo1!t b:1XBL"c7t&RMo(A|q||gzC^^^ozCtu޵k? ΄}$)A# H`_HgB>0L,LЈ)c)L)p=HȉȉD @ D$ L'b: 1dHk`Oi>bL}$[٦>NdB>NF>N6}$rLh`:)G"F Lo$FƠcs SW0ڕ}Tb_ezCD1eFLJSbLjzC#hpȉ"t1N3!rLo&BLoBLo2!rL 50-$Ĵ1-L}7%{o(k~^+ОCw[-2}$rLofBLoFLo6}$rLoBLoALo Gr R +P+}ʞQ3Ubt20Lǘ2s(D1!'ŦDJ魈魦D Lgb:1aHt&!3әɶ`sQB8K8 81lBLg#Mo LK1-ALK<P }]=`@*YLSƔ9SPs581ssMoy#tb:ZCb_ 7}$rLh`dH#%i6B>ކ|1]BBL" M50ۃG<>ҙ̾veժgHd1raLc" N,"""G"N L$NNs9wi`z!w!w950]Lbtqc_czC  >.!"ǴTRBLKR"tK 1]. >F>70>YLYSVƔSVkGXXXizC䘮t!U7Dj LWb1]mzC䘮t !k5ɶ`J+4XA c cVӵELך940]G:t]P1{C'P\ ^+={(%H}FGJf_b#ş R^gxz?0 @bYg#PJBNT"'*M n tb1ݨFBL7"M* L1BLz 1k@/c~GiɅۤyT !͟ BoI_eɛ*_)?o1׍/ [wT:|}i~ϏK8y'KB~_ eS?$q~Fi^yq>&i6.8>ϰX~gK_X~Eh/;l[~={w.^೬9 # >9)ybf;~s;wߑ-79~np+ρ?:~qhwJ_ur_y-'cLW>WVkԀՄ5`5ր 95bZ֘s7nt!MbÈ׆'Rmx֦!+6WlxmmymlT׆S1lYSC^q٫gƙglTx RpA C>>r)5;X3G mņ6]~Æi؏Md)jņņA+چM bK {a?6 /L Igkęqf?ƙqzYLh`rɁ[bFaߜ*)c/ixptyN5\>=~~_(,V*|X.wz}YL\dzjFbpmx1,#  o𛭱dR\ %oЗ))p5p5EX&YZ 'X>.gag_E rMg>ldZ1mG\W͟]r5C">&o"3>,Ѿk\s#$>4^++#pt~'ˍno_ӿNtI#g#cl3gMHY%|<Fz> )6 6~xoճiѾu _V}UYV!ְZA?qt{G_<\[c_mw#/CXFrhǎeiPq?DDzscY)U3/e%! 3!>X>˸Rή*ئՋoSpUR@ |.aO1@律x>DfM߬> 6>5~ᖬ9W;>$mwO @μ"U򪶚Gt`DWs[0a,Xfcok Tjp !"Rs zx0XtJx_eq9o*a'=#?B##-Y3Q/1]2qKũ3ܡn}_#G9m}q 89VDYo at]І + Wf1]:O-:ز^͵#|?ٟ\rR&x{dAo^skSe>}58~GWIŢZ/zdT&}[GQ0BGqt y:jS/ں>Xo˵{ra#DEKb8a:<&frX9a-L\f@;(Kb8a:<&frX9,Ξ19䰘9Æډaψ(BGqt y:L39roX+{TG3909.(|H' q\80C|a&uc˵5L$=/rX#:N$Lqxt:a&֍9a|aswÄG:!tHB8$t2L39sX9frX'9uq"a: qȜ39naeV+М39,V\^byxX:T(t&LGБ0a#la&}9,ײ-s>9yuP(L#'a:BG2L39r10b0hr0a:r a˜39+Hcf4(VVLX[GY ffbtTSlvRMbtVѭx0eQ{ё-QG-O Ӟta?W /DmOzI߲ώYLj҂ʬid9:VZHw_Gbֱ5BFHx҅~!_HgJ2먐>*2yt479s҂~$#uo 7BY!}VHVxr>--tձB(+M`qfϬ |||_(Bz^hu*tz,r>[:ˬBi<`]]!:[)\_3U:9evˌp`mڍO ~-0 -<(ilhs:l҂*U*r󶵵REuXX6s.S7Rۜmgi $nmj>k;}M=>V=Y  sVytپ\k}ѾW4mn|UZ߂nVl{O[ ;NwX\ Y˾҆ wun/:c0l6l$BMF9cQDZqeffݕV]Ys+L?]*?XVZz}e=O.9+hkEo!ӤE"#;sw&nŶ++{Eki]alƜXjkڝ9% wYQW ZaFbb8Zd;Oe9|Dd x0  # A2#0rw¿/>^V X`!0rw¿/>^V X 4N9x] pUwz2A 5Ze%< Ÿ($AEj@Pv}{}rڭ|l֥eR>V\xt3L҃wa9}stRe,]ދ]$˹{Ҹ"g^36drfyٳ3RV>*Okc dqpQ0>XUϽ}O}%Ix d# ^!?5p3pG \a'Frbrb#c:RӑDLG>940E(t#c:ZфFLG'SrHG¦ >i𱈐E"G"tc11}$rLj`:ӱ j-o DHTkyH`ߡWb820Ø2GozCNt! 7DmFmm7DD L'b:1Lca7T1v >Nۑ9wdAx7Di\ +`0IIOB't,#AN RXGG:Cg#1e2aL1e#p'$ĝȉ;Mb L 1-FLMӻ40ӻӻL)N!t b:%HAw/x2dH T >N%#ME>N5}$rLi`:i4G"t 1N:FCun`uxr}̾uNlV}bJFL)!)%SJLo(p}58q7!'FNmzCt!37DL Lgb:1izCޣ=ރޓlAo1JP\ ܹW ={5{/!wEkHާ}އgHt!YA]H#8A#}>rf ߚQ'fT>G")5bl˜2clG \~ NOȉ>9s40Ct#cZi)!i#c:WӹEL&:FG =F < >#<<"t 17!rL40-#Ĵ 1- :FjoDzC7̾uNlV}7DSЈ)Ɣ0<`zC#,BN,@N,0!rL50-'Ĵ1-WQ+UWh_A_aH.t!! Ӆɶ`GJiq!!>9VbZV>9UVbZVu#y"H}cao }'|@:P>L,TkĔj˜R1sD !'j5Db Lb1]l1"tK1].1!rLk50%Ĵ1Mu 9|aPTK 1:Ĵ1]2BL!˂ǖwi&}Mg_qٗaQLYSƔS7׽5u=z SoH>郄>>hH䘮t!+DC>DCCɶ`#GJÄ||#cRӕDLW>940]E*tUP1RHMH}>f_Ͼٗb#1˜ƔG=F#4hp ȉ"t 1]6!rLQBLEL5!rLh`5d[zC~ %i1B>>||1]ZBL"kMou#tb.!jo({tš=Ms RٗR~ocAy51k@8X{ xR}e\ /+9b2^/[x2~O "*-"eV)b{ŞwŸY#[q7(qbW^U\}-ʸHw*a{xKWô*R?JMO/]_TxxQ~߹^{#r/~1?^~{Kϸ >U{Ϫg^j/YLi/8Sw**|=ᖗٖ}-/{[^fZW -`yC-UU{#,/UUU{\MWYրOjԀOրOb vC& L7b 1dݐcObbg556Wb:Oc, 9ڱpN֥Z(uj}wG1=}:O3̫c<ί3?C>P 9u:Υ3 uVVsE뭊K U{l֘s['f+S)ÚșF 4r9h 4{nvE-oA|Not+![ӭX> ; GCr}:s}:%1t:Wզ'vO|{3]Co+75['si¹4էZ}iykoCfC^Gu4N(Z_=Z 0Vm+ñ6l6uhf }6@mf-;c lXñ6| D8P6(ƪ .qlzI  l0qƞ;5N8d=_Z@< 9Ӳz{mjí!dž1lm\aφ? hi|64Æl0f1Y#&5ƤfIA7/}kzSrE3-iAδz]"tbԛ䘾Ke/_lZ3pt5l }6;6aM yg צ86lI@g7>^CnDMgymjÌcÑ6!/8μg^&3/cy95d!%gvkpf7!gv#gv=!tbԐ䘶j`Ji+bښr/6 `0"5fXcԐolxgâTdžbESC ƞ~lcӢmYrφ>^Fm0M φW}6ܚ665$Aiӈ3mq Le݋^ Lb1w`2,f>OQ a)s v>9œNxSwο Z.?aߝ- >Y>>C,[}e,LYG1>=Z-,v܆taI7n~rA&S}x[WW5\%sr~>]P:>6-gPDtrg26OcbqĹ}bRЯg4 | =oW-Dz5>؟^Op@o'C\8,L1rV~~NUc\/NL >g!8>ە39|6~rRGWcj-miѾJ?U_ 9ƺ4ϒi~BAkog8[#Z}۱/SXfŲpӡر, j=?ד)ѱTlU37MB~A,A,3Xy[UhA,egn`WelQɷ_N"a ~YTmi>>>geX>UI>>#]}_C13>Op X2 wWp@b9\^vpw ߗ\^."H+E",J=,$?xU+9ލ3LoCg+2nNO^):sYغ$ 99ڑ&~Jqۿ*Ȯom;RpB=CSge~~lC?ι3Yk+Jsa{\hni}Sz#3Xy؋x[o= X3ҍg1~c, :d siNON k߆XZWVuOEa;4bF:]?]?~=#\JTwKTwIT%;%%MڨXXXnX>#*ܢ^.63-JvKZ$n&mܲ[wqerK[wqK纸e76ŭwq]Ҹ.n-]]2%-\'^Z**;U+~ʨd7,DSJMC|)47^xPF"\ Jފ9On4#7G{k~}[~{̕kGH^w?zە^vyͣz5ٕKWpD/REgbRGH vX.}DJG\ Kd/`rYLJ/ BW:/VJk;Ӽ7Kf)ԙ?ʙv-73-g!C:ioJ7LiaE>.+3;wLJsL˒YR:ϴ0K(r{D*iAi%cJTܙ('J鉼V&ˤtYc[tCc !wHr'\ K;{\5W_;9$5l}* ?Er0]ka?dzNLo-QT[SZWQZ*,sKRZ}=OYh_*zl^yΔVV̋Ȍk!fn9+OקG9H]ZWS-| ZhaW UV:m ˪+W\Z]l_?^hq7KWvuZm*q+ʗ te!]keVhr`qMeiӕY>M*ufGDEԾ~}נ}_ ZYt 96n:ovv4rM;V*:;Ѷ5q5j˖V/ 2Nl-e=Ĺq'AIñgدϲo}z{_cWo3Jkm+:`Z6_mqmFKs)YV]W"griUhc=،tGzW)Dd x0  # A2]Jpߦ0`!y]Jpߦ04N9Gx] |E=W&pAXVOtQH"! QtՀ\r*+"O . )."f몚T_3=aZ~~{Uuu @/ ~oP9(0: nR?Y:1Eqq?`~uOumOu|TXX `3d~su}(0 ^BO pٳ^"`J˛x/+e(#XOG ,ASyUDWPW.xtyL0C6 ~CWHշՁuy6 nٟe<:w$R~P)O&1-h?!9.K\mwLy$ +wJ -N/p q :A:|6ⷥ gNZLB3SәB3'd= -N.guꢹN}/hf~t3~O3~̯YbD>'{TSe$c mbV61+61KĬ3)9Ct:; NNg >utN:B:s:DstS9{HtXW3cZst0tN$zboB솘pZ)\#^ALp~p=Op(j:k5~B^i^RBw{[]އ^QEOo*e:_@?` 1_(\_F" R-nt ׵x#k_xy-\l"mK<ˉ({=Lt^\9>d^t+ xt Wϋ i| _p W)׷{>b.\7ⵙ 3 ^n*e2b &N\?\{f,$2--\\_(]cmӸ}:_x #:p} MR^?AcZU%8\_p"ud|y/sGbyGR6v+/xMC1=µ!($0ZJY?Nf|I~m03#ش9ƃo]•!؞٘_9jLp-U>Dj =:[k9G\V7(\iE1n *W)XV>b)tk)\w. PҮR6q}ܠ^2j.\yu rsGyܫr~pg1ZuEG-\\p=Fp-Beukrl0µ^\r5rU2>]iЅx;%j#([*W) {b7k^@Z#YS|5 Y?A;|]|+em!Ř>lYvZ/*1~$ftu:w!N1l_~ptS?pME+MQ W)bVڪpZuǵpA{ סo(p~po} }k hh־%eJv#>x2d UhC0H\{\yYmukߤ]us="cCX[:(!}_6@lf 6 eຒx.JU<QRo[4uf_<~}+h_\k|]9~[ ,*ez$ȹC\O\?H\gعJD\C7㺼w[Âq Mk}}}۫rןf{ݜa^\ϱyt.P9þF$eJ:]8:b;Vp>qݍs4n]~!.Kko*eJnc:ǂ@%>xr]H{0fW)kUՊx+V:V\d^4WuI:_Wn_W~+7㛼AQF}Ϗksľ̅x2\9j W)WZ9/ѸyiE^sާ9+%t_K4% }-AiFAl[ /3ȱ7?Bh :L+1eS`-{. D\{✷2:׵Wwnh>l!eນհ=wKnk-؆&9qǺ9 8gGHY?~{6w0g߳~p=Xw~1aўJkL&ezBCـ{ 3 s*b\#v{,}\eFg28O! W/s^=}\f6 o]oAݐR}bĴ ZS zƟ_i*e&ڵzv`iף]ouh >k}aıf{Zb{m棽6"3,\k5!twi W)׺߻pm,#Ҟq6=-\li&cZ:wlR+q+ R ={6{νox~bQϭ=֞ϭ=yJ)$t:?:/t:;RItA u@tA*_v{RNcPZƖXle1:wxInT=DH~Xڊw =\ř>B{rM.J&& \tG)iA:-HN N ν(:]NPNga}^ ZG?NdYL}]O4[h2Z&=KAe^I^ʼRs]-֯̿Ge2[l2I2eޛa/dWV^Ӝ >^3?/I?/Ia^"TTH'I0:):)Lu|a1Gi"~N{&*x6ts+0+ea~WH2FUľ4)K{^wi~Kw&n~6h? n:׽Hc?$~w}cd|H K/\KN.pk݇x'=wg`@ⵊa9}t~>9Eiv㗉O=jvUs3|sV?YY?/8VSE/CvD12B ӽ>MH$ـ{\ J<^YcC;\(}@brgu#=̿*6< i򌆥QJg~vX_OpLgЀ(ӣI,K7Sn(֚353cFNt#mC9fYZ\jiJ[-1ACrY)vb}]Z1cs'}X8NJi%y |MF_*iMRaQ2,cUƲX|+_~+#3Ee~VJ8=toF41ۯ95Sn'$9v0p (&nuz@v ÐkC;$Z0!@Dv@ˉxF+0iC r^.`92݆Ve 4y6-7-3p;mhSF9Bأ7"Ue PiNSjݦcM8tR&B{ ̶MF=s4ʕ͎D >rlX!j׳llhG.F9~``[96n[ర7Zn]v Ǎ G)hheg,̶: |[3ڥl3e r^њlCV0fYk=[CM}hMٿel #(給 = mg {|s.sn9n7-{ɆV%p4IDXN콾3gn{7:b1y/ܾ[a҇vOo<F60@QݐPvnE$Z0X7܅6Ap@WG`!LzF-le>GR4N9hx ǟݗsiJJQ29:9:D%*NFLMWs9KE*BP?Rt$gkewho}]2((0QND𿧇+ҩ9׶n~Ƞ!r:՝uUii׸* 0hDT Iٚ憀>*M>h%a"#f%Ӑhv]L}  F8"ɐBIpg]Q4г}(]egjjqd.|vW+4dž!UDpPo*<*~Σ 侷;tRz e:V3n/Y{7R?kJE FNr8i{ kM%l_fOo]wY l9OTkrʭ4c>W,RyojeB C5'.?"/CFN9rXzZ}u;|Dͯ-I;e뚞34!!|#pqZ;wjٚ}iȀR/m~z_2ly`Fjܷπg*!Z W~!ApOk{ҥq}ײQqp~RvT=~!ױ_iCFP}L~oKL+_2#˴L?>eڷ e7eWi!՚B6-W]5,ճ j3I|#ߏ$7w369}^Iy%+9}2m&P'Nu49Jʴʴ ˴,22YwRPSXe 0+lXWk׏CgU~ q6v[pba5eyn([?X{A*h{ k5X;+ 3&3Yk$`N'vl`}MūKzNl^F($c] S.6Zc/qu-V=>yXcm}7n= -h?ºUGy ppa]>k0{$Ϊl`bSvakS0>{~eaU~΁vh/{ <_Xc!i ud]udG{ĵYP_ӞgrkgMڌG*IΊXG@wq>sak[x-`U% u|0 tZ֫q0caU~>ݭ:X ŕt }nX'ykaU~V4Eu.6]cyi%(Mo+๾å !lֿ~~.kAe] fד;7&?籋Eʰc鼆pa`s 1Ygĺrr kn`G֓Cb\]vZnOZ륭3!2߂̅u^@obc43[,U&cmwy1pkgFt(D7X*[?X_^K֗\coD,V3 ־&'@x۟J=ڵ^z"ŚyװXY~P{㹰DX {X읭;ªl`}Dg=Q7b<9ªl`z" 1aMn/ژ0 yweLlv@+X+z_暀5^X{b[ªl`w +#0r u;R9wܳA"S;qXZl8._`=0SXX;>q]=!6m4` zYb8nYn.;Z?`maU~zR$ ֓N 'j !M'.mV!#6fpϪl`t[Å!XƏmaU~ pn[?^F Hnm2^Λy_n֍b}0M큻ªl`彬}8qa]@0zp*[?XWĹb؉CU0?} \sKU@YlDXU`,[Ӆ4qTV䫲A: ׁ \ekt~h0"C "Pe*˜7O qny k>XKm"?Y6kȅG@c=&FB[X;^\X7 /jz`kUzvP6^?g Se+㙆m"\>3^ Ɠ3-"xuVek]h? \X 5Ry,5_-ʶ3xU{ĭ9S{?ϙ'#j/Q:ʴx=p=cJyU2+ez,ӻ*e: e:$e:DT.Rcغ81Djڤ?ɞe  ZwdCC"Y!{lk$-̶ęmƧ?g=lsFIҬi^S?ߴ϶Ҝ1m$3qiuH#چ¾c;ಮB-Wy\`>;Όs -¿E_)^ /3{B5ZdgU,;3Z"f[Rǫk k^8N%ٮ] Ka&հcJy/COae:{R^#P#RX#dHoqH|V ~ K?<|] 0aXl0|W2w=^'K?Moq6~8'q#8K -qq^%emęg< yd HٞG=zqdTdT d,QcD qq5۟1>#vyeYEu3xo^sHV<(sDeNi+߫iZ*0.{CA2]ci!#mh/|fo& Gpb_8ZSG$DCADktxE;UeaO&MQ?sGqp[jhmoD` Y~QU󷊌Gtg4(e_95xT,<-xR-jEK("@Z;:0Iij-.#jG{J7Tkm.K.rPqwCts-KSMo8֔O9a>7SyB3|S\~T\d6+R#[*5I^턲sd˺مߑ,i2WG^=%G _Y5xrT({{HűcN,9;Go~8ja* Zts="׷/׶J 6PMm% j!blj+F>j6Sʳ}B]joS:J5yUۅ 6j l8=4Ū4]ZpB(,]+wP- j?MdSۉXHfԦݪZOOi"5nX;],۩c -R}^f(34\ϫNz q;)V{'E.,|CSP 9Rx"&ꇻļq B}զBmdj;t},뵔&˷ϑ)0&SXEmjʯfH[SvzQU՚_tM6 4ZOWK8:=c{4JҞ&S_O=mj㨷TcW=KwԦ"HO4DRO)S-@ͱ=DJ5yU+yO4Vgβ/ګ]sU'>zIԶыR}Ւ=g|Bg4} >rVYmQQ|j;P[䠶$ijZZlUm9)vSs;ئ- B[9hձI>{<.TcW jsPkSkYmejo$V{#ՖQ%MJNZ0I;}jc,sC+ӓ5jkJZI]9)Um*glҏ O;u >jQUJ3XmVU(aoai-æ-1d#AmELmEZUT04M5~3=Cob J+W>LɳaS4Ys#6hocvmZG--״@[mj6ƾjj[3`9`>`>vV?Rei}4ȁ}^󭲱4V">j7`)ƶsJ tM9Lwlϥ jeSNm%^tMmbjSjɞo WMm.j6:!W;C,<ͧ댗L"=yHRz},{.6c,lgF] 53"ծJov bMZ(ɳI՘{#tTcWomj>k^&S/nwqqY1CmjR>jc'olj!ƾxD{!|:+~qFbq!5+4f`OE SFívRkghi|SXԥ k` `]s/2D-R:du2U /Dd x0  # A2-v|3= '9-`!-v|3= '9\4N9MxŶO (U"^aQ1 ADX]#9ETT'( `p]1";SU53=8-j~SkA9DmT LG־mfO9 j-k/'qD #А^}IQlsS@)U8h#ˋk'#f%,=zDytϩ> H$W} ͟͟nG4GzCUuM6:ϐN]idҫ|qz.u'-/aS@D͟J8C[K/ZU>'nZK\PwRB߯p!\#wㇱMþǟ^Iۑ&ϔ#]vk 2*B#zmHtJe5І\u nN\xTBF(af(X&>~K左yo0 吪 %+Y8803_0y_ɰdaTpW@~נmvGɐ:ܡr>v'h]캶ʖq"֒&=G}uc)-#8w Ixv}C:SfSu:Xo:AbWu:濣NV]Ϛj~p;mwoH{JB{_#2GdO{OS>p}vZO̢OT>1r_{#u:*::tԟNܳu::::twItLu:&u:F=yT듎{:a|ſj jR}d7ktߘkǨkg#QT^P_܉ Q78pe)4'T۲?q>[}Xjs?XK:á]' lZ׎0Xk*AxUA1vGڅR\;`q;~6Vm+;1 kinFUmƪm`=u |$S\Xg)kA0CxUDspQӅ@bZmVm`~@qˆ +Ve`4Ȫm`=Wcq>kcն~rC;4juj>rsQY&M  +p+Y姪bJ b!%k}$**ùWҶ^)}ڹ.n[wguUb :ubu>~C;4>ua6/σ5, v{\X[`< 0!,j[?XC{ Pu\k㱀96sZhς`Ŝu g.q6j[?X~56`ƍoW"6Vmu3geg*о#,]g^ y`]+U{Wxj]X9?-"ƪm`}F٫Gnk0> j[?Xrݘ\WZ?q)lоVH.p8K6Vm"hwZ7.ǵ x$)_0Yua]kmx#dն~^-sqǵ\`:@6kuUQ= zLa#k Q+V%Ǭ.pgO]srfj[?X 9^.bdx>icն~ۗ=  ց(ɀy+6VmX!qlva V\lW\˅p~z;@UZ8ڵ-F2Vf=U0W!UK<ą5kooȲ-Ÿecն\˺WA_qm,Gr:ɪm`XHqovoX770s֯@>]X},X/j[?Xs8 ڇ"Z5$5=XA+wMVmО#zjH3h\9 6Vm!y?p"©.uzX"̳j[?Xst s2֥v1j! myYɚ\(ߎ6VmklgLu]X<5zz ccն~џXG.qbۀ+[?m`]Or~&gmRB:;rƪm`e?< a/r= Z>z6жY;=!"l6xϖŰ[u} S\X|[1Xjcն~FhW }I^D v*:rYĵw5Ƒ {^^zCą5׹ /y)EEUڼZ+(@Võg ]o3ƪm`o,%+2`e.k/8hAw+]XLIW X'*mbB\j0g#lbcն~փ>z5XyO7o26Vm;Fri'23 0w^UU^υI4UyBN3s |wWn8™JFI,YDy4ב3"FI򼿿3g쯴o݊[ۼst>Ӓq YR>9!E|reN'fPXUNSuo'+LU~դ RI~̋Wj/uGIEm^̶+<̋K4&N<%Ls-Smi^LLiz3M4'4˗iޜLmY6:96:9mtjF=?Cu2%::dJp E[G^&YA)nqz@~XCmꏰX觖*?ރ*s'W}1 =c5}q,yg|۳ aYnZxnR@~OׇѾч>wae+vNg-pp7m}K"f/݂{辶j*yiLsHR nqG쁶rSo\5[%=X(gʳv*X5]Z~Vh-BVܒfp8)VCQ6hG tky6 ĿЎ6c3R㐟s[hm -UA#Άx~"ͯȨ߲5aϹP}Xb D,KXI% KELQǪL HIzLTRR V[.]#*W d?' ѷV!= j[(g;IJcu<վ1,0hOoi(R{g<ؽC_폜]9 jeOdJI`%qE6A1PA.Ǹiفp *z:*kǴZ(Z-Q Jj 1jvCBq̫Zuq,9E1jDB$9^ =E2jQ71j5rH=Ǽ-BPO(5ZN6.S'ڃtRXZb]P{H5i5OwYP?ZY\, /Ův?uQj3-T"&O 9Um* vڽtRWtC9FJc^F q'Mu iJc֬KQL+M?̡v+ݯ8դ݃oIIjeՆ,1sPj6l^snGN'l4磄&;ԊQRc^nڍIjm 4jO?HqՊ:vRdjҵ;ԖR/1.Khͧ;js1S-|Ji<9IIs:nz,8AChCm*J@q̫9POӕǼt=IiCm=8UQZ*f-V˱XoyiiF gjhFG2sleH\X"8Um<-hCm8-Sj6֋ʡvQj6wϮv=8ަu:ˡ w7RmiӹiRB :ɡNQj󪶁jj"eS{˦PXA!SUj󪶜%P[H(5i`v9tC:\q-ct͢f>d:P}ǼJ)*)6:]Huz$f:Zqt'QvvǼw{YqkNQKv-fJcֱs봭~1up-j^A@> Zadani font odlomkaViV Obi na tablica4 l4a .k. Bez popisaLL [ Tekst fusnote$a$CJOJQJaJ>&> [Referenca fusnoteH*jC@j [Uvu eno tijelo teksta$dh`a$OJQJmH sH :B@": [ Tijelo tekstax6U@16 x Hiperveza >*B*ph8 @B8 5MPodno~je  p#2)@Q2 5M Broj straniceN"@N @q Opis slike xx56OJQJ\aJ*+X:;sopQR   *+%%'',,U/V/r/11212255^7_7:>;>8G9GMMM(Q)QmUnUrWsW[[ _ _aacc!i"immm$m|o}ompnprrttttvwwwQxRxyy<{={||/}0}1}2}3}4}5}6}7}_}a}b}}}}t~./de{|Z[\]^_܅ޅ߅{|}~ކ҈ӈԈՈֈ׈؈وڈۈ܈689 !"#$vxy{|MNOPQRSTUVҏӏŐƐpqstvwyz|}0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000@000@000@000@000@0@0@0@0@0@000    004x0000000 $$$' _Am0~ΚNQSUXZ\`.;Hmwt}6TORTVWY[]^_dP`(8f}+BE]tw͐XXX t t t t t t  '!!8@0(  B S  ? OLE_LINK1 OLE_LINK2 _Toc115874686 _Toc115874687 _Toc115874690 _Toc115874691 _Toc115874692 _Toc115874693 _Toc115874694 _Toc115874695 _Toc1158746967}b}e_܈$VƐ **]}}ۅ݆5uciG(l :iG(|5iG(ܠiG(L:iG(5iG(5iG(5iG(:iG(L:iG( :iG(:iG(:iG(|5iG(|iG(5iG(iG(<iG(|iG(iG(iG(<iG( :iG( :iG(, :iG(l :iG( :iG( :iG(, :iG(l :iG( :iG( :iG(, :iG(l :iG( :iG( :iG(, :iG(l :iG( :iG( :iG(iG(<iG(|iG(iG(iG(<iG(|iG(iG(iG(<iG(|iG(iG(iG(<iG(|iG(iG(iG(<iG(|iG(iG(T:iG(:iG(:iG(:iG(T:iG(:iG(:iG(:iG(T:iG(:iG(:iG(:iG(T:iG(:iG(:iG(:iG(T:iG(:iG(:iG(:iG(T:iG(:iG(4:iG(t:iG(:iG(:iG(4:iG(t:iG(:iG(:iG(4:iG(t:iG(:iG(:iG(4:iG(t:iG(:iG(:iG(4:iG(t:+;;IL > H H ZZkvvGG\\((**iUiU[[u}}~ ~~~~&~~~~~!!' E"" qqȅȅ,``       !"#$%&'()*+,-/.0123465789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`ab:EOOR   D O O aap||II^^((**kUkU[[u}~~~~%~.~.~~~~~&//MɁ**(yyЅЅ3ff  !"#$%&'()*+,-/.0123465789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`ab=_*urn:schemas-microsoft-com:office:smarttags PlaceType=^*urn:schemas-microsoft-com:office:smarttags PlaceName8@*urn:schemas-microsoft-com:office:smarttagsCityB[*urn:schemas-microsoft-com:office:smarttagscountry-region9`*urn:schemas-microsoft-com:office:smarttagsplace>c*urn:schemas-microsoft-com:office:smarttags PersonName i Jurica `imurina ProductIDccc`_^``[[[``[[`[[`[`[`[`[`[`[`[`[`@[`[``[[^_[`^_`[`[`^_[`[[`[^_[`[[[`[[^_[[`[`[[``[[`[[`[`[`[`[[`[+12:<@X^s+.++D/I///p7u7??GEOEMMNMXM}QQQQUUWWWW"\+\ ____ccee2i9ijjsj‚ĂłȂpqqssttvwyz|}si$;%t+~+88aWjW]]m`o`҅օpqqssttvwyz|}333333333mmpqqssttvwyz|}pqqssttvwyz|}Cgn,XP Q 2 ="D\kgmkki*3R1$&Y%&'|J'W'n( )$,^.a.T/q/3B0H1m1w3z5%6|&6#27f7sB9yB=0A2B~EQF?HvIfKPCMWR(JR}RUTrV"VUV[@[\ \4^be^(_Bbmc_dg_oFr)su~HuhuWYvq$w}ryuykzK{6{+[ Wa+W#M\!Bc',;?Kc9&G}a*^xa5bzDP/Yd#k o'[lt{ 6VUa+9@q5M;A^-#oKd\x !1YVb%*%!ii9wl+B@R @/0pp@p8pt@p@UnknownGz Times New Roman5Symbol3& z Arial"1Vù&Vù&!&p}K p}K !4d&&2QHX ?+[2)R&D ACTIVITY AND PATENTS IN CEE COUNTRIESJurica `imurinakorisnikOh+'0$ 8D d p | ,R&D ACTIVITY AND PATENTS IN CEE COUNTRIESJurica imurina Normal.dot korisnik2Microsoft Office Word@@@@p}՜.+,D՜.+,p, hp  Ekonomski fakultet Zagreb K& *R&D ACTIVITY AND PATENTS IN CEE COUNTRIES Naslovx 8@ _PID_HLINKSA00mailto:ssmolic@efzg.hr_h_mailto:tgelo@efzg.hr_a_mailto:jsimurina@efzg.hr_  !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdeghijklmnopqrstuvwxyz{|}~     Root Entry FP6%Data f1TableCWordDocument8SummaryInformation(DocumentSummaryInformation8CompObjs  F!Microsoft Office Wordov dokument MSWordDocWord.Document.89q